repository_name
stringclasses
316 values
func_path_in_repository
stringlengths
6
223
func_name
stringlengths
1
134
language
stringclasses
1 value
func_code_string
stringlengths
57
65.5k
func_documentation_string
stringlengths
1
46.3k
split_name
stringclasses
1 value
func_code_url
stringlengths
91
315
called_functions
listlengths
1
156
enclosing_scope
stringlengths
2
1.48M
sentinelsat/sentinelsat
sentinelsat/sentinel.py
SentinelAPI.download_all
python
def download_all(self, products, directory_path='.', max_attempts=10, checksum=True): product_ids = list(products) self.logger.info("Will download %d products", len(product_ids)) return_values = OrderedDict() last_exception = None for i, product_id in enumerate(products): for attempt_num in range(max_attempts): try: product_info = self.download(product_id, directory_path, checksum) return_values[product_id] = product_info break except (KeyboardInterrupt, SystemExit): raise except InvalidChecksumError as e: last_exception = e self.logger.warning( "Invalid checksum. The downloaded file for '%s' is corrupted.", product_id) except SentinelAPILTAError as e: last_exception = e self.logger.exception("There was an error retrieving %s from the LTA", product_id) break except Exception as e: last_exception = e self.logger.exception("There was an error downloading %s", product_id) self.logger.info("%s/%s products downloaded", i + 1, len(product_ids)) failed = set(products) - set(return_values) # split up sucessfully processed products into downloaded and only triggered retrieval from the LTA triggered = OrderedDict([(k, v) for k, v in return_values.items() if v['Online'] is False]) downloaded = OrderedDict([(k, v) for k, v in return_values.items() if v['Online'] is True]) if len(failed) == len(product_ids) and last_exception is not None: raise last_exception return downloaded, triggered, failed
Download a list of products. Takes a list of product IDs as input. This means that the return value of query() can be passed directly to this method. File names on the server are used for the downloaded files, e.g. "S1A_EW_GRDH_1SDH_20141003T003840_20141003T003920_002658_002F54_4DD1.zip". In case of interruptions or other exceptions, downloading will restart from where it left off. Downloading is attempted at most max_attempts times to avoid getting stuck with unrecoverable errors. Parameters ---------- products : list List of product IDs directory_path : string Directory where the downloaded files will be downloaded max_attempts : int, optional Number of allowed retries before giving up downloading a product. Defaults to 10. checksum : bool, optional If True, verify the downloaded files' integrity by checking its MD5 checksum. Throws InvalidChecksumError if the checksum does not match. Defaults to True. Raises ------ Raises the most recent downloading exception if all downloads failed. Returns ------- dict[string, dict] A dictionary containing the return value from download() for each successfully downloaded product. dict[string, dict] A dictionary containing the product information for products whose retrieval from the long term archive was successfully triggered. set[string] The list of products that failed to download.
train
https://github.com/sentinelsat/sentinelsat/blob/eacfd79ff4e7e939147db9dfdd393c67d64eecaa/sentinelsat/sentinel.py#L554-L627
[ "def download(self, id, directory_path='.', checksum=True):\n \"\"\"Download a product.\n\n Uses the filename on the server for the downloaded file, e.g.\n \"S1A_EW_GRDH_1SDH_20141003T003840_20141003T003920_002658_002F54_4DD1.zip\".\n\n Incomplete downloads are continued and complete files are skipped.\n\n Parameters\n ----------\n id : string\n UUID of the product, e.g. 'a8dd0cfd-613e-45ce-868c-d79177b916ed'\n directory_path : string, optional\n Where the file will be downloaded\n checksum : bool, optional\n If True, verify the downloaded file's integrity by checking its MD5 checksum.\n Throws InvalidChecksumError if the checksum does not match.\n Defaults to True.\n\n Returns\n -------\n product_info : dict\n Dictionary containing the product's info from get_product_info() as well as\n the path on disk.\n\n Raises\n ------\n InvalidChecksumError\n If the MD5 checksum does not match the checksum on the server.\n \"\"\"\n product_info = self.get_product_odata(id)\n path = join(directory_path, product_info['title'] + '.zip')\n product_info['path'] = path\n product_info['downloaded_bytes'] = 0\n\n self.logger.info('Downloading %s to %s', id, path)\n\n if exists(path):\n # We assume that the product has been downloaded and is complete\n return product_info\n\n # An incomplete download triggers the retrieval from the LTA if the product is not online\n if not product_info['Online']:\n self.logger.warning(\n 'Product %s is not online. Triggering retrieval from long term archive.',\n product_info['id'])\n self._trigger_offline_retrieval(product_info['url'])\n return product_info\n\n # Use a temporary file for downloading\n temp_path = path + '.incomplete'\n\n skip_download = False\n if exists(temp_path):\n if getsize(temp_path) > product_info['size']:\n self.logger.warning(\n \"Existing incomplete file %s is larger than the expected final size\"\n \" (%s vs %s bytes). Deleting it.\",\n str(temp_path), getsize(temp_path), product_info['size'])\n remove(temp_path)\n elif getsize(temp_path) == product_info['size']:\n if self._md5_compare(temp_path, product_info['md5']):\n skip_download = True\n else:\n # Log a warning since this should never happen\n self.logger.warning(\n \"Existing incomplete file %s appears to be fully downloaded but \"\n \"its checksum is incorrect. Deleting it.\",\n str(temp_path))\n remove(temp_path)\n else:\n # continue downloading\n self.logger.info(\n \"Download will resume from existing incomplete file %s.\", temp_path)\n pass\n\n if not skip_download:\n # Store the number of downloaded bytes for unit tests\n product_info['downloaded_bytes'] = self._download(\n product_info['url'], temp_path, self.session, product_info['size'])\n\n # Check integrity with MD5 checksum\n if checksum is True:\n if not self._md5_compare(temp_path, product_info['md5']):\n remove(temp_path)\n raise InvalidChecksumError('File corrupt: checksums do not match')\n\n # Download successful, rename the temporary file to its proper name\n shutil.move(temp_path, path)\n return product_info\n" ]
class SentinelAPI: """Class to connect to Copernicus Open Access Hub, search and download imagery. Parameters ---------- user : string username for DataHub set to None to use ~/.netrc password : string password for DataHub set to None to use ~/.netrc api_url : string, optional URL of the DataHub defaults to 'https://scihub.copernicus.eu/apihub' show_progressbars : bool Whether progressbars should be shown or not, e.g. during download. Defaults to True. timeout : float or tuple, optional How long to wait for DataHub response (in seconds). Tuple (connect, read) allowed. Attributes ---------- session : requests.Session Session to connect to DataHub api_url : str URL to the DataHub page_size : int Number of results per query page. Current value: 100 (maximum allowed on ApiHub) timeout : float or tuple How long to wait for DataHub response (in seconds). """ logger = logging.getLogger('sentinelsat.SentinelAPI') def __init__(self, user, password, api_url='https://scihub.copernicus.eu/apihub/', show_progressbars=True, timeout=None): self.session = requests.Session() if user and password: self.session.auth = (user, password) self.api_url = api_url if api_url.endswith('/') else api_url + '/' self.page_size = 100 self.user_agent = 'sentinelsat/' + sentinelsat_version self.session.headers['User-Agent'] = self.user_agent self.show_progressbars = show_progressbars self.timeout = timeout # For unit tests self._last_query = None self._last_response = None def query(self, area=None, date=None, raw=None, area_relation='Intersects', order_by=None, limit=None, offset=0, **keywords): """Query the OpenSearch API with the coordinates of an area, a date interval and any other search keywords accepted by the API. Parameters ---------- area : str, optional The area of interest formatted as a Well-Known Text string. date : tuple of (str or datetime) or str, optional A time interval filter based on the Sensing Start Time of the products. Expects a tuple of (start, end), e.g. ("NOW-1DAY", "NOW"). The timestamps can be either a Python datetime or a string in one of the following formats: - yyyyMMdd - yyyy-MM-ddThh:mm:ss.SSSZ (ISO-8601) - yyyy-MM-ddThh:mm:ssZ - NOW - NOW-<n>DAY(S) (or HOUR(S), MONTH(S), etc.) - NOW+<n>DAY(S) - yyyy-MM-ddThh:mm:ssZ-<n>DAY(S) - NOW/DAY (or HOUR, MONTH etc.) - rounds the value to the given unit Alternatively, an already fully formatted string such as "[NOW-1DAY TO NOW]" can be used as well. raw : str, optional Additional query text that will be appended to the query. area_relation : {'Intersects', 'Contains', 'IsWithin'}, optional What relation to use for testing the AOI. Case insensitive. - Intersects: true if the AOI and the footprint intersect (default) - Contains: true if the AOI is inside the footprint - IsWithin: true if the footprint is inside the AOI order_by: str, optional A comma-separated list of fields to order by (on server side). Prefix the field name by '+' or '-' to sort in ascending or descending order, respectively. Ascending order is used if prefix is omitted. Example: "cloudcoverpercentage, -beginposition". limit: int, optional Maximum number of products returned. Defaults to no limit. offset: int, optional The number of results to skip. Defaults to 0. **keywords Additional keywords can be used to specify other query parameters, e.g. `relativeorbitnumber=70`. See https://scihub.copernicus.eu/twiki/do/view/SciHubUserGuide/3FullTextSearch for a full list. Range values can be passed as two-element tuples, e.g. `cloudcoverpercentage=(0, 30)`. `None` can be used in range values for one-sided ranges, e.g. `orbitnumber=(16302, None)`. Ranges with no bounds (`orbitnumber=(None, None)`) will not be included in the query. The time interval formats accepted by the `date` parameter can also be used with any other parameters that expect time intervals (that is: 'beginposition', 'endposition', 'date', 'creationdate', and 'ingestiondate'). Returns ------- dict[string, dict] Products returned by the query as a dictionary with the product ID as the key and the product's attributes (a dictionary) as the value. """ query = self.format_query(area, date, raw, area_relation, **keywords) self.logger.debug("Running query: order_by=%s, limit=%s, offset=%s, query=%s", order_by, limit, offset, query) formatted_order_by = _format_order_by(order_by) response, count = self._load_query(query, formatted_order_by, limit, offset) self.logger.info("Found %s products", count) return _parse_opensearch_response(response) @staticmethod def format_query(area=None, date=None, raw=None, area_relation='Intersects', **keywords): """Create a OpenSearch API query string. """ if area_relation.lower() not in {"intersects", "contains", "iswithin"}: raise ValueError("Incorrect AOI relation provided ({})".format(area_relation)) # Check for duplicate keywords kw_lower = set(x.lower() for x in keywords) if (len(kw_lower) != len(keywords) or (date is not None and 'beginposition' in kw_lower) or (area is not None and 'footprint' in kw_lower)): raise ValueError("Query contains duplicate keywords. Note that query keywords are case-insensitive.") query_parts = [] if date is not None: keywords['beginPosition'] = date for attr, value in sorted(keywords.items()): # Escape spaces, where appropriate if isinstance(value, string_types): value = value.strip() if not any(value.startswith(s[0]) and value.endswith(s[1]) for s in ['[]', '{}', '//', '()']): value = re.sub(r'\s', r'\ ', value, re.M) # Handle date keywords # Keywords from https://github.com/SentinelDataHub/DataHubSystem/search?q=text/date+iso8601 date_attrs = ['beginposition', 'endposition', 'date', 'creationdate', 'ingestiondate'] if attr.lower() in date_attrs: # Automatically format date-type attributes if isinstance(value, string_types) and ' TO ' in value: # This is a string already formatted as a date interval, # e.g. '[NOW-1DAY TO NOW]' pass elif not isinstance(value, string_types) and len(value) == 2: value = (format_query_date(value[0]), format_query_date(value[1])) else: raise ValueError("Date-type query parameter '{}' expects a two-element tuple " "of str or datetime objects. Received {}".format(attr, value)) # Handle ranged values if isinstance(value, (list, tuple)): # Handle value ranges if len(value) == 2: # Allow None to be used as a unlimited bound value = ['*' if x is None else x for x in value] if all(x == '*' for x in value): continue value = '[{} TO {}]'.format(*value) else: raise ValueError("Invalid number of elements in list. Expected 2, received " "{}".format(len(value))) query_parts.append('{}:{}'.format(attr, value)) if raw: query_parts.append(raw) if area is not None: query_parts.append('footprint:"{}({})"'.format(area_relation, area)) return ' '.join(query_parts) def query_raw(self, query, order_by=None, limit=None, offset=0): """ Do a full-text query on the OpenSearch API using the format specified in https://scihub.copernicus.eu/twiki/do/view/SciHubUserGuide/3FullTextSearch DEPRECATED: use :meth:`query(raw=...) <.query>` instead. This method will be removed in the next major release. Parameters ---------- query : str The query string. order_by: str, optional A comma-separated list of fields to order by (on server side). Prefix the field name by '+' or '-' to sort in ascending or descending order, respectively. Ascending order is used, if prefix is omitted. Example: "cloudcoverpercentage, -beginposition". limit: int, optional Maximum number of products returned. Defaults to no limit. offset: int, optional The number of results to skip. Defaults to 0. Returns ------- dict[string, dict] Products returned by the query as a dictionary with the product ID as the key and the product's attributes (a dictionary) as the value. """ warnings.warn( "query_raw() has been merged with query(). use query(raw=...) instead.", PendingDeprecationWarning ) return self.query(raw=query, order_by=order_by, limit=limit, offset=offset) def count(self, area=None, date=None, raw=None, area_relation='Intersects', **keywords): """Get the number of products matching a query. Accepted parameters are identical to :meth:`SentinelAPI.query()`. This is a significantly more efficient alternative to doing `len(api.query())`, which can take minutes to run for queries matching thousands of products. Returns ------- int The number of products matching a query. """ for kw in ['order_by', 'limit', 'offset']: # Allow these function arguments to be included for compatibility with query(), # but ignore them. if kw in keywords: del keywords[kw] query = self.format_query(area, date, raw, area_relation, **keywords) _, total_count = self._load_query(query, limit=0) return total_count def _load_query(self, query, order_by=None, limit=None, offset=0): products, count = self._load_subquery(query, order_by, limit, offset) # repeat query until all results have been loaded max_offset = count if limit is not None: max_offset = min(count, offset + limit) if max_offset > offset + self.page_size: progress = self._tqdm(desc="Querying products", initial=self.page_size, total=max_offset - offset, unit=' products') for new_offset in range(offset + self.page_size, max_offset, self.page_size): new_limit = limit if limit is not None: new_limit = limit - new_offset + offset ret = self._load_subquery(query, order_by, new_limit, new_offset)[0] progress.update(len(ret)) products += ret progress.close() return products, count def _load_subquery(self, query, order_by=None, limit=None, offset=0): # store last query (for testing) self._last_query = query self.logger.debug("Sub-query: offset=%s, limit=%s", offset, limit) # load query results url = self._format_url(order_by, limit, offset) response = self.session.post(url, {'q': query}, auth=self.session.auth, headers={'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8'}, timeout=self.timeout) _check_scihub_response(response) # store last status code (for testing) self._last_response = response # parse response content try: json_feed = response.json()['feed'] if json_feed['opensearch:totalResults'] is None: # We are using some unintended behavior of the server that a null is # returned as the total results value when the query string was incorrect. raise SentinelAPIError( 'Invalid query string. Check the parameters and format.', response) total_results = int(json_feed['opensearch:totalResults']) except (ValueError, KeyError): raise SentinelAPIError('API response not valid. JSON decoding failed.', response) products = json_feed.get('entry', []) # this verification is necessary because if the query returns only # one product, self.products will be a dict not a list if isinstance(products, dict): products = [products] return products, total_results def _format_url(self, order_by=None, limit=None, offset=0): if limit is None: limit = self.page_size limit = min(limit, self.page_size) url = 'search?format=json&rows={}'.format(limit) url += '&start={}'.format(offset) if order_by: url += '&orderby={}'.format(order_by) return urljoin(self.api_url, url) @staticmethod def to_geojson(products): """Return the products from a query response as a GeoJSON with the values in their appropriate Python types. """ feature_list = [] for i, (product_id, props) in enumerate(products.items()): props = props.copy() props['id'] = product_id poly = geomet.wkt.loads(props['footprint']) del props['footprint'] del props['gmlfootprint'] # Fix "'datetime' is not JSON serializable" for k, v in props.items(): if isinstance(v, (date, datetime)): props[k] = v.strftime('%Y-%m-%dT%H:%M:%S.%fZ') feature_list.append( geojson.Feature(geometry=poly, id=i, properties=props) ) return geojson.FeatureCollection(feature_list) @staticmethod def to_dataframe(products): """Return the products from a query response as a Pandas DataFrame with the values in their appropriate Python types. """ try: import pandas as pd except ImportError: raise ImportError("to_dataframe requires the optional dependency Pandas.") return pd.DataFrame.from_dict(products, orient='index') @staticmethod def to_geodataframe(products): """Return the products from a query response as a GeoPandas GeoDataFrame with the values in their appropriate Python types. """ try: import geopandas as gpd import shapely.wkt except ImportError: raise ImportError("to_geodataframe requires the optional dependencies GeoPandas and Shapely.") crs = {'init': 'epsg:4326'} # WGS84 if len(products) == 0: return gpd.GeoDataFrame(crs=crs) df = SentinelAPI.to_dataframe(products) geometry = [shapely.wkt.loads(fp) for fp in df['footprint']] # remove useless columns df.drop(['footprint', 'gmlfootprint'], axis=1, inplace=True) return gpd.GeoDataFrame(df, crs=crs, geometry=geometry) def get_product_odata(self, id, full=False): """Access OData API to get info about a product. Returns a dict containing the id, title, size, md5sum, date, footprint and download url of the product. The date field corresponds to the Start ContentDate value. If `full` is set to True, then the full, detailed metadata of the product is returned in addition to the above. Parameters ---------- id : string The UUID of the product to query full : bool Whether to get the full metadata for the Product. False by default. Returns ------- dict[str, Any] A dictionary with an item for each metadata attribute Notes ----- For a full list of mappings between the OpenSearch (Solr) and OData attribute names see the following definition files: https://github.com/SentinelDataHub/DataHubSystem/blob/master/addon/sentinel-1/src/main/resources/META-INF/sentinel-1.owl https://github.com/SentinelDataHub/DataHubSystem/blob/master/addon/sentinel-2/src/main/resources/META-INF/sentinel-2.owl https://github.com/SentinelDataHub/DataHubSystem/blob/master/addon/sentinel-3/src/main/resources/META-INF/sentinel-3.owl """ url = urljoin(self.api_url, u"odata/v1/Products('{}')?$format=json".format(id)) if full: url += '&$expand=Attributes' response = self.session.get(url, auth=self.session.auth, timeout=self.timeout) _check_scihub_response(response) values = _parse_odata_response(response.json()['d']) return values def _trigger_offline_retrieval(self, url): """ Triggers retrieval of an offline product Trying to download an offline product triggers its retrieval from the long term archive. The returned HTTP status code conveys whether this was successful. Parameters ---------- url : string URL for downloading the product Notes ----- https://scihub.copernicus.eu/userguide/LongTermArchive """ with self.session.get(url, auth=self.session.auth, timeout=self.timeout) as r: # check https://scihub.copernicus.eu/userguide/LongTermArchive#HTTP_Status_codes if r.status_code == 202: self.logger.info("Accepted for retrieval") elif r.status_code == 503: self.logger.error("Request not accepted") raise SentinelAPILTAError('Request for retrieval from LTA not accepted', r) elif r.status_code == 403: self.logger.error("Requests exceed user quota") raise SentinelAPILTAError('Requests for retrieval from LTA exceed user quota', r) elif r.status_code == 500: # should not happen self.logger.error("Trying to download an offline product") raise SentinelAPILTAError('Trying to download an offline product', r) return r.status_code def download(self, id, directory_path='.', checksum=True): """Download a product. Uses the filename on the server for the downloaded file, e.g. "S1A_EW_GRDH_1SDH_20141003T003840_20141003T003920_002658_002F54_4DD1.zip". Incomplete downloads are continued and complete files are skipped. Parameters ---------- id : string UUID of the product, e.g. 'a8dd0cfd-613e-45ce-868c-d79177b916ed' directory_path : string, optional Where the file will be downloaded checksum : bool, optional If True, verify the downloaded file's integrity by checking its MD5 checksum. Throws InvalidChecksumError if the checksum does not match. Defaults to True. Returns ------- product_info : dict Dictionary containing the product's info from get_product_info() as well as the path on disk. Raises ------ InvalidChecksumError If the MD5 checksum does not match the checksum on the server. """ product_info = self.get_product_odata(id) path = join(directory_path, product_info['title'] + '.zip') product_info['path'] = path product_info['downloaded_bytes'] = 0 self.logger.info('Downloading %s to %s', id, path) if exists(path): # We assume that the product has been downloaded and is complete return product_info # An incomplete download triggers the retrieval from the LTA if the product is not online if not product_info['Online']: self.logger.warning( 'Product %s is not online. Triggering retrieval from long term archive.', product_info['id']) self._trigger_offline_retrieval(product_info['url']) return product_info # Use a temporary file for downloading temp_path = path + '.incomplete' skip_download = False if exists(temp_path): if getsize(temp_path) > product_info['size']: self.logger.warning( "Existing incomplete file %s is larger than the expected final size" " (%s vs %s bytes). Deleting it.", str(temp_path), getsize(temp_path), product_info['size']) remove(temp_path) elif getsize(temp_path) == product_info['size']: if self._md5_compare(temp_path, product_info['md5']): skip_download = True else: # Log a warning since this should never happen self.logger.warning( "Existing incomplete file %s appears to be fully downloaded but " "its checksum is incorrect. Deleting it.", str(temp_path)) remove(temp_path) else: # continue downloading self.logger.info( "Download will resume from existing incomplete file %s.", temp_path) pass if not skip_download: # Store the number of downloaded bytes for unit tests product_info['downloaded_bytes'] = self._download( product_info['url'], temp_path, self.session, product_info['size']) # Check integrity with MD5 checksum if checksum is True: if not self._md5_compare(temp_path, product_info['md5']): remove(temp_path) raise InvalidChecksumError('File corrupt: checksums do not match') # Download successful, rename the temporary file to its proper name shutil.move(temp_path, path) return product_info @staticmethod def get_products_size(products): """Return the total file size in GB of all products in the OpenSearch response.""" size_total = 0 for title, props in products.items(): size_product = props["size"] size_value = float(size_product.split(" ")[0]) size_unit = str(size_product.split(" ")[1]) if size_unit == "MB": size_value /= 1024. if size_unit == "KB": size_value /= 1024. * 1024. size_total += size_value return round(size_total, 2) @staticmethod def check_query_length(query): """Determine whether a query to the OpenSearch API is too long. The length of a query string is limited to approximately 3938 characters but any special characters (that is, not alphanumeric or -_.*) will take up more space. Parameters ---------- query : str The query string Returns ------- float Ratio of the query length to the maximum length """ # The server uses the Java's URLEncoder implementation internally, which we are replicating here effective_length = len(quote_plus(query, safe="-_.*").replace('~', '%7E')) return effective_length / 3938 def _query_names(self, names): """Find products by their names, e.g. S1A_EW_GRDH_1SDH_20141003T003840_20141003T003920_002658_002F54_4DD1. Note that duplicates exist on server, so multiple products can be returned for each name. Parameters ---------- names : list[string] List of product names. Returns ------- dict[string, dict[str, dict]] A dictionary mapping each name to a dictionary which contains the products with that name (with ID as the key). """ def chunks(l, n): """Yield successive n-sized chunks from l.""" for i in range(0, len(l), n): yield l[i:i + n] products = {} # 40 names per query fits reasonably well inside the query limit for chunk in chunks(names, 40): query = " OR ".join(chunk) products.update(self.query(raw=query)) # Group the products output = OrderedDict((name, dict()) for name in names) for id, metadata in products.items(): name = metadata['identifier'] output[name][id] = metadata return output def check_files(self, paths=None, ids=None, directory=None, delete=False): """Verify the integrity of product files on disk. Integrity is checked by comparing the size and checksum of the file with the respective values on the server. The input can be a list of products to check or a list of IDs and a directory. In cases where multiple products with different IDs exist on the server for given product name, the file is considered to be correct if any of them matches the file size and checksum. A warning is logged in such situations. The corrupt products' OData info is included in the return value to make it easier to re-download the products, if necessary. Parameters ---------- paths : list[string] List of product file paths. ids : list[string] List of product IDs. directory : string Directory where the files are located, if checking based on product IDs. delete : bool Whether to delete corrupt products. Defaults to False. Returns ------- dict[str, list[dict]] A dictionary listing the invalid or missing files. The dictionary maps the corrupt file paths to a list of OData dictionaries of matching products on the server (as returned by :meth:`SentinelAPI.get_product_odata()`). """ if not ids and not paths: raise ValueError("Must provide either file paths or product IDs and a directory") if ids and not directory: raise ValueError("Directory value missing") paths = paths or [] ids = ids or [] def name_from_path(path): return splitext(basename(path))[0] # Get product IDs corresponding to the files on disk names = [] if paths: names = list(map(name_from_path, paths)) result = self._query_names(names) for product_dicts in result.values(): ids += list(product_dicts) names_from_paths = set(names) ids = set(ids) # Collect the OData information for each product # Product name -> list of matching odata dicts product_infos = defaultdict(list) for id in ids: odata = self.get_product_odata(id) name = odata['title'] product_infos[name].append(odata) # Collect if name not in names_from_paths: paths.append(join(directory, name + '.zip')) # Now go over the list of products and check them corrupt = {} for path in paths: name = name_from_path(path) if len(product_infos[name]) > 1: self.logger.warning("{} matches multiple products on server".format(path)) if not exists(path): # We will consider missing files as corrupt also self.logger.info("{} does not exist on disk".format(path)) corrupt[path] = product_infos[name] continue is_fine = False for product_info in product_infos[name]: if (getsize(path) == product_info['size'] and self._md5_compare(path, product_info['md5'])): is_fine = True break if not is_fine: self.logger.info("{} is corrupt".format(path)) corrupt[path] = product_infos[name] if delete: remove(path) return corrupt def _md5_compare(self, file_path, checksum, block_size=2 ** 13): """Compare a given MD5 checksum with one calculated from a file.""" with closing(self._tqdm(desc="MD5 checksumming", total=getsize(file_path), unit="B", unit_scale=True)) as progress: md5 = hashlib.md5() with open(file_path, "rb") as f: while True: block_data = f.read(block_size) if not block_data: break md5.update(block_data) progress.update(len(block_data)) return md5.hexdigest().lower() == checksum.lower() def _download(self, url, path, session, file_size): headers = {} continuing = exists(path) if continuing: already_downloaded_bytes = getsize(path) headers = {'Range': 'bytes={}-'.format(already_downloaded_bytes)} else: already_downloaded_bytes = 0 downloaded_bytes = 0 with closing(session.get(url, stream=True, auth=session.auth, headers=headers, timeout=self.timeout)) as r, \ closing(self._tqdm(desc="Downloading", total=file_size, unit="B", unit_scale=True, initial=already_downloaded_bytes)) as progress: _check_scihub_response(r, test_json=False) chunk_size = 2 ** 20 # download in 1 MB chunks mode = 'ab' if continuing else 'wb' with open(path, mode) as f: for chunk in r.iter_content(chunk_size=chunk_size): if chunk: # filter out keep-alive new chunks f.write(chunk) progress.update(len(chunk)) downloaded_bytes += len(chunk) # Return the number of bytes downloaded return downloaded_bytes def _tqdm(self, **kwargs): """tqdm progressbar wrapper. May be overridden to customize progressbar behavior""" kwargs.update({'disable': not self.show_progressbars}) return tqdm(**kwargs)
sentinelsat/sentinelsat
sentinelsat/sentinel.py
SentinelAPI.get_products_size
python
def get_products_size(products): size_total = 0 for title, props in products.items(): size_product = props["size"] size_value = float(size_product.split(" ")[0]) size_unit = str(size_product.split(" ")[1]) if size_unit == "MB": size_value /= 1024. if size_unit == "KB": size_value /= 1024. * 1024. size_total += size_value return round(size_total, 2)
Return the total file size in GB of all products in the OpenSearch response.
train
https://github.com/sentinelsat/sentinelsat/blob/eacfd79ff4e7e939147db9dfdd393c67d64eecaa/sentinelsat/sentinel.py#L630-L642
null
class SentinelAPI: """Class to connect to Copernicus Open Access Hub, search and download imagery. Parameters ---------- user : string username for DataHub set to None to use ~/.netrc password : string password for DataHub set to None to use ~/.netrc api_url : string, optional URL of the DataHub defaults to 'https://scihub.copernicus.eu/apihub' show_progressbars : bool Whether progressbars should be shown or not, e.g. during download. Defaults to True. timeout : float or tuple, optional How long to wait for DataHub response (in seconds). Tuple (connect, read) allowed. Attributes ---------- session : requests.Session Session to connect to DataHub api_url : str URL to the DataHub page_size : int Number of results per query page. Current value: 100 (maximum allowed on ApiHub) timeout : float or tuple How long to wait for DataHub response (in seconds). """ logger = logging.getLogger('sentinelsat.SentinelAPI') def __init__(self, user, password, api_url='https://scihub.copernicus.eu/apihub/', show_progressbars=True, timeout=None): self.session = requests.Session() if user and password: self.session.auth = (user, password) self.api_url = api_url if api_url.endswith('/') else api_url + '/' self.page_size = 100 self.user_agent = 'sentinelsat/' + sentinelsat_version self.session.headers['User-Agent'] = self.user_agent self.show_progressbars = show_progressbars self.timeout = timeout # For unit tests self._last_query = None self._last_response = None def query(self, area=None, date=None, raw=None, area_relation='Intersects', order_by=None, limit=None, offset=0, **keywords): """Query the OpenSearch API with the coordinates of an area, a date interval and any other search keywords accepted by the API. Parameters ---------- area : str, optional The area of interest formatted as a Well-Known Text string. date : tuple of (str or datetime) or str, optional A time interval filter based on the Sensing Start Time of the products. Expects a tuple of (start, end), e.g. ("NOW-1DAY", "NOW"). The timestamps can be either a Python datetime or a string in one of the following formats: - yyyyMMdd - yyyy-MM-ddThh:mm:ss.SSSZ (ISO-8601) - yyyy-MM-ddThh:mm:ssZ - NOW - NOW-<n>DAY(S) (or HOUR(S), MONTH(S), etc.) - NOW+<n>DAY(S) - yyyy-MM-ddThh:mm:ssZ-<n>DAY(S) - NOW/DAY (or HOUR, MONTH etc.) - rounds the value to the given unit Alternatively, an already fully formatted string such as "[NOW-1DAY TO NOW]" can be used as well. raw : str, optional Additional query text that will be appended to the query. area_relation : {'Intersects', 'Contains', 'IsWithin'}, optional What relation to use for testing the AOI. Case insensitive. - Intersects: true if the AOI and the footprint intersect (default) - Contains: true if the AOI is inside the footprint - IsWithin: true if the footprint is inside the AOI order_by: str, optional A comma-separated list of fields to order by (on server side). Prefix the field name by '+' or '-' to sort in ascending or descending order, respectively. Ascending order is used if prefix is omitted. Example: "cloudcoverpercentage, -beginposition". limit: int, optional Maximum number of products returned. Defaults to no limit. offset: int, optional The number of results to skip. Defaults to 0. **keywords Additional keywords can be used to specify other query parameters, e.g. `relativeorbitnumber=70`. See https://scihub.copernicus.eu/twiki/do/view/SciHubUserGuide/3FullTextSearch for a full list. Range values can be passed as two-element tuples, e.g. `cloudcoverpercentage=(0, 30)`. `None` can be used in range values for one-sided ranges, e.g. `orbitnumber=(16302, None)`. Ranges with no bounds (`orbitnumber=(None, None)`) will not be included in the query. The time interval formats accepted by the `date` parameter can also be used with any other parameters that expect time intervals (that is: 'beginposition', 'endposition', 'date', 'creationdate', and 'ingestiondate'). Returns ------- dict[string, dict] Products returned by the query as a dictionary with the product ID as the key and the product's attributes (a dictionary) as the value. """ query = self.format_query(area, date, raw, area_relation, **keywords) self.logger.debug("Running query: order_by=%s, limit=%s, offset=%s, query=%s", order_by, limit, offset, query) formatted_order_by = _format_order_by(order_by) response, count = self._load_query(query, formatted_order_by, limit, offset) self.logger.info("Found %s products", count) return _parse_opensearch_response(response) @staticmethod def format_query(area=None, date=None, raw=None, area_relation='Intersects', **keywords): """Create a OpenSearch API query string. """ if area_relation.lower() not in {"intersects", "contains", "iswithin"}: raise ValueError("Incorrect AOI relation provided ({})".format(area_relation)) # Check for duplicate keywords kw_lower = set(x.lower() for x in keywords) if (len(kw_lower) != len(keywords) or (date is not None and 'beginposition' in kw_lower) or (area is not None and 'footprint' in kw_lower)): raise ValueError("Query contains duplicate keywords. Note that query keywords are case-insensitive.") query_parts = [] if date is not None: keywords['beginPosition'] = date for attr, value in sorted(keywords.items()): # Escape spaces, where appropriate if isinstance(value, string_types): value = value.strip() if not any(value.startswith(s[0]) and value.endswith(s[1]) for s in ['[]', '{}', '//', '()']): value = re.sub(r'\s', r'\ ', value, re.M) # Handle date keywords # Keywords from https://github.com/SentinelDataHub/DataHubSystem/search?q=text/date+iso8601 date_attrs = ['beginposition', 'endposition', 'date', 'creationdate', 'ingestiondate'] if attr.lower() in date_attrs: # Automatically format date-type attributes if isinstance(value, string_types) and ' TO ' in value: # This is a string already formatted as a date interval, # e.g. '[NOW-1DAY TO NOW]' pass elif not isinstance(value, string_types) and len(value) == 2: value = (format_query_date(value[0]), format_query_date(value[1])) else: raise ValueError("Date-type query parameter '{}' expects a two-element tuple " "of str or datetime objects. Received {}".format(attr, value)) # Handle ranged values if isinstance(value, (list, tuple)): # Handle value ranges if len(value) == 2: # Allow None to be used as a unlimited bound value = ['*' if x is None else x for x in value] if all(x == '*' for x in value): continue value = '[{} TO {}]'.format(*value) else: raise ValueError("Invalid number of elements in list. Expected 2, received " "{}".format(len(value))) query_parts.append('{}:{}'.format(attr, value)) if raw: query_parts.append(raw) if area is not None: query_parts.append('footprint:"{}({})"'.format(area_relation, area)) return ' '.join(query_parts) def query_raw(self, query, order_by=None, limit=None, offset=0): """ Do a full-text query on the OpenSearch API using the format specified in https://scihub.copernicus.eu/twiki/do/view/SciHubUserGuide/3FullTextSearch DEPRECATED: use :meth:`query(raw=...) <.query>` instead. This method will be removed in the next major release. Parameters ---------- query : str The query string. order_by: str, optional A comma-separated list of fields to order by (on server side). Prefix the field name by '+' or '-' to sort in ascending or descending order, respectively. Ascending order is used, if prefix is omitted. Example: "cloudcoverpercentage, -beginposition". limit: int, optional Maximum number of products returned. Defaults to no limit. offset: int, optional The number of results to skip. Defaults to 0. Returns ------- dict[string, dict] Products returned by the query as a dictionary with the product ID as the key and the product's attributes (a dictionary) as the value. """ warnings.warn( "query_raw() has been merged with query(). use query(raw=...) instead.", PendingDeprecationWarning ) return self.query(raw=query, order_by=order_by, limit=limit, offset=offset) def count(self, area=None, date=None, raw=None, area_relation='Intersects', **keywords): """Get the number of products matching a query. Accepted parameters are identical to :meth:`SentinelAPI.query()`. This is a significantly more efficient alternative to doing `len(api.query())`, which can take minutes to run for queries matching thousands of products. Returns ------- int The number of products matching a query. """ for kw in ['order_by', 'limit', 'offset']: # Allow these function arguments to be included for compatibility with query(), # but ignore them. if kw in keywords: del keywords[kw] query = self.format_query(area, date, raw, area_relation, **keywords) _, total_count = self._load_query(query, limit=0) return total_count def _load_query(self, query, order_by=None, limit=None, offset=0): products, count = self._load_subquery(query, order_by, limit, offset) # repeat query until all results have been loaded max_offset = count if limit is not None: max_offset = min(count, offset + limit) if max_offset > offset + self.page_size: progress = self._tqdm(desc="Querying products", initial=self.page_size, total=max_offset - offset, unit=' products') for new_offset in range(offset + self.page_size, max_offset, self.page_size): new_limit = limit if limit is not None: new_limit = limit - new_offset + offset ret = self._load_subquery(query, order_by, new_limit, new_offset)[0] progress.update(len(ret)) products += ret progress.close() return products, count def _load_subquery(self, query, order_by=None, limit=None, offset=0): # store last query (for testing) self._last_query = query self.logger.debug("Sub-query: offset=%s, limit=%s", offset, limit) # load query results url = self._format_url(order_by, limit, offset) response = self.session.post(url, {'q': query}, auth=self.session.auth, headers={'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8'}, timeout=self.timeout) _check_scihub_response(response) # store last status code (for testing) self._last_response = response # parse response content try: json_feed = response.json()['feed'] if json_feed['opensearch:totalResults'] is None: # We are using some unintended behavior of the server that a null is # returned as the total results value when the query string was incorrect. raise SentinelAPIError( 'Invalid query string. Check the parameters and format.', response) total_results = int(json_feed['opensearch:totalResults']) except (ValueError, KeyError): raise SentinelAPIError('API response not valid. JSON decoding failed.', response) products = json_feed.get('entry', []) # this verification is necessary because if the query returns only # one product, self.products will be a dict not a list if isinstance(products, dict): products = [products] return products, total_results def _format_url(self, order_by=None, limit=None, offset=0): if limit is None: limit = self.page_size limit = min(limit, self.page_size) url = 'search?format=json&rows={}'.format(limit) url += '&start={}'.format(offset) if order_by: url += '&orderby={}'.format(order_by) return urljoin(self.api_url, url) @staticmethod def to_geojson(products): """Return the products from a query response as a GeoJSON with the values in their appropriate Python types. """ feature_list = [] for i, (product_id, props) in enumerate(products.items()): props = props.copy() props['id'] = product_id poly = geomet.wkt.loads(props['footprint']) del props['footprint'] del props['gmlfootprint'] # Fix "'datetime' is not JSON serializable" for k, v in props.items(): if isinstance(v, (date, datetime)): props[k] = v.strftime('%Y-%m-%dT%H:%M:%S.%fZ') feature_list.append( geojson.Feature(geometry=poly, id=i, properties=props) ) return geojson.FeatureCollection(feature_list) @staticmethod def to_dataframe(products): """Return the products from a query response as a Pandas DataFrame with the values in their appropriate Python types. """ try: import pandas as pd except ImportError: raise ImportError("to_dataframe requires the optional dependency Pandas.") return pd.DataFrame.from_dict(products, orient='index') @staticmethod def to_geodataframe(products): """Return the products from a query response as a GeoPandas GeoDataFrame with the values in their appropriate Python types. """ try: import geopandas as gpd import shapely.wkt except ImportError: raise ImportError("to_geodataframe requires the optional dependencies GeoPandas and Shapely.") crs = {'init': 'epsg:4326'} # WGS84 if len(products) == 0: return gpd.GeoDataFrame(crs=crs) df = SentinelAPI.to_dataframe(products) geometry = [shapely.wkt.loads(fp) for fp in df['footprint']] # remove useless columns df.drop(['footprint', 'gmlfootprint'], axis=1, inplace=True) return gpd.GeoDataFrame(df, crs=crs, geometry=geometry) def get_product_odata(self, id, full=False): """Access OData API to get info about a product. Returns a dict containing the id, title, size, md5sum, date, footprint and download url of the product. The date field corresponds to the Start ContentDate value. If `full` is set to True, then the full, detailed metadata of the product is returned in addition to the above. Parameters ---------- id : string The UUID of the product to query full : bool Whether to get the full metadata for the Product. False by default. Returns ------- dict[str, Any] A dictionary with an item for each metadata attribute Notes ----- For a full list of mappings between the OpenSearch (Solr) and OData attribute names see the following definition files: https://github.com/SentinelDataHub/DataHubSystem/blob/master/addon/sentinel-1/src/main/resources/META-INF/sentinel-1.owl https://github.com/SentinelDataHub/DataHubSystem/blob/master/addon/sentinel-2/src/main/resources/META-INF/sentinel-2.owl https://github.com/SentinelDataHub/DataHubSystem/blob/master/addon/sentinel-3/src/main/resources/META-INF/sentinel-3.owl """ url = urljoin(self.api_url, u"odata/v1/Products('{}')?$format=json".format(id)) if full: url += '&$expand=Attributes' response = self.session.get(url, auth=self.session.auth, timeout=self.timeout) _check_scihub_response(response) values = _parse_odata_response(response.json()['d']) return values def _trigger_offline_retrieval(self, url): """ Triggers retrieval of an offline product Trying to download an offline product triggers its retrieval from the long term archive. The returned HTTP status code conveys whether this was successful. Parameters ---------- url : string URL for downloading the product Notes ----- https://scihub.copernicus.eu/userguide/LongTermArchive """ with self.session.get(url, auth=self.session.auth, timeout=self.timeout) as r: # check https://scihub.copernicus.eu/userguide/LongTermArchive#HTTP_Status_codes if r.status_code == 202: self.logger.info("Accepted for retrieval") elif r.status_code == 503: self.logger.error("Request not accepted") raise SentinelAPILTAError('Request for retrieval from LTA not accepted', r) elif r.status_code == 403: self.logger.error("Requests exceed user quota") raise SentinelAPILTAError('Requests for retrieval from LTA exceed user quota', r) elif r.status_code == 500: # should not happen self.logger.error("Trying to download an offline product") raise SentinelAPILTAError('Trying to download an offline product', r) return r.status_code def download(self, id, directory_path='.', checksum=True): """Download a product. Uses the filename on the server for the downloaded file, e.g. "S1A_EW_GRDH_1SDH_20141003T003840_20141003T003920_002658_002F54_4DD1.zip". Incomplete downloads are continued and complete files are skipped. Parameters ---------- id : string UUID of the product, e.g. 'a8dd0cfd-613e-45ce-868c-d79177b916ed' directory_path : string, optional Where the file will be downloaded checksum : bool, optional If True, verify the downloaded file's integrity by checking its MD5 checksum. Throws InvalidChecksumError if the checksum does not match. Defaults to True. Returns ------- product_info : dict Dictionary containing the product's info from get_product_info() as well as the path on disk. Raises ------ InvalidChecksumError If the MD5 checksum does not match the checksum on the server. """ product_info = self.get_product_odata(id) path = join(directory_path, product_info['title'] + '.zip') product_info['path'] = path product_info['downloaded_bytes'] = 0 self.logger.info('Downloading %s to %s', id, path) if exists(path): # We assume that the product has been downloaded and is complete return product_info # An incomplete download triggers the retrieval from the LTA if the product is not online if not product_info['Online']: self.logger.warning( 'Product %s is not online. Triggering retrieval from long term archive.', product_info['id']) self._trigger_offline_retrieval(product_info['url']) return product_info # Use a temporary file for downloading temp_path = path + '.incomplete' skip_download = False if exists(temp_path): if getsize(temp_path) > product_info['size']: self.logger.warning( "Existing incomplete file %s is larger than the expected final size" " (%s vs %s bytes). Deleting it.", str(temp_path), getsize(temp_path), product_info['size']) remove(temp_path) elif getsize(temp_path) == product_info['size']: if self._md5_compare(temp_path, product_info['md5']): skip_download = True else: # Log a warning since this should never happen self.logger.warning( "Existing incomplete file %s appears to be fully downloaded but " "its checksum is incorrect. Deleting it.", str(temp_path)) remove(temp_path) else: # continue downloading self.logger.info( "Download will resume from existing incomplete file %s.", temp_path) pass if not skip_download: # Store the number of downloaded bytes for unit tests product_info['downloaded_bytes'] = self._download( product_info['url'], temp_path, self.session, product_info['size']) # Check integrity with MD5 checksum if checksum is True: if not self._md5_compare(temp_path, product_info['md5']): remove(temp_path) raise InvalidChecksumError('File corrupt: checksums do not match') # Download successful, rename the temporary file to its proper name shutil.move(temp_path, path) return product_info def download_all(self, products, directory_path='.', max_attempts=10, checksum=True): """Download a list of products. Takes a list of product IDs as input. This means that the return value of query() can be passed directly to this method. File names on the server are used for the downloaded files, e.g. "S1A_EW_GRDH_1SDH_20141003T003840_20141003T003920_002658_002F54_4DD1.zip". In case of interruptions or other exceptions, downloading will restart from where it left off. Downloading is attempted at most max_attempts times to avoid getting stuck with unrecoverable errors. Parameters ---------- products : list List of product IDs directory_path : string Directory where the downloaded files will be downloaded max_attempts : int, optional Number of allowed retries before giving up downloading a product. Defaults to 10. checksum : bool, optional If True, verify the downloaded files' integrity by checking its MD5 checksum. Throws InvalidChecksumError if the checksum does not match. Defaults to True. Raises ------ Raises the most recent downloading exception if all downloads failed. Returns ------- dict[string, dict] A dictionary containing the return value from download() for each successfully downloaded product. dict[string, dict] A dictionary containing the product information for products whose retrieval from the long term archive was successfully triggered. set[string] The list of products that failed to download. """ product_ids = list(products) self.logger.info("Will download %d products", len(product_ids)) return_values = OrderedDict() last_exception = None for i, product_id in enumerate(products): for attempt_num in range(max_attempts): try: product_info = self.download(product_id, directory_path, checksum) return_values[product_id] = product_info break except (KeyboardInterrupt, SystemExit): raise except InvalidChecksumError as e: last_exception = e self.logger.warning( "Invalid checksum. The downloaded file for '%s' is corrupted.", product_id) except SentinelAPILTAError as e: last_exception = e self.logger.exception("There was an error retrieving %s from the LTA", product_id) break except Exception as e: last_exception = e self.logger.exception("There was an error downloading %s", product_id) self.logger.info("%s/%s products downloaded", i + 1, len(product_ids)) failed = set(products) - set(return_values) # split up sucessfully processed products into downloaded and only triggered retrieval from the LTA triggered = OrderedDict([(k, v) for k, v in return_values.items() if v['Online'] is False]) downloaded = OrderedDict([(k, v) for k, v in return_values.items() if v['Online'] is True]) if len(failed) == len(product_ids) and last_exception is not None: raise last_exception return downloaded, triggered, failed @staticmethod @staticmethod def check_query_length(query): """Determine whether a query to the OpenSearch API is too long. The length of a query string is limited to approximately 3938 characters but any special characters (that is, not alphanumeric or -_.*) will take up more space. Parameters ---------- query : str The query string Returns ------- float Ratio of the query length to the maximum length """ # The server uses the Java's URLEncoder implementation internally, which we are replicating here effective_length = len(quote_plus(query, safe="-_.*").replace('~', '%7E')) return effective_length / 3938 def _query_names(self, names): """Find products by their names, e.g. S1A_EW_GRDH_1SDH_20141003T003840_20141003T003920_002658_002F54_4DD1. Note that duplicates exist on server, so multiple products can be returned for each name. Parameters ---------- names : list[string] List of product names. Returns ------- dict[string, dict[str, dict]] A dictionary mapping each name to a dictionary which contains the products with that name (with ID as the key). """ def chunks(l, n): """Yield successive n-sized chunks from l.""" for i in range(0, len(l), n): yield l[i:i + n] products = {} # 40 names per query fits reasonably well inside the query limit for chunk in chunks(names, 40): query = " OR ".join(chunk) products.update(self.query(raw=query)) # Group the products output = OrderedDict((name, dict()) for name in names) for id, metadata in products.items(): name = metadata['identifier'] output[name][id] = metadata return output def check_files(self, paths=None, ids=None, directory=None, delete=False): """Verify the integrity of product files on disk. Integrity is checked by comparing the size and checksum of the file with the respective values on the server. The input can be a list of products to check or a list of IDs and a directory. In cases where multiple products with different IDs exist on the server for given product name, the file is considered to be correct if any of them matches the file size and checksum. A warning is logged in such situations. The corrupt products' OData info is included in the return value to make it easier to re-download the products, if necessary. Parameters ---------- paths : list[string] List of product file paths. ids : list[string] List of product IDs. directory : string Directory where the files are located, if checking based on product IDs. delete : bool Whether to delete corrupt products. Defaults to False. Returns ------- dict[str, list[dict]] A dictionary listing the invalid or missing files. The dictionary maps the corrupt file paths to a list of OData dictionaries of matching products on the server (as returned by :meth:`SentinelAPI.get_product_odata()`). """ if not ids and not paths: raise ValueError("Must provide either file paths or product IDs and a directory") if ids and not directory: raise ValueError("Directory value missing") paths = paths or [] ids = ids or [] def name_from_path(path): return splitext(basename(path))[0] # Get product IDs corresponding to the files on disk names = [] if paths: names = list(map(name_from_path, paths)) result = self._query_names(names) for product_dicts in result.values(): ids += list(product_dicts) names_from_paths = set(names) ids = set(ids) # Collect the OData information for each product # Product name -> list of matching odata dicts product_infos = defaultdict(list) for id in ids: odata = self.get_product_odata(id) name = odata['title'] product_infos[name].append(odata) # Collect if name not in names_from_paths: paths.append(join(directory, name + '.zip')) # Now go over the list of products and check them corrupt = {} for path in paths: name = name_from_path(path) if len(product_infos[name]) > 1: self.logger.warning("{} matches multiple products on server".format(path)) if not exists(path): # We will consider missing files as corrupt also self.logger.info("{} does not exist on disk".format(path)) corrupt[path] = product_infos[name] continue is_fine = False for product_info in product_infos[name]: if (getsize(path) == product_info['size'] and self._md5_compare(path, product_info['md5'])): is_fine = True break if not is_fine: self.logger.info("{} is corrupt".format(path)) corrupt[path] = product_infos[name] if delete: remove(path) return corrupt def _md5_compare(self, file_path, checksum, block_size=2 ** 13): """Compare a given MD5 checksum with one calculated from a file.""" with closing(self._tqdm(desc="MD5 checksumming", total=getsize(file_path), unit="B", unit_scale=True)) as progress: md5 = hashlib.md5() with open(file_path, "rb") as f: while True: block_data = f.read(block_size) if not block_data: break md5.update(block_data) progress.update(len(block_data)) return md5.hexdigest().lower() == checksum.lower() def _download(self, url, path, session, file_size): headers = {} continuing = exists(path) if continuing: already_downloaded_bytes = getsize(path) headers = {'Range': 'bytes={}-'.format(already_downloaded_bytes)} else: already_downloaded_bytes = 0 downloaded_bytes = 0 with closing(session.get(url, stream=True, auth=session.auth, headers=headers, timeout=self.timeout)) as r, \ closing(self._tqdm(desc="Downloading", total=file_size, unit="B", unit_scale=True, initial=already_downloaded_bytes)) as progress: _check_scihub_response(r, test_json=False) chunk_size = 2 ** 20 # download in 1 MB chunks mode = 'ab' if continuing else 'wb' with open(path, mode) as f: for chunk in r.iter_content(chunk_size=chunk_size): if chunk: # filter out keep-alive new chunks f.write(chunk) progress.update(len(chunk)) downloaded_bytes += len(chunk) # Return the number of bytes downloaded return downloaded_bytes def _tqdm(self, **kwargs): """tqdm progressbar wrapper. May be overridden to customize progressbar behavior""" kwargs.update({'disable': not self.show_progressbars}) return tqdm(**kwargs)
sentinelsat/sentinelsat
sentinelsat/sentinel.py
SentinelAPI._query_names
python
def _query_names(self, names): def chunks(l, n): """Yield successive n-sized chunks from l.""" for i in range(0, len(l), n): yield l[i:i + n] products = {} # 40 names per query fits reasonably well inside the query limit for chunk in chunks(names, 40): query = " OR ".join(chunk) products.update(self.query(raw=query)) # Group the products output = OrderedDict((name, dict()) for name in names) for id, metadata in products.items(): name = metadata['identifier'] output[name][id] = metadata return output
Find products by their names, e.g. S1A_EW_GRDH_1SDH_20141003T003840_20141003T003920_002658_002F54_4DD1. Note that duplicates exist on server, so multiple products can be returned for each name. Parameters ---------- names : list[string] List of product names. Returns ------- dict[string, dict[str, dict]] A dictionary mapping each name to a dictionary which contains the products with that name (with ID as the key).
train
https://github.com/sentinelsat/sentinelsat/blob/eacfd79ff4e7e939147db9dfdd393c67d64eecaa/sentinelsat/sentinel.py#L665-L700
null
class SentinelAPI: """Class to connect to Copernicus Open Access Hub, search and download imagery. Parameters ---------- user : string username for DataHub set to None to use ~/.netrc password : string password for DataHub set to None to use ~/.netrc api_url : string, optional URL of the DataHub defaults to 'https://scihub.copernicus.eu/apihub' show_progressbars : bool Whether progressbars should be shown or not, e.g. during download. Defaults to True. timeout : float or tuple, optional How long to wait for DataHub response (in seconds). Tuple (connect, read) allowed. Attributes ---------- session : requests.Session Session to connect to DataHub api_url : str URL to the DataHub page_size : int Number of results per query page. Current value: 100 (maximum allowed on ApiHub) timeout : float or tuple How long to wait for DataHub response (in seconds). """ logger = logging.getLogger('sentinelsat.SentinelAPI') def __init__(self, user, password, api_url='https://scihub.copernicus.eu/apihub/', show_progressbars=True, timeout=None): self.session = requests.Session() if user and password: self.session.auth = (user, password) self.api_url = api_url if api_url.endswith('/') else api_url + '/' self.page_size = 100 self.user_agent = 'sentinelsat/' + sentinelsat_version self.session.headers['User-Agent'] = self.user_agent self.show_progressbars = show_progressbars self.timeout = timeout # For unit tests self._last_query = None self._last_response = None def query(self, area=None, date=None, raw=None, area_relation='Intersects', order_by=None, limit=None, offset=0, **keywords): """Query the OpenSearch API with the coordinates of an area, a date interval and any other search keywords accepted by the API. Parameters ---------- area : str, optional The area of interest formatted as a Well-Known Text string. date : tuple of (str or datetime) or str, optional A time interval filter based on the Sensing Start Time of the products. Expects a tuple of (start, end), e.g. ("NOW-1DAY", "NOW"). The timestamps can be either a Python datetime or a string in one of the following formats: - yyyyMMdd - yyyy-MM-ddThh:mm:ss.SSSZ (ISO-8601) - yyyy-MM-ddThh:mm:ssZ - NOW - NOW-<n>DAY(S) (or HOUR(S), MONTH(S), etc.) - NOW+<n>DAY(S) - yyyy-MM-ddThh:mm:ssZ-<n>DAY(S) - NOW/DAY (or HOUR, MONTH etc.) - rounds the value to the given unit Alternatively, an already fully formatted string such as "[NOW-1DAY TO NOW]" can be used as well. raw : str, optional Additional query text that will be appended to the query. area_relation : {'Intersects', 'Contains', 'IsWithin'}, optional What relation to use for testing the AOI. Case insensitive. - Intersects: true if the AOI and the footprint intersect (default) - Contains: true if the AOI is inside the footprint - IsWithin: true if the footprint is inside the AOI order_by: str, optional A comma-separated list of fields to order by (on server side). Prefix the field name by '+' or '-' to sort in ascending or descending order, respectively. Ascending order is used if prefix is omitted. Example: "cloudcoverpercentage, -beginposition". limit: int, optional Maximum number of products returned. Defaults to no limit. offset: int, optional The number of results to skip. Defaults to 0. **keywords Additional keywords can be used to specify other query parameters, e.g. `relativeorbitnumber=70`. See https://scihub.copernicus.eu/twiki/do/view/SciHubUserGuide/3FullTextSearch for a full list. Range values can be passed as two-element tuples, e.g. `cloudcoverpercentage=(0, 30)`. `None` can be used in range values for one-sided ranges, e.g. `orbitnumber=(16302, None)`. Ranges with no bounds (`orbitnumber=(None, None)`) will not be included in the query. The time interval formats accepted by the `date` parameter can also be used with any other parameters that expect time intervals (that is: 'beginposition', 'endposition', 'date', 'creationdate', and 'ingestiondate'). Returns ------- dict[string, dict] Products returned by the query as a dictionary with the product ID as the key and the product's attributes (a dictionary) as the value. """ query = self.format_query(area, date, raw, area_relation, **keywords) self.logger.debug("Running query: order_by=%s, limit=%s, offset=%s, query=%s", order_by, limit, offset, query) formatted_order_by = _format_order_by(order_by) response, count = self._load_query(query, formatted_order_by, limit, offset) self.logger.info("Found %s products", count) return _parse_opensearch_response(response) @staticmethod def format_query(area=None, date=None, raw=None, area_relation='Intersects', **keywords): """Create a OpenSearch API query string. """ if area_relation.lower() not in {"intersects", "contains", "iswithin"}: raise ValueError("Incorrect AOI relation provided ({})".format(area_relation)) # Check for duplicate keywords kw_lower = set(x.lower() for x in keywords) if (len(kw_lower) != len(keywords) or (date is not None and 'beginposition' in kw_lower) or (area is not None and 'footprint' in kw_lower)): raise ValueError("Query contains duplicate keywords. Note that query keywords are case-insensitive.") query_parts = [] if date is not None: keywords['beginPosition'] = date for attr, value in sorted(keywords.items()): # Escape spaces, where appropriate if isinstance(value, string_types): value = value.strip() if not any(value.startswith(s[0]) and value.endswith(s[1]) for s in ['[]', '{}', '//', '()']): value = re.sub(r'\s', r'\ ', value, re.M) # Handle date keywords # Keywords from https://github.com/SentinelDataHub/DataHubSystem/search?q=text/date+iso8601 date_attrs = ['beginposition', 'endposition', 'date', 'creationdate', 'ingestiondate'] if attr.lower() in date_attrs: # Automatically format date-type attributes if isinstance(value, string_types) and ' TO ' in value: # This is a string already formatted as a date interval, # e.g. '[NOW-1DAY TO NOW]' pass elif not isinstance(value, string_types) and len(value) == 2: value = (format_query_date(value[0]), format_query_date(value[1])) else: raise ValueError("Date-type query parameter '{}' expects a two-element tuple " "of str or datetime objects. Received {}".format(attr, value)) # Handle ranged values if isinstance(value, (list, tuple)): # Handle value ranges if len(value) == 2: # Allow None to be used as a unlimited bound value = ['*' if x is None else x for x in value] if all(x == '*' for x in value): continue value = '[{} TO {}]'.format(*value) else: raise ValueError("Invalid number of elements in list. Expected 2, received " "{}".format(len(value))) query_parts.append('{}:{}'.format(attr, value)) if raw: query_parts.append(raw) if area is not None: query_parts.append('footprint:"{}({})"'.format(area_relation, area)) return ' '.join(query_parts) def query_raw(self, query, order_by=None, limit=None, offset=0): """ Do a full-text query on the OpenSearch API using the format specified in https://scihub.copernicus.eu/twiki/do/view/SciHubUserGuide/3FullTextSearch DEPRECATED: use :meth:`query(raw=...) <.query>` instead. This method will be removed in the next major release. Parameters ---------- query : str The query string. order_by: str, optional A comma-separated list of fields to order by (on server side). Prefix the field name by '+' or '-' to sort in ascending or descending order, respectively. Ascending order is used, if prefix is omitted. Example: "cloudcoverpercentage, -beginposition". limit: int, optional Maximum number of products returned. Defaults to no limit. offset: int, optional The number of results to skip. Defaults to 0. Returns ------- dict[string, dict] Products returned by the query as a dictionary with the product ID as the key and the product's attributes (a dictionary) as the value. """ warnings.warn( "query_raw() has been merged with query(). use query(raw=...) instead.", PendingDeprecationWarning ) return self.query(raw=query, order_by=order_by, limit=limit, offset=offset) def count(self, area=None, date=None, raw=None, area_relation='Intersects', **keywords): """Get the number of products matching a query. Accepted parameters are identical to :meth:`SentinelAPI.query()`. This is a significantly more efficient alternative to doing `len(api.query())`, which can take minutes to run for queries matching thousands of products. Returns ------- int The number of products matching a query. """ for kw in ['order_by', 'limit', 'offset']: # Allow these function arguments to be included for compatibility with query(), # but ignore them. if kw in keywords: del keywords[kw] query = self.format_query(area, date, raw, area_relation, **keywords) _, total_count = self._load_query(query, limit=0) return total_count def _load_query(self, query, order_by=None, limit=None, offset=0): products, count = self._load_subquery(query, order_by, limit, offset) # repeat query until all results have been loaded max_offset = count if limit is not None: max_offset = min(count, offset + limit) if max_offset > offset + self.page_size: progress = self._tqdm(desc="Querying products", initial=self.page_size, total=max_offset - offset, unit=' products') for new_offset in range(offset + self.page_size, max_offset, self.page_size): new_limit = limit if limit is not None: new_limit = limit - new_offset + offset ret = self._load_subquery(query, order_by, new_limit, new_offset)[0] progress.update(len(ret)) products += ret progress.close() return products, count def _load_subquery(self, query, order_by=None, limit=None, offset=0): # store last query (for testing) self._last_query = query self.logger.debug("Sub-query: offset=%s, limit=%s", offset, limit) # load query results url = self._format_url(order_by, limit, offset) response = self.session.post(url, {'q': query}, auth=self.session.auth, headers={'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8'}, timeout=self.timeout) _check_scihub_response(response) # store last status code (for testing) self._last_response = response # parse response content try: json_feed = response.json()['feed'] if json_feed['opensearch:totalResults'] is None: # We are using some unintended behavior of the server that a null is # returned as the total results value when the query string was incorrect. raise SentinelAPIError( 'Invalid query string. Check the parameters and format.', response) total_results = int(json_feed['opensearch:totalResults']) except (ValueError, KeyError): raise SentinelAPIError('API response not valid. JSON decoding failed.', response) products = json_feed.get('entry', []) # this verification is necessary because if the query returns only # one product, self.products will be a dict not a list if isinstance(products, dict): products = [products] return products, total_results def _format_url(self, order_by=None, limit=None, offset=0): if limit is None: limit = self.page_size limit = min(limit, self.page_size) url = 'search?format=json&rows={}'.format(limit) url += '&start={}'.format(offset) if order_by: url += '&orderby={}'.format(order_by) return urljoin(self.api_url, url) @staticmethod def to_geojson(products): """Return the products from a query response as a GeoJSON with the values in their appropriate Python types. """ feature_list = [] for i, (product_id, props) in enumerate(products.items()): props = props.copy() props['id'] = product_id poly = geomet.wkt.loads(props['footprint']) del props['footprint'] del props['gmlfootprint'] # Fix "'datetime' is not JSON serializable" for k, v in props.items(): if isinstance(v, (date, datetime)): props[k] = v.strftime('%Y-%m-%dT%H:%M:%S.%fZ') feature_list.append( geojson.Feature(geometry=poly, id=i, properties=props) ) return geojson.FeatureCollection(feature_list) @staticmethod def to_dataframe(products): """Return the products from a query response as a Pandas DataFrame with the values in their appropriate Python types. """ try: import pandas as pd except ImportError: raise ImportError("to_dataframe requires the optional dependency Pandas.") return pd.DataFrame.from_dict(products, orient='index') @staticmethod def to_geodataframe(products): """Return the products from a query response as a GeoPandas GeoDataFrame with the values in their appropriate Python types. """ try: import geopandas as gpd import shapely.wkt except ImportError: raise ImportError("to_geodataframe requires the optional dependencies GeoPandas and Shapely.") crs = {'init': 'epsg:4326'} # WGS84 if len(products) == 0: return gpd.GeoDataFrame(crs=crs) df = SentinelAPI.to_dataframe(products) geometry = [shapely.wkt.loads(fp) for fp in df['footprint']] # remove useless columns df.drop(['footprint', 'gmlfootprint'], axis=1, inplace=True) return gpd.GeoDataFrame(df, crs=crs, geometry=geometry) def get_product_odata(self, id, full=False): """Access OData API to get info about a product. Returns a dict containing the id, title, size, md5sum, date, footprint and download url of the product. The date field corresponds to the Start ContentDate value. If `full` is set to True, then the full, detailed metadata of the product is returned in addition to the above. Parameters ---------- id : string The UUID of the product to query full : bool Whether to get the full metadata for the Product. False by default. Returns ------- dict[str, Any] A dictionary with an item for each metadata attribute Notes ----- For a full list of mappings between the OpenSearch (Solr) and OData attribute names see the following definition files: https://github.com/SentinelDataHub/DataHubSystem/blob/master/addon/sentinel-1/src/main/resources/META-INF/sentinel-1.owl https://github.com/SentinelDataHub/DataHubSystem/blob/master/addon/sentinel-2/src/main/resources/META-INF/sentinel-2.owl https://github.com/SentinelDataHub/DataHubSystem/blob/master/addon/sentinel-3/src/main/resources/META-INF/sentinel-3.owl """ url = urljoin(self.api_url, u"odata/v1/Products('{}')?$format=json".format(id)) if full: url += '&$expand=Attributes' response = self.session.get(url, auth=self.session.auth, timeout=self.timeout) _check_scihub_response(response) values = _parse_odata_response(response.json()['d']) return values def _trigger_offline_retrieval(self, url): """ Triggers retrieval of an offline product Trying to download an offline product triggers its retrieval from the long term archive. The returned HTTP status code conveys whether this was successful. Parameters ---------- url : string URL for downloading the product Notes ----- https://scihub.copernicus.eu/userguide/LongTermArchive """ with self.session.get(url, auth=self.session.auth, timeout=self.timeout) as r: # check https://scihub.copernicus.eu/userguide/LongTermArchive#HTTP_Status_codes if r.status_code == 202: self.logger.info("Accepted for retrieval") elif r.status_code == 503: self.logger.error("Request not accepted") raise SentinelAPILTAError('Request for retrieval from LTA not accepted', r) elif r.status_code == 403: self.logger.error("Requests exceed user quota") raise SentinelAPILTAError('Requests for retrieval from LTA exceed user quota', r) elif r.status_code == 500: # should not happen self.logger.error("Trying to download an offline product") raise SentinelAPILTAError('Trying to download an offline product', r) return r.status_code def download(self, id, directory_path='.', checksum=True): """Download a product. Uses the filename on the server for the downloaded file, e.g. "S1A_EW_GRDH_1SDH_20141003T003840_20141003T003920_002658_002F54_4DD1.zip". Incomplete downloads are continued and complete files are skipped. Parameters ---------- id : string UUID of the product, e.g. 'a8dd0cfd-613e-45ce-868c-d79177b916ed' directory_path : string, optional Where the file will be downloaded checksum : bool, optional If True, verify the downloaded file's integrity by checking its MD5 checksum. Throws InvalidChecksumError if the checksum does not match. Defaults to True. Returns ------- product_info : dict Dictionary containing the product's info from get_product_info() as well as the path on disk. Raises ------ InvalidChecksumError If the MD5 checksum does not match the checksum on the server. """ product_info = self.get_product_odata(id) path = join(directory_path, product_info['title'] + '.zip') product_info['path'] = path product_info['downloaded_bytes'] = 0 self.logger.info('Downloading %s to %s', id, path) if exists(path): # We assume that the product has been downloaded and is complete return product_info # An incomplete download triggers the retrieval from the LTA if the product is not online if not product_info['Online']: self.logger.warning( 'Product %s is not online. Triggering retrieval from long term archive.', product_info['id']) self._trigger_offline_retrieval(product_info['url']) return product_info # Use a temporary file for downloading temp_path = path + '.incomplete' skip_download = False if exists(temp_path): if getsize(temp_path) > product_info['size']: self.logger.warning( "Existing incomplete file %s is larger than the expected final size" " (%s vs %s bytes). Deleting it.", str(temp_path), getsize(temp_path), product_info['size']) remove(temp_path) elif getsize(temp_path) == product_info['size']: if self._md5_compare(temp_path, product_info['md5']): skip_download = True else: # Log a warning since this should never happen self.logger.warning( "Existing incomplete file %s appears to be fully downloaded but " "its checksum is incorrect. Deleting it.", str(temp_path)) remove(temp_path) else: # continue downloading self.logger.info( "Download will resume from existing incomplete file %s.", temp_path) pass if not skip_download: # Store the number of downloaded bytes for unit tests product_info['downloaded_bytes'] = self._download( product_info['url'], temp_path, self.session, product_info['size']) # Check integrity with MD5 checksum if checksum is True: if not self._md5_compare(temp_path, product_info['md5']): remove(temp_path) raise InvalidChecksumError('File corrupt: checksums do not match') # Download successful, rename the temporary file to its proper name shutil.move(temp_path, path) return product_info def download_all(self, products, directory_path='.', max_attempts=10, checksum=True): """Download a list of products. Takes a list of product IDs as input. This means that the return value of query() can be passed directly to this method. File names on the server are used for the downloaded files, e.g. "S1A_EW_GRDH_1SDH_20141003T003840_20141003T003920_002658_002F54_4DD1.zip". In case of interruptions or other exceptions, downloading will restart from where it left off. Downloading is attempted at most max_attempts times to avoid getting stuck with unrecoverable errors. Parameters ---------- products : list List of product IDs directory_path : string Directory where the downloaded files will be downloaded max_attempts : int, optional Number of allowed retries before giving up downloading a product. Defaults to 10. checksum : bool, optional If True, verify the downloaded files' integrity by checking its MD5 checksum. Throws InvalidChecksumError if the checksum does not match. Defaults to True. Raises ------ Raises the most recent downloading exception if all downloads failed. Returns ------- dict[string, dict] A dictionary containing the return value from download() for each successfully downloaded product. dict[string, dict] A dictionary containing the product information for products whose retrieval from the long term archive was successfully triggered. set[string] The list of products that failed to download. """ product_ids = list(products) self.logger.info("Will download %d products", len(product_ids)) return_values = OrderedDict() last_exception = None for i, product_id in enumerate(products): for attempt_num in range(max_attempts): try: product_info = self.download(product_id, directory_path, checksum) return_values[product_id] = product_info break except (KeyboardInterrupt, SystemExit): raise except InvalidChecksumError as e: last_exception = e self.logger.warning( "Invalid checksum. The downloaded file for '%s' is corrupted.", product_id) except SentinelAPILTAError as e: last_exception = e self.logger.exception("There was an error retrieving %s from the LTA", product_id) break except Exception as e: last_exception = e self.logger.exception("There was an error downloading %s", product_id) self.logger.info("%s/%s products downloaded", i + 1, len(product_ids)) failed = set(products) - set(return_values) # split up sucessfully processed products into downloaded and only triggered retrieval from the LTA triggered = OrderedDict([(k, v) for k, v in return_values.items() if v['Online'] is False]) downloaded = OrderedDict([(k, v) for k, v in return_values.items() if v['Online'] is True]) if len(failed) == len(product_ids) and last_exception is not None: raise last_exception return downloaded, triggered, failed @staticmethod def get_products_size(products): """Return the total file size in GB of all products in the OpenSearch response.""" size_total = 0 for title, props in products.items(): size_product = props["size"] size_value = float(size_product.split(" ")[0]) size_unit = str(size_product.split(" ")[1]) if size_unit == "MB": size_value /= 1024. if size_unit == "KB": size_value /= 1024. * 1024. size_total += size_value return round(size_total, 2) @staticmethod def check_query_length(query): """Determine whether a query to the OpenSearch API is too long. The length of a query string is limited to approximately 3938 characters but any special characters (that is, not alphanumeric or -_.*) will take up more space. Parameters ---------- query : str The query string Returns ------- float Ratio of the query length to the maximum length """ # The server uses the Java's URLEncoder implementation internally, which we are replicating here effective_length = len(quote_plus(query, safe="-_.*").replace('~', '%7E')) return effective_length / 3938 def check_files(self, paths=None, ids=None, directory=None, delete=False): """Verify the integrity of product files on disk. Integrity is checked by comparing the size and checksum of the file with the respective values on the server. The input can be a list of products to check or a list of IDs and a directory. In cases where multiple products with different IDs exist on the server for given product name, the file is considered to be correct if any of them matches the file size and checksum. A warning is logged in such situations. The corrupt products' OData info is included in the return value to make it easier to re-download the products, if necessary. Parameters ---------- paths : list[string] List of product file paths. ids : list[string] List of product IDs. directory : string Directory where the files are located, if checking based on product IDs. delete : bool Whether to delete corrupt products. Defaults to False. Returns ------- dict[str, list[dict]] A dictionary listing the invalid or missing files. The dictionary maps the corrupt file paths to a list of OData dictionaries of matching products on the server (as returned by :meth:`SentinelAPI.get_product_odata()`). """ if not ids and not paths: raise ValueError("Must provide either file paths or product IDs and a directory") if ids and not directory: raise ValueError("Directory value missing") paths = paths or [] ids = ids or [] def name_from_path(path): return splitext(basename(path))[0] # Get product IDs corresponding to the files on disk names = [] if paths: names = list(map(name_from_path, paths)) result = self._query_names(names) for product_dicts in result.values(): ids += list(product_dicts) names_from_paths = set(names) ids = set(ids) # Collect the OData information for each product # Product name -> list of matching odata dicts product_infos = defaultdict(list) for id in ids: odata = self.get_product_odata(id) name = odata['title'] product_infos[name].append(odata) # Collect if name not in names_from_paths: paths.append(join(directory, name + '.zip')) # Now go over the list of products and check them corrupt = {} for path in paths: name = name_from_path(path) if len(product_infos[name]) > 1: self.logger.warning("{} matches multiple products on server".format(path)) if not exists(path): # We will consider missing files as corrupt also self.logger.info("{} does not exist on disk".format(path)) corrupt[path] = product_infos[name] continue is_fine = False for product_info in product_infos[name]: if (getsize(path) == product_info['size'] and self._md5_compare(path, product_info['md5'])): is_fine = True break if not is_fine: self.logger.info("{} is corrupt".format(path)) corrupt[path] = product_infos[name] if delete: remove(path) return corrupt def _md5_compare(self, file_path, checksum, block_size=2 ** 13): """Compare a given MD5 checksum with one calculated from a file.""" with closing(self._tqdm(desc="MD5 checksumming", total=getsize(file_path), unit="B", unit_scale=True)) as progress: md5 = hashlib.md5() with open(file_path, "rb") as f: while True: block_data = f.read(block_size) if not block_data: break md5.update(block_data) progress.update(len(block_data)) return md5.hexdigest().lower() == checksum.lower() def _download(self, url, path, session, file_size): headers = {} continuing = exists(path) if continuing: already_downloaded_bytes = getsize(path) headers = {'Range': 'bytes={}-'.format(already_downloaded_bytes)} else: already_downloaded_bytes = 0 downloaded_bytes = 0 with closing(session.get(url, stream=True, auth=session.auth, headers=headers, timeout=self.timeout)) as r, \ closing(self._tqdm(desc="Downloading", total=file_size, unit="B", unit_scale=True, initial=already_downloaded_bytes)) as progress: _check_scihub_response(r, test_json=False) chunk_size = 2 ** 20 # download in 1 MB chunks mode = 'ab' if continuing else 'wb' with open(path, mode) as f: for chunk in r.iter_content(chunk_size=chunk_size): if chunk: # filter out keep-alive new chunks f.write(chunk) progress.update(len(chunk)) downloaded_bytes += len(chunk) # Return the number of bytes downloaded return downloaded_bytes def _tqdm(self, **kwargs): """tqdm progressbar wrapper. May be overridden to customize progressbar behavior""" kwargs.update({'disable': not self.show_progressbars}) return tqdm(**kwargs)
sentinelsat/sentinelsat
sentinelsat/sentinel.py
SentinelAPI.check_files
python
def check_files(self, paths=None, ids=None, directory=None, delete=False): if not ids and not paths: raise ValueError("Must provide either file paths or product IDs and a directory") if ids and not directory: raise ValueError("Directory value missing") paths = paths or [] ids = ids or [] def name_from_path(path): return splitext(basename(path))[0] # Get product IDs corresponding to the files on disk names = [] if paths: names = list(map(name_from_path, paths)) result = self._query_names(names) for product_dicts in result.values(): ids += list(product_dicts) names_from_paths = set(names) ids = set(ids) # Collect the OData information for each product # Product name -> list of matching odata dicts product_infos = defaultdict(list) for id in ids: odata = self.get_product_odata(id) name = odata['title'] product_infos[name].append(odata) # Collect if name not in names_from_paths: paths.append(join(directory, name + '.zip')) # Now go over the list of products and check them corrupt = {} for path in paths: name = name_from_path(path) if len(product_infos[name]) > 1: self.logger.warning("{} matches multiple products on server".format(path)) if not exists(path): # We will consider missing files as corrupt also self.logger.info("{} does not exist on disk".format(path)) corrupt[path] = product_infos[name] continue is_fine = False for product_info in product_infos[name]: if (getsize(path) == product_info['size'] and self._md5_compare(path, product_info['md5'])): is_fine = True break if not is_fine: self.logger.info("{} is corrupt".format(path)) corrupt[path] = product_infos[name] if delete: remove(path) return corrupt
Verify the integrity of product files on disk. Integrity is checked by comparing the size and checksum of the file with the respective values on the server. The input can be a list of products to check or a list of IDs and a directory. In cases where multiple products with different IDs exist on the server for given product name, the file is considered to be correct if any of them matches the file size and checksum. A warning is logged in such situations. The corrupt products' OData info is included in the return value to make it easier to re-download the products, if necessary. Parameters ---------- paths : list[string] List of product file paths. ids : list[string] List of product IDs. directory : string Directory where the files are located, if checking based on product IDs. delete : bool Whether to delete corrupt products. Defaults to False. Returns ------- dict[str, list[dict]] A dictionary listing the invalid or missing files. The dictionary maps the corrupt file paths to a list of OData dictionaries of matching products on the server (as returned by :meth:`SentinelAPI.get_product_odata()`).
train
https://github.com/sentinelsat/sentinelsat/blob/eacfd79ff4e7e939147db9dfdd393c67d64eecaa/sentinelsat/sentinel.py#L702-L793
null
class SentinelAPI: """Class to connect to Copernicus Open Access Hub, search and download imagery. Parameters ---------- user : string username for DataHub set to None to use ~/.netrc password : string password for DataHub set to None to use ~/.netrc api_url : string, optional URL of the DataHub defaults to 'https://scihub.copernicus.eu/apihub' show_progressbars : bool Whether progressbars should be shown or not, e.g. during download. Defaults to True. timeout : float or tuple, optional How long to wait for DataHub response (in seconds). Tuple (connect, read) allowed. Attributes ---------- session : requests.Session Session to connect to DataHub api_url : str URL to the DataHub page_size : int Number of results per query page. Current value: 100 (maximum allowed on ApiHub) timeout : float or tuple How long to wait for DataHub response (in seconds). """ logger = logging.getLogger('sentinelsat.SentinelAPI') def __init__(self, user, password, api_url='https://scihub.copernicus.eu/apihub/', show_progressbars=True, timeout=None): self.session = requests.Session() if user and password: self.session.auth = (user, password) self.api_url = api_url if api_url.endswith('/') else api_url + '/' self.page_size = 100 self.user_agent = 'sentinelsat/' + sentinelsat_version self.session.headers['User-Agent'] = self.user_agent self.show_progressbars = show_progressbars self.timeout = timeout # For unit tests self._last_query = None self._last_response = None def query(self, area=None, date=None, raw=None, area_relation='Intersects', order_by=None, limit=None, offset=0, **keywords): """Query the OpenSearch API with the coordinates of an area, a date interval and any other search keywords accepted by the API. Parameters ---------- area : str, optional The area of interest formatted as a Well-Known Text string. date : tuple of (str or datetime) or str, optional A time interval filter based on the Sensing Start Time of the products. Expects a tuple of (start, end), e.g. ("NOW-1DAY", "NOW"). The timestamps can be either a Python datetime or a string in one of the following formats: - yyyyMMdd - yyyy-MM-ddThh:mm:ss.SSSZ (ISO-8601) - yyyy-MM-ddThh:mm:ssZ - NOW - NOW-<n>DAY(S) (or HOUR(S), MONTH(S), etc.) - NOW+<n>DAY(S) - yyyy-MM-ddThh:mm:ssZ-<n>DAY(S) - NOW/DAY (or HOUR, MONTH etc.) - rounds the value to the given unit Alternatively, an already fully formatted string such as "[NOW-1DAY TO NOW]" can be used as well. raw : str, optional Additional query text that will be appended to the query. area_relation : {'Intersects', 'Contains', 'IsWithin'}, optional What relation to use for testing the AOI. Case insensitive. - Intersects: true if the AOI and the footprint intersect (default) - Contains: true if the AOI is inside the footprint - IsWithin: true if the footprint is inside the AOI order_by: str, optional A comma-separated list of fields to order by (on server side). Prefix the field name by '+' or '-' to sort in ascending or descending order, respectively. Ascending order is used if prefix is omitted. Example: "cloudcoverpercentage, -beginposition". limit: int, optional Maximum number of products returned. Defaults to no limit. offset: int, optional The number of results to skip. Defaults to 0. **keywords Additional keywords can be used to specify other query parameters, e.g. `relativeorbitnumber=70`. See https://scihub.copernicus.eu/twiki/do/view/SciHubUserGuide/3FullTextSearch for a full list. Range values can be passed as two-element tuples, e.g. `cloudcoverpercentage=(0, 30)`. `None` can be used in range values for one-sided ranges, e.g. `orbitnumber=(16302, None)`. Ranges with no bounds (`orbitnumber=(None, None)`) will not be included in the query. The time interval formats accepted by the `date` parameter can also be used with any other parameters that expect time intervals (that is: 'beginposition', 'endposition', 'date', 'creationdate', and 'ingestiondate'). Returns ------- dict[string, dict] Products returned by the query as a dictionary with the product ID as the key and the product's attributes (a dictionary) as the value. """ query = self.format_query(area, date, raw, area_relation, **keywords) self.logger.debug("Running query: order_by=%s, limit=%s, offset=%s, query=%s", order_by, limit, offset, query) formatted_order_by = _format_order_by(order_by) response, count = self._load_query(query, formatted_order_by, limit, offset) self.logger.info("Found %s products", count) return _parse_opensearch_response(response) @staticmethod def format_query(area=None, date=None, raw=None, area_relation='Intersects', **keywords): """Create a OpenSearch API query string. """ if area_relation.lower() not in {"intersects", "contains", "iswithin"}: raise ValueError("Incorrect AOI relation provided ({})".format(area_relation)) # Check for duplicate keywords kw_lower = set(x.lower() for x in keywords) if (len(kw_lower) != len(keywords) or (date is not None and 'beginposition' in kw_lower) or (area is not None and 'footprint' in kw_lower)): raise ValueError("Query contains duplicate keywords. Note that query keywords are case-insensitive.") query_parts = [] if date is not None: keywords['beginPosition'] = date for attr, value in sorted(keywords.items()): # Escape spaces, where appropriate if isinstance(value, string_types): value = value.strip() if not any(value.startswith(s[0]) and value.endswith(s[1]) for s in ['[]', '{}', '//', '()']): value = re.sub(r'\s', r'\ ', value, re.M) # Handle date keywords # Keywords from https://github.com/SentinelDataHub/DataHubSystem/search?q=text/date+iso8601 date_attrs = ['beginposition', 'endposition', 'date', 'creationdate', 'ingestiondate'] if attr.lower() in date_attrs: # Automatically format date-type attributes if isinstance(value, string_types) and ' TO ' in value: # This is a string already formatted as a date interval, # e.g. '[NOW-1DAY TO NOW]' pass elif not isinstance(value, string_types) and len(value) == 2: value = (format_query_date(value[0]), format_query_date(value[1])) else: raise ValueError("Date-type query parameter '{}' expects a two-element tuple " "of str or datetime objects. Received {}".format(attr, value)) # Handle ranged values if isinstance(value, (list, tuple)): # Handle value ranges if len(value) == 2: # Allow None to be used as a unlimited bound value = ['*' if x is None else x for x in value] if all(x == '*' for x in value): continue value = '[{} TO {}]'.format(*value) else: raise ValueError("Invalid number of elements in list. Expected 2, received " "{}".format(len(value))) query_parts.append('{}:{}'.format(attr, value)) if raw: query_parts.append(raw) if area is not None: query_parts.append('footprint:"{}({})"'.format(area_relation, area)) return ' '.join(query_parts) def query_raw(self, query, order_by=None, limit=None, offset=0): """ Do a full-text query on the OpenSearch API using the format specified in https://scihub.copernicus.eu/twiki/do/view/SciHubUserGuide/3FullTextSearch DEPRECATED: use :meth:`query(raw=...) <.query>` instead. This method will be removed in the next major release. Parameters ---------- query : str The query string. order_by: str, optional A comma-separated list of fields to order by (on server side). Prefix the field name by '+' or '-' to sort in ascending or descending order, respectively. Ascending order is used, if prefix is omitted. Example: "cloudcoverpercentage, -beginposition". limit: int, optional Maximum number of products returned. Defaults to no limit. offset: int, optional The number of results to skip. Defaults to 0. Returns ------- dict[string, dict] Products returned by the query as a dictionary with the product ID as the key and the product's attributes (a dictionary) as the value. """ warnings.warn( "query_raw() has been merged with query(). use query(raw=...) instead.", PendingDeprecationWarning ) return self.query(raw=query, order_by=order_by, limit=limit, offset=offset) def count(self, area=None, date=None, raw=None, area_relation='Intersects', **keywords): """Get the number of products matching a query. Accepted parameters are identical to :meth:`SentinelAPI.query()`. This is a significantly more efficient alternative to doing `len(api.query())`, which can take minutes to run for queries matching thousands of products. Returns ------- int The number of products matching a query. """ for kw in ['order_by', 'limit', 'offset']: # Allow these function arguments to be included for compatibility with query(), # but ignore them. if kw in keywords: del keywords[kw] query = self.format_query(area, date, raw, area_relation, **keywords) _, total_count = self._load_query(query, limit=0) return total_count def _load_query(self, query, order_by=None, limit=None, offset=0): products, count = self._load_subquery(query, order_by, limit, offset) # repeat query until all results have been loaded max_offset = count if limit is not None: max_offset = min(count, offset + limit) if max_offset > offset + self.page_size: progress = self._tqdm(desc="Querying products", initial=self.page_size, total=max_offset - offset, unit=' products') for new_offset in range(offset + self.page_size, max_offset, self.page_size): new_limit = limit if limit is not None: new_limit = limit - new_offset + offset ret = self._load_subquery(query, order_by, new_limit, new_offset)[0] progress.update(len(ret)) products += ret progress.close() return products, count def _load_subquery(self, query, order_by=None, limit=None, offset=0): # store last query (for testing) self._last_query = query self.logger.debug("Sub-query: offset=%s, limit=%s", offset, limit) # load query results url = self._format_url(order_by, limit, offset) response = self.session.post(url, {'q': query}, auth=self.session.auth, headers={'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8'}, timeout=self.timeout) _check_scihub_response(response) # store last status code (for testing) self._last_response = response # parse response content try: json_feed = response.json()['feed'] if json_feed['opensearch:totalResults'] is None: # We are using some unintended behavior of the server that a null is # returned as the total results value when the query string was incorrect. raise SentinelAPIError( 'Invalid query string. Check the parameters and format.', response) total_results = int(json_feed['opensearch:totalResults']) except (ValueError, KeyError): raise SentinelAPIError('API response not valid. JSON decoding failed.', response) products = json_feed.get('entry', []) # this verification is necessary because if the query returns only # one product, self.products will be a dict not a list if isinstance(products, dict): products = [products] return products, total_results def _format_url(self, order_by=None, limit=None, offset=0): if limit is None: limit = self.page_size limit = min(limit, self.page_size) url = 'search?format=json&rows={}'.format(limit) url += '&start={}'.format(offset) if order_by: url += '&orderby={}'.format(order_by) return urljoin(self.api_url, url) @staticmethod def to_geojson(products): """Return the products from a query response as a GeoJSON with the values in their appropriate Python types. """ feature_list = [] for i, (product_id, props) in enumerate(products.items()): props = props.copy() props['id'] = product_id poly = geomet.wkt.loads(props['footprint']) del props['footprint'] del props['gmlfootprint'] # Fix "'datetime' is not JSON serializable" for k, v in props.items(): if isinstance(v, (date, datetime)): props[k] = v.strftime('%Y-%m-%dT%H:%M:%S.%fZ') feature_list.append( geojson.Feature(geometry=poly, id=i, properties=props) ) return geojson.FeatureCollection(feature_list) @staticmethod def to_dataframe(products): """Return the products from a query response as a Pandas DataFrame with the values in their appropriate Python types. """ try: import pandas as pd except ImportError: raise ImportError("to_dataframe requires the optional dependency Pandas.") return pd.DataFrame.from_dict(products, orient='index') @staticmethod def to_geodataframe(products): """Return the products from a query response as a GeoPandas GeoDataFrame with the values in their appropriate Python types. """ try: import geopandas as gpd import shapely.wkt except ImportError: raise ImportError("to_geodataframe requires the optional dependencies GeoPandas and Shapely.") crs = {'init': 'epsg:4326'} # WGS84 if len(products) == 0: return gpd.GeoDataFrame(crs=crs) df = SentinelAPI.to_dataframe(products) geometry = [shapely.wkt.loads(fp) for fp in df['footprint']] # remove useless columns df.drop(['footprint', 'gmlfootprint'], axis=1, inplace=True) return gpd.GeoDataFrame(df, crs=crs, geometry=geometry) def get_product_odata(self, id, full=False): """Access OData API to get info about a product. Returns a dict containing the id, title, size, md5sum, date, footprint and download url of the product. The date field corresponds to the Start ContentDate value. If `full` is set to True, then the full, detailed metadata of the product is returned in addition to the above. Parameters ---------- id : string The UUID of the product to query full : bool Whether to get the full metadata for the Product. False by default. Returns ------- dict[str, Any] A dictionary with an item for each metadata attribute Notes ----- For a full list of mappings between the OpenSearch (Solr) and OData attribute names see the following definition files: https://github.com/SentinelDataHub/DataHubSystem/blob/master/addon/sentinel-1/src/main/resources/META-INF/sentinel-1.owl https://github.com/SentinelDataHub/DataHubSystem/blob/master/addon/sentinel-2/src/main/resources/META-INF/sentinel-2.owl https://github.com/SentinelDataHub/DataHubSystem/blob/master/addon/sentinel-3/src/main/resources/META-INF/sentinel-3.owl """ url = urljoin(self.api_url, u"odata/v1/Products('{}')?$format=json".format(id)) if full: url += '&$expand=Attributes' response = self.session.get(url, auth=self.session.auth, timeout=self.timeout) _check_scihub_response(response) values = _parse_odata_response(response.json()['d']) return values def _trigger_offline_retrieval(self, url): """ Triggers retrieval of an offline product Trying to download an offline product triggers its retrieval from the long term archive. The returned HTTP status code conveys whether this was successful. Parameters ---------- url : string URL for downloading the product Notes ----- https://scihub.copernicus.eu/userguide/LongTermArchive """ with self.session.get(url, auth=self.session.auth, timeout=self.timeout) as r: # check https://scihub.copernicus.eu/userguide/LongTermArchive#HTTP_Status_codes if r.status_code == 202: self.logger.info("Accepted for retrieval") elif r.status_code == 503: self.logger.error("Request not accepted") raise SentinelAPILTAError('Request for retrieval from LTA not accepted', r) elif r.status_code == 403: self.logger.error("Requests exceed user quota") raise SentinelAPILTAError('Requests for retrieval from LTA exceed user quota', r) elif r.status_code == 500: # should not happen self.logger.error("Trying to download an offline product") raise SentinelAPILTAError('Trying to download an offline product', r) return r.status_code def download(self, id, directory_path='.', checksum=True): """Download a product. Uses the filename on the server for the downloaded file, e.g. "S1A_EW_GRDH_1SDH_20141003T003840_20141003T003920_002658_002F54_4DD1.zip". Incomplete downloads are continued and complete files are skipped. Parameters ---------- id : string UUID of the product, e.g. 'a8dd0cfd-613e-45ce-868c-d79177b916ed' directory_path : string, optional Where the file will be downloaded checksum : bool, optional If True, verify the downloaded file's integrity by checking its MD5 checksum. Throws InvalidChecksumError if the checksum does not match. Defaults to True. Returns ------- product_info : dict Dictionary containing the product's info from get_product_info() as well as the path on disk. Raises ------ InvalidChecksumError If the MD5 checksum does not match the checksum on the server. """ product_info = self.get_product_odata(id) path = join(directory_path, product_info['title'] + '.zip') product_info['path'] = path product_info['downloaded_bytes'] = 0 self.logger.info('Downloading %s to %s', id, path) if exists(path): # We assume that the product has been downloaded and is complete return product_info # An incomplete download triggers the retrieval from the LTA if the product is not online if not product_info['Online']: self.logger.warning( 'Product %s is not online. Triggering retrieval from long term archive.', product_info['id']) self._trigger_offline_retrieval(product_info['url']) return product_info # Use a temporary file for downloading temp_path = path + '.incomplete' skip_download = False if exists(temp_path): if getsize(temp_path) > product_info['size']: self.logger.warning( "Existing incomplete file %s is larger than the expected final size" " (%s vs %s bytes). Deleting it.", str(temp_path), getsize(temp_path), product_info['size']) remove(temp_path) elif getsize(temp_path) == product_info['size']: if self._md5_compare(temp_path, product_info['md5']): skip_download = True else: # Log a warning since this should never happen self.logger.warning( "Existing incomplete file %s appears to be fully downloaded but " "its checksum is incorrect. Deleting it.", str(temp_path)) remove(temp_path) else: # continue downloading self.logger.info( "Download will resume from existing incomplete file %s.", temp_path) pass if not skip_download: # Store the number of downloaded bytes for unit tests product_info['downloaded_bytes'] = self._download( product_info['url'], temp_path, self.session, product_info['size']) # Check integrity with MD5 checksum if checksum is True: if not self._md5_compare(temp_path, product_info['md5']): remove(temp_path) raise InvalidChecksumError('File corrupt: checksums do not match') # Download successful, rename the temporary file to its proper name shutil.move(temp_path, path) return product_info def download_all(self, products, directory_path='.', max_attempts=10, checksum=True): """Download a list of products. Takes a list of product IDs as input. This means that the return value of query() can be passed directly to this method. File names on the server are used for the downloaded files, e.g. "S1A_EW_GRDH_1SDH_20141003T003840_20141003T003920_002658_002F54_4DD1.zip". In case of interruptions or other exceptions, downloading will restart from where it left off. Downloading is attempted at most max_attempts times to avoid getting stuck with unrecoverable errors. Parameters ---------- products : list List of product IDs directory_path : string Directory where the downloaded files will be downloaded max_attempts : int, optional Number of allowed retries before giving up downloading a product. Defaults to 10. checksum : bool, optional If True, verify the downloaded files' integrity by checking its MD5 checksum. Throws InvalidChecksumError if the checksum does not match. Defaults to True. Raises ------ Raises the most recent downloading exception if all downloads failed. Returns ------- dict[string, dict] A dictionary containing the return value from download() for each successfully downloaded product. dict[string, dict] A dictionary containing the product information for products whose retrieval from the long term archive was successfully triggered. set[string] The list of products that failed to download. """ product_ids = list(products) self.logger.info("Will download %d products", len(product_ids)) return_values = OrderedDict() last_exception = None for i, product_id in enumerate(products): for attempt_num in range(max_attempts): try: product_info = self.download(product_id, directory_path, checksum) return_values[product_id] = product_info break except (KeyboardInterrupt, SystemExit): raise except InvalidChecksumError as e: last_exception = e self.logger.warning( "Invalid checksum. The downloaded file for '%s' is corrupted.", product_id) except SentinelAPILTAError as e: last_exception = e self.logger.exception("There was an error retrieving %s from the LTA", product_id) break except Exception as e: last_exception = e self.logger.exception("There was an error downloading %s", product_id) self.logger.info("%s/%s products downloaded", i + 1, len(product_ids)) failed = set(products) - set(return_values) # split up sucessfully processed products into downloaded and only triggered retrieval from the LTA triggered = OrderedDict([(k, v) for k, v in return_values.items() if v['Online'] is False]) downloaded = OrderedDict([(k, v) for k, v in return_values.items() if v['Online'] is True]) if len(failed) == len(product_ids) and last_exception is not None: raise last_exception return downloaded, triggered, failed @staticmethod def get_products_size(products): """Return the total file size in GB of all products in the OpenSearch response.""" size_total = 0 for title, props in products.items(): size_product = props["size"] size_value = float(size_product.split(" ")[0]) size_unit = str(size_product.split(" ")[1]) if size_unit == "MB": size_value /= 1024. if size_unit == "KB": size_value /= 1024. * 1024. size_total += size_value return round(size_total, 2) @staticmethod def check_query_length(query): """Determine whether a query to the OpenSearch API is too long. The length of a query string is limited to approximately 3938 characters but any special characters (that is, not alphanumeric or -_.*) will take up more space. Parameters ---------- query : str The query string Returns ------- float Ratio of the query length to the maximum length """ # The server uses the Java's URLEncoder implementation internally, which we are replicating here effective_length = len(quote_plus(query, safe="-_.*").replace('~', '%7E')) return effective_length / 3938 def _query_names(self, names): """Find products by their names, e.g. S1A_EW_GRDH_1SDH_20141003T003840_20141003T003920_002658_002F54_4DD1. Note that duplicates exist on server, so multiple products can be returned for each name. Parameters ---------- names : list[string] List of product names. Returns ------- dict[string, dict[str, dict]] A dictionary mapping each name to a dictionary which contains the products with that name (with ID as the key). """ def chunks(l, n): """Yield successive n-sized chunks from l.""" for i in range(0, len(l), n): yield l[i:i + n] products = {} # 40 names per query fits reasonably well inside the query limit for chunk in chunks(names, 40): query = " OR ".join(chunk) products.update(self.query(raw=query)) # Group the products output = OrderedDict((name, dict()) for name in names) for id, metadata in products.items(): name = metadata['identifier'] output[name][id] = metadata return output def _md5_compare(self, file_path, checksum, block_size=2 ** 13): """Compare a given MD5 checksum with one calculated from a file.""" with closing(self._tqdm(desc="MD5 checksumming", total=getsize(file_path), unit="B", unit_scale=True)) as progress: md5 = hashlib.md5() with open(file_path, "rb") as f: while True: block_data = f.read(block_size) if not block_data: break md5.update(block_data) progress.update(len(block_data)) return md5.hexdigest().lower() == checksum.lower() def _download(self, url, path, session, file_size): headers = {} continuing = exists(path) if continuing: already_downloaded_bytes = getsize(path) headers = {'Range': 'bytes={}-'.format(already_downloaded_bytes)} else: already_downloaded_bytes = 0 downloaded_bytes = 0 with closing(session.get(url, stream=True, auth=session.auth, headers=headers, timeout=self.timeout)) as r, \ closing(self._tqdm(desc="Downloading", total=file_size, unit="B", unit_scale=True, initial=already_downloaded_bytes)) as progress: _check_scihub_response(r, test_json=False) chunk_size = 2 ** 20 # download in 1 MB chunks mode = 'ab' if continuing else 'wb' with open(path, mode) as f: for chunk in r.iter_content(chunk_size=chunk_size): if chunk: # filter out keep-alive new chunks f.write(chunk) progress.update(len(chunk)) downloaded_bytes += len(chunk) # Return the number of bytes downloaded return downloaded_bytes def _tqdm(self, **kwargs): """tqdm progressbar wrapper. May be overridden to customize progressbar behavior""" kwargs.update({'disable': not self.show_progressbars}) return tqdm(**kwargs)
sentinelsat/sentinelsat
sentinelsat/sentinel.py
SentinelAPI._md5_compare
python
def _md5_compare(self, file_path, checksum, block_size=2 ** 13): with closing(self._tqdm(desc="MD5 checksumming", total=getsize(file_path), unit="B", unit_scale=True)) as progress: md5 = hashlib.md5() with open(file_path, "rb") as f: while True: block_data = f.read(block_size) if not block_data: break md5.update(block_data) progress.update(len(block_data)) return md5.hexdigest().lower() == checksum.lower()
Compare a given MD5 checksum with one calculated from a file.
train
https://github.com/sentinelsat/sentinelsat/blob/eacfd79ff4e7e939147db9dfdd393c67d64eecaa/sentinelsat/sentinel.py#L795-L807
[ "def _tqdm(self, **kwargs):\n \"\"\"tqdm progressbar wrapper. May be overridden to customize progressbar behavior\"\"\"\n kwargs.update({'disable': not self.show_progressbars})\n return tqdm(**kwargs)\n" ]
class SentinelAPI: """Class to connect to Copernicus Open Access Hub, search and download imagery. Parameters ---------- user : string username for DataHub set to None to use ~/.netrc password : string password for DataHub set to None to use ~/.netrc api_url : string, optional URL of the DataHub defaults to 'https://scihub.copernicus.eu/apihub' show_progressbars : bool Whether progressbars should be shown or not, e.g. during download. Defaults to True. timeout : float or tuple, optional How long to wait for DataHub response (in seconds). Tuple (connect, read) allowed. Attributes ---------- session : requests.Session Session to connect to DataHub api_url : str URL to the DataHub page_size : int Number of results per query page. Current value: 100 (maximum allowed on ApiHub) timeout : float or tuple How long to wait for DataHub response (in seconds). """ logger = logging.getLogger('sentinelsat.SentinelAPI') def __init__(self, user, password, api_url='https://scihub.copernicus.eu/apihub/', show_progressbars=True, timeout=None): self.session = requests.Session() if user and password: self.session.auth = (user, password) self.api_url = api_url if api_url.endswith('/') else api_url + '/' self.page_size = 100 self.user_agent = 'sentinelsat/' + sentinelsat_version self.session.headers['User-Agent'] = self.user_agent self.show_progressbars = show_progressbars self.timeout = timeout # For unit tests self._last_query = None self._last_response = None def query(self, area=None, date=None, raw=None, area_relation='Intersects', order_by=None, limit=None, offset=0, **keywords): """Query the OpenSearch API with the coordinates of an area, a date interval and any other search keywords accepted by the API. Parameters ---------- area : str, optional The area of interest formatted as a Well-Known Text string. date : tuple of (str or datetime) or str, optional A time interval filter based on the Sensing Start Time of the products. Expects a tuple of (start, end), e.g. ("NOW-1DAY", "NOW"). The timestamps can be either a Python datetime or a string in one of the following formats: - yyyyMMdd - yyyy-MM-ddThh:mm:ss.SSSZ (ISO-8601) - yyyy-MM-ddThh:mm:ssZ - NOW - NOW-<n>DAY(S) (or HOUR(S), MONTH(S), etc.) - NOW+<n>DAY(S) - yyyy-MM-ddThh:mm:ssZ-<n>DAY(S) - NOW/DAY (or HOUR, MONTH etc.) - rounds the value to the given unit Alternatively, an already fully formatted string such as "[NOW-1DAY TO NOW]" can be used as well. raw : str, optional Additional query text that will be appended to the query. area_relation : {'Intersects', 'Contains', 'IsWithin'}, optional What relation to use for testing the AOI. Case insensitive. - Intersects: true if the AOI and the footprint intersect (default) - Contains: true if the AOI is inside the footprint - IsWithin: true if the footprint is inside the AOI order_by: str, optional A comma-separated list of fields to order by (on server side). Prefix the field name by '+' or '-' to sort in ascending or descending order, respectively. Ascending order is used if prefix is omitted. Example: "cloudcoverpercentage, -beginposition". limit: int, optional Maximum number of products returned. Defaults to no limit. offset: int, optional The number of results to skip. Defaults to 0. **keywords Additional keywords can be used to specify other query parameters, e.g. `relativeorbitnumber=70`. See https://scihub.copernicus.eu/twiki/do/view/SciHubUserGuide/3FullTextSearch for a full list. Range values can be passed as two-element tuples, e.g. `cloudcoverpercentage=(0, 30)`. `None` can be used in range values for one-sided ranges, e.g. `orbitnumber=(16302, None)`. Ranges with no bounds (`orbitnumber=(None, None)`) will not be included in the query. The time interval formats accepted by the `date` parameter can also be used with any other parameters that expect time intervals (that is: 'beginposition', 'endposition', 'date', 'creationdate', and 'ingestiondate'). Returns ------- dict[string, dict] Products returned by the query as a dictionary with the product ID as the key and the product's attributes (a dictionary) as the value. """ query = self.format_query(area, date, raw, area_relation, **keywords) self.logger.debug("Running query: order_by=%s, limit=%s, offset=%s, query=%s", order_by, limit, offset, query) formatted_order_by = _format_order_by(order_by) response, count = self._load_query(query, formatted_order_by, limit, offset) self.logger.info("Found %s products", count) return _parse_opensearch_response(response) @staticmethod def format_query(area=None, date=None, raw=None, area_relation='Intersects', **keywords): """Create a OpenSearch API query string. """ if area_relation.lower() not in {"intersects", "contains", "iswithin"}: raise ValueError("Incorrect AOI relation provided ({})".format(area_relation)) # Check for duplicate keywords kw_lower = set(x.lower() for x in keywords) if (len(kw_lower) != len(keywords) or (date is not None and 'beginposition' in kw_lower) or (area is not None and 'footprint' in kw_lower)): raise ValueError("Query contains duplicate keywords. Note that query keywords are case-insensitive.") query_parts = [] if date is not None: keywords['beginPosition'] = date for attr, value in sorted(keywords.items()): # Escape spaces, where appropriate if isinstance(value, string_types): value = value.strip() if not any(value.startswith(s[0]) and value.endswith(s[1]) for s in ['[]', '{}', '//', '()']): value = re.sub(r'\s', r'\ ', value, re.M) # Handle date keywords # Keywords from https://github.com/SentinelDataHub/DataHubSystem/search?q=text/date+iso8601 date_attrs = ['beginposition', 'endposition', 'date', 'creationdate', 'ingestiondate'] if attr.lower() in date_attrs: # Automatically format date-type attributes if isinstance(value, string_types) and ' TO ' in value: # This is a string already formatted as a date interval, # e.g. '[NOW-1DAY TO NOW]' pass elif not isinstance(value, string_types) and len(value) == 2: value = (format_query_date(value[0]), format_query_date(value[1])) else: raise ValueError("Date-type query parameter '{}' expects a two-element tuple " "of str or datetime objects. Received {}".format(attr, value)) # Handle ranged values if isinstance(value, (list, tuple)): # Handle value ranges if len(value) == 2: # Allow None to be used as a unlimited bound value = ['*' if x is None else x for x in value] if all(x == '*' for x in value): continue value = '[{} TO {}]'.format(*value) else: raise ValueError("Invalid number of elements in list. Expected 2, received " "{}".format(len(value))) query_parts.append('{}:{}'.format(attr, value)) if raw: query_parts.append(raw) if area is not None: query_parts.append('footprint:"{}({})"'.format(area_relation, area)) return ' '.join(query_parts) def query_raw(self, query, order_by=None, limit=None, offset=0): """ Do a full-text query on the OpenSearch API using the format specified in https://scihub.copernicus.eu/twiki/do/view/SciHubUserGuide/3FullTextSearch DEPRECATED: use :meth:`query(raw=...) <.query>` instead. This method will be removed in the next major release. Parameters ---------- query : str The query string. order_by: str, optional A comma-separated list of fields to order by (on server side). Prefix the field name by '+' or '-' to sort in ascending or descending order, respectively. Ascending order is used, if prefix is omitted. Example: "cloudcoverpercentage, -beginposition". limit: int, optional Maximum number of products returned. Defaults to no limit. offset: int, optional The number of results to skip. Defaults to 0. Returns ------- dict[string, dict] Products returned by the query as a dictionary with the product ID as the key and the product's attributes (a dictionary) as the value. """ warnings.warn( "query_raw() has been merged with query(). use query(raw=...) instead.", PendingDeprecationWarning ) return self.query(raw=query, order_by=order_by, limit=limit, offset=offset) def count(self, area=None, date=None, raw=None, area_relation='Intersects', **keywords): """Get the number of products matching a query. Accepted parameters are identical to :meth:`SentinelAPI.query()`. This is a significantly more efficient alternative to doing `len(api.query())`, which can take minutes to run for queries matching thousands of products. Returns ------- int The number of products matching a query. """ for kw in ['order_by', 'limit', 'offset']: # Allow these function arguments to be included for compatibility with query(), # but ignore them. if kw in keywords: del keywords[kw] query = self.format_query(area, date, raw, area_relation, **keywords) _, total_count = self._load_query(query, limit=0) return total_count def _load_query(self, query, order_by=None, limit=None, offset=0): products, count = self._load_subquery(query, order_by, limit, offset) # repeat query until all results have been loaded max_offset = count if limit is not None: max_offset = min(count, offset + limit) if max_offset > offset + self.page_size: progress = self._tqdm(desc="Querying products", initial=self.page_size, total=max_offset - offset, unit=' products') for new_offset in range(offset + self.page_size, max_offset, self.page_size): new_limit = limit if limit is not None: new_limit = limit - new_offset + offset ret = self._load_subquery(query, order_by, new_limit, new_offset)[0] progress.update(len(ret)) products += ret progress.close() return products, count def _load_subquery(self, query, order_by=None, limit=None, offset=0): # store last query (for testing) self._last_query = query self.logger.debug("Sub-query: offset=%s, limit=%s", offset, limit) # load query results url = self._format_url(order_by, limit, offset) response = self.session.post(url, {'q': query}, auth=self.session.auth, headers={'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8'}, timeout=self.timeout) _check_scihub_response(response) # store last status code (for testing) self._last_response = response # parse response content try: json_feed = response.json()['feed'] if json_feed['opensearch:totalResults'] is None: # We are using some unintended behavior of the server that a null is # returned as the total results value when the query string was incorrect. raise SentinelAPIError( 'Invalid query string. Check the parameters and format.', response) total_results = int(json_feed['opensearch:totalResults']) except (ValueError, KeyError): raise SentinelAPIError('API response not valid. JSON decoding failed.', response) products = json_feed.get('entry', []) # this verification is necessary because if the query returns only # one product, self.products will be a dict not a list if isinstance(products, dict): products = [products] return products, total_results def _format_url(self, order_by=None, limit=None, offset=0): if limit is None: limit = self.page_size limit = min(limit, self.page_size) url = 'search?format=json&rows={}'.format(limit) url += '&start={}'.format(offset) if order_by: url += '&orderby={}'.format(order_by) return urljoin(self.api_url, url) @staticmethod def to_geojson(products): """Return the products from a query response as a GeoJSON with the values in their appropriate Python types. """ feature_list = [] for i, (product_id, props) in enumerate(products.items()): props = props.copy() props['id'] = product_id poly = geomet.wkt.loads(props['footprint']) del props['footprint'] del props['gmlfootprint'] # Fix "'datetime' is not JSON serializable" for k, v in props.items(): if isinstance(v, (date, datetime)): props[k] = v.strftime('%Y-%m-%dT%H:%M:%S.%fZ') feature_list.append( geojson.Feature(geometry=poly, id=i, properties=props) ) return geojson.FeatureCollection(feature_list) @staticmethod def to_dataframe(products): """Return the products from a query response as a Pandas DataFrame with the values in their appropriate Python types. """ try: import pandas as pd except ImportError: raise ImportError("to_dataframe requires the optional dependency Pandas.") return pd.DataFrame.from_dict(products, orient='index') @staticmethod def to_geodataframe(products): """Return the products from a query response as a GeoPandas GeoDataFrame with the values in their appropriate Python types. """ try: import geopandas as gpd import shapely.wkt except ImportError: raise ImportError("to_geodataframe requires the optional dependencies GeoPandas and Shapely.") crs = {'init': 'epsg:4326'} # WGS84 if len(products) == 0: return gpd.GeoDataFrame(crs=crs) df = SentinelAPI.to_dataframe(products) geometry = [shapely.wkt.loads(fp) for fp in df['footprint']] # remove useless columns df.drop(['footprint', 'gmlfootprint'], axis=1, inplace=True) return gpd.GeoDataFrame(df, crs=crs, geometry=geometry) def get_product_odata(self, id, full=False): """Access OData API to get info about a product. Returns a dict containing the id, title, size, md5sum, date, footprint and download url of the product. The date field corresponds to the Start ContentDate value. If `full` is set to True, then the full, detailed metadata of the product is returned in addition to the above. Parameters ---------- id : string The UUID of the product to query full : bool Whether to get the full metadata for the Product. False by default. Returns ------- dict[str, Any] A dictionary with an item for each metadata attribute Notes ----- For a full list of mappings between the OpenSearch (Solr) and OData attribute names see the following definition files: https://github.com/SentinelDataHub/DataHubSystem/blob/master/addon/sentinel-1/src/main/resources/META-INF/sentinel-1.owl https://github.com/SentinelDataHub/DataHubSystem/blob/master/addon/sentinel-2/src/main/resources/META-INF/sentinel-2.owl https://github.com/SentinelDataHub/DataHubSystem/blob/master/addon/sentinel-3/src/main/resources/META-INF/sentinel-3.owl """ url = urljoin(self.api_url, u"odata/v1/Products('{}')?$format=json".format(id)) if full: url += '&$expand=Attributes' response = self.session.get(url, auth=self.session.auth, timeout=self.timeout) _check_scihub_response(response) values = _parse_odata_response(response.json()['d']) return values def _trigger_offline_retrieval(self, url): """ Triggers retrieval of an offline product Trying to download an offline product triggers its retrieval from the long term archive. The returned HTTP status code conveys whether this was successful. Parameters ---------- url : string URL for downloading the product Notes ----- https://scihub.copernicus.eu/userguide/LongTermArchive """ with self.session.get(url, auth=self.session.auth, timeout=self.timeout) as r: # check https://scihub.copernicus.eu/userguide/LongTermArchive#HTTP_Status_codes if r.status_code == 202: self.logger.info("Accepted for retrieval") elif r.status_code == 503: self.logger.error("Request not accepted") raise SentinelAPILTAError('Request for retrieval from LTA not accepted', r) elif r.status_code == 403: self.logger.error("Requests exceed user quota") raise SentinelAPILTAError('Requests for retrieval from LTA exceed user quota', r) elif r.status_code == 500: # should not happen self.logger.error("Trying to download an offline product") raise SentinelAPILTAError('Trying to download an offline product', r) return r.status_code def download(self, id, directory_path='.', checksum=True): """Download a product. Uses the filename on the server for the downloaded file, e.g. "S1A_EW_GRDH_1SDH_20141003T003840_20141003T003920_002658_002F54_4DD1.zip". Incomplete downloads are continued and complete files are skipped. Parameters ---------- id : string UUID of the product, e.g. 'a8dd0cfd-613e-45ce-868c-d79177b916ed' directory_path : string, optional Where the file will be downloaded checksum : bool, optional If True, verify the downloaded file's integrity by checking its MD5 checksum. Throws InvalidChecksumError if the checksum does not match. Defaults to True. Returns ------- product_info : dict Dictionary containing the product's info from get_product_info() as well as the path on disk. Raises ------ InvalidChecksumError If the MD5 checksum does not match the checksum on the server. """ product_info = self.get_product_odata(id) path = join(directory_path, product_info['title'] + '.zip') product_info['path'] = path product_info['downloaded_bytes'] = 0 self.logger.info('Downloading %s to %s', id, path) if exists(path): # We assume that the product has been downloaded and is complete return product_info # An incomplete download triggers the retrieval from the LTA if the product is not online if not product_info['Online']: self.logger.warning( 'Product %s is not online. Triggering retrieval from long term archive.', product_info['id']) self._trigger_offline_retrieval(product_info['url']) return product_info # Use a temporary file for downloading temp_path = path + '.incomplete' skip_download = False if exists(temp_path): if getsize(temp_path) > product_info['size']: self.logger.warning( "Existing incomplete file %s is larger than the expected final size" " (%s vs %s bytes). Deleting it.", str(temp_path), getsize(temp_path), product_info['size']) remove(temp_path) elif getsize(temp_path) == product_info['size']: if self._md5_compare(temp_path, product_info['md5']): skip_download = True else: # Log a warning since this should never happen self.logger.warning( "Existing incomplete file %s appears to be fully downloaded but " "its checksum is incorrect. Deleting it.", str(temp_path)) remove(temp_path) else: # continue downloading self.logger.info( "Download will resume from existing incomplete file %s.", temp_path) pass if not skip_download: # Store the number of downloaded bytes for unit tests product_info['downloaded_bytes'] = self._download( product_info['url'], temp_path, self.session, product_info['size']) # Check integrity with MD5 checksum if checksum is True: if not self._md5_compare(temp_path, product_info['md5']): remove(temp_path) raise InvalidChecksumError('File corrupt: checksums do not match') # Download successful, rename the temporary file to its proper name shutil.move(temp_path, path) return product_info def download_all(self, products, directory_path='.', max_attempts=10, checksum=True): """Download a list of products. Takes a list of product IDs as input. This means that the return value of query() can be passed directly to this method. File names on the server are used for the downloaded files, e.g. "S1A_EW_GRDH_1SDH_20141003T003840_20141003T003920_002658_002F54_4DD1.zip". In case of interruptions or other exceptions, downloading will restart from where it left off. Downloading is attempted at most max_attempts times to avoid getting stuck with unrecoverable errors. Parameters ---------- products : list List of product IDs directory_path : string Directory where the downloaded files will be downloaded max_attempts : int, optional Number of allowed retries before giving up downloading a product. Defaults to 10. checksum : bool, optional If True, verify the downloaded files' integrity by checking its MD5 checksum. Throws InvalidChecksumError if the checksum does not match. Defaults to True. Raises ------ Raises the most recent downloading exception if all downloads failed. Returns ------- dict[string, dict] A dictionary containing the return value from download() for each successfully downloaded product. dict[string, dict] A dictionary containing the product information for products whose retrieval from the long term archive was successfully triggered. set[string] The list of products that failed to download. """ product_ids = list(products) self.logger.info("Will download %d products", len(product_ids)) return_values = OrderedDict() last_exception = None for i, product_id in enumerate(products): for attempt_num in range(max_attempts): try: product_info = self.download(product_id, directory_path, checksum) return_values[product_id] = product_info break except (KeyboardInterrupt, SystemExit): raise except InvalidChecksumError as e: last_exception = e self.logger.warning( "Invalid checksum. The downloaded file for '%s' is corrupted.", product_id) except SentinelAPILTAError as e: last_exception = e self.logger.exception("There was an error retrieving %s from the LTA", product_id) break except Exception as e: last_exception = e self.logger.exception("There was an error downloading %s", product_id) self.logger.info("%s/%s products downloaded", i + 1, len(product_ids)) failed = set(products) - set(return_values) # split up sucessfully processed products into downloaded and only triggered retrieval from the LTA triggered = OrderedDict([(k, v) for k, v in return_values.items() if v['Online'] is False]) downloaded = OrderedDict([(k, v) for k, v in return_values.items() if v['Online'] is True]) if len(failed) == len(product_ids) and last_exception is not None: raise last_exception return downloaded, triggered, failed @staticmethod def get_products_size(products): """Return the total file size in GB of all products in the OpenSearch response.""" size_total = 0 for title, props in products.items(): size_product = props["size"] size_value = float(size_product.split(" ")[0]) size_unit = str(size_product.split(" ")[1]) if size_unit == "MB": size_value /= 1024. if size_unit == "KB": size_value /= 1024. * 1024. size_total += size_value return round(size_total, 2) @staticmethod def check_query_length(query): """Determine whether a query to the OpenSearch API is too long. The length of a query string is limited to approximately 3938 characters but any special characters (that is, not alphanumeric or -_.*) will take up more space. Parameters ---------- query : str The query string Returns ------- float Ratio of the query length to the maximum length """ # The server uses the Java's URLEncoder implementation internally, which we are replicating here effective_length = len(quote_plus(query, safe="-_.*").replace('~', '%7E')) return effective_length / 3938 def _query_names(self, names): """Find products by their names, e.g. S1A_EW_GRDH_1SDH_20141003T003840_20141003T003920_002658_002F54_4DD1. Note that duplicates exist on server, so multiple products can be returned for each name. Parameters ---------- names : list[string] List of product names. Returns ------- dict[string, dict[str, dict]] A dictionary mapping each name to a dictionary which contains the products with that name (with ID as the key). """ def chunks(l, n): """Yield successive n-sized chunks from l.""" for i in range(0, len(l), n): yield l[i:i + n] products = {} # 40 names per query fits reasonably well inside the query limit for chunk in chunks(names, 40): query = " OR ".join(chunk) products.update(self.query(raw=query)) # Group the products output = OrderedDict((name, dict()) for name in names) for id, metadata in products.items(): name = metadata['identifier'] output[name][id] = metadata return output def check_files(self, paths=None, ids=None, directory=None, delete=False): """Verify the integrity of product files on disk. Integrity is checked by comparing the size and checksum of the file with the respective values on the server. The input can be a list of products to check or a list of IDs and a directory. In cases where multiple products with different IDs exist on the server for given product name, the file is considered to be correct if any of them matches the file size and checksum. A warning is logged in such situations. The corrupt products' OData info is included in the return value to make it easier to re-download the products, if necessary. Parameters ---------- paths : list[string] List of product file paths. ids : list[string] List of product IDs. directory : string Directory where the files are located, if checking based on product IDs. delete : bool Whether to delete corrupt products. Defaults to False. Returns ------- dict[str, list[dict]] A dictionary listing the invalid or missing files. The dictionary maps the corrupt file paths to a list of OData dictionaries of matching products on the server (as returned by :meth:`SentinelAPI.get_product_odata()`). """ if not ids and not paths: raise ValueError("Must provide either file paths or product IDs and a directory") if ids and not directory: raise ValueError("Directory value missing") paths = paths or [] ids = ids or [] def name_from_path(path): return splitext(basename(path))[0] # Get product IDs corresponding to the files on disk names = [] if paths: names = list(map(name_from_path, paths)) result = self._query_names(names) for product_dicts in result.values(): ids += list(product_dicts) names_from_paths = set(names) ids = set(ids) # Collect the OData information for each product # Product name -> list of matching odata dicts product_infos = defaultdict(list) for id in ids: odata = self.get_product_odata(id) name = odata['title'] product_infos[name].append(odata) # Collect if name not in names_from_paths: paths.append(join(directory, name + '.zip')) # Now go over the list of products and check them corrupt = {} for path in paths: name = name_from_path(path) if len(product_infos[name]) > 1: self.logger.warning("{} matches multiple products on server".format(path)) if not exists(path): # We will consider missing files as corrupt also self.logger.info("{} does not exist on disk".format(path)) corrupt[path] = product_infos[name] continue is_fine = False for product_info in product_infos[name]: if (getsize(path) == product_info['size'] and self._md5_compare(path, product_info['md5'])): is_fine = True break if not is_fine: self.logger.info("{} is corrupt".format(path)) corrupt[path] = product_infos[name] if delete: remove(path) return corrupt def _download(self, url, path, session, file_size): headers = {} continuing = exists(path) if continuing: already_downloaded_bytes = getsize(path) headers = {'Range': 'bytes={}-'.format(already_downloaded_bytes)} else: already_downloaded_bytes = 0 downloaded_bytes = 0 with closing(session.get(url, stream=True, auth=session.auth, headers=headers, timeout=self.timeout)) as r, \ closing(self._tqdm(desc="Downloading", total=file_size, unit="B", unit_scale=True, initial=already_downloaded_bytes)) as progress: _check_scihub_response(r, test_json=False) chunk_size = 2 ** 20 # download in 1 MB chunks mode = 'ab' if continuing else 'wb' with open(path, mode) as f: for chunk in r.iter_content(chunk_size=chunk_size): if chunk: # filter out keep-alive new chunks f.write(chunk) progress.update(len(chunk)) downloaded_bytes += len(chunk) # Return the number of bytes downloaded return downloaded_bytes def _tqdm(self, **kwargs): """tqdm progressbar wrapper. May be overridden to customize progressbar behavior""" kwargs.update({'disable': not self.show_progressbars}) return tqdm(**kwargs)
deanmalmgren/textract
textract/parsers/odt_parser.py
Parser.to_string
python
def to_string(self): buff = u"" for child in self.content.iter(): if child.tag in [self.qn('text:p'), self.qn('text:h')]: buff += self.text_to_string(child) + "\n" # remove last newline char if buff: buff = buff[:-1] return buff
Converts the document to a string.
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/odt_parser.py#L19-L28
[ "def qn(self, namespace):\n \"\"\"Connect tag prefix to longer namespace\"\"\"\n nsmap = {\n 'text': 'urn:oasis:names:tc:opendocument:xmlns:text:1.0',\n }\n spl = namespace.split(':')\n return '{{{}}}{}'.format(nsmap[spl[0]], spl[1])\n" ]
class Parser(BaseParser): """Extract text from open document files. """ def extract(self, filename, **kwargs): # Inspiration from # https://github.com/odoo/odoo/blob/master/addons/document/odt2txt.py with open(filename, 'rb') as stream: zip_stream = zipfile.ZipFile(stream) self.content = ET.fromstring(zip_stream.read("content.xml")) return self.to_string() def text_to_string(self, element): buff = u"" if element.text is not None: buff += element.text for child in element: if child.tag == self.qn('text:tab'): buff += "\t" if child.tail is not None: buff += child.tail elif child.tag == self.qn('text:s'): buff += u" " if child.get(self.qn('text:c')) is not None: buff += u" " * (int(child.get(self.qn('text:c'))) - 1) if child.tail is not None: buff += child.tail else: buff += self.text_to_string(child) if element.tail is not None: buff += element.tail return buff def qn(self, namespace): """Connect tag prefix to longer namespace""" nsmap = { 'text': 'urn:oasis:names:tc:opendocument:xmlns:text:1.0', } spl = namespace.split(':') return '{{{}}}{}'.format(nsmap[spl[0]], spl[1])
deanmalmgren/textract
textract/parsers/odt_parser.py
Parser.qn
python
def qn(self, namespace): nsmap = { 'text': 'urn:oasis:names:tc:opendocument:xmlns:text:1.0', } spl = namespace.split(':') return '{{{}}}{}'.format(nsmap[spl[0]], spl[1])
Connect tag prefix to longer namespace
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/odt_parser.py#L51-L57
null
class Parser(BaseParser): """Extract text from open document files. """ def extract(self, filename, **kwargs): # Inspiration from # https://github.com/odoo/odoo/blob/master/addons/document/odt2txt.py with open(filename, 'rb') as stream: zip_stream = zipfile.ZipFile(stream) self.content = ET.fromstring(zip_stream.read("content.xml")) return self.to_string() def to_string(self): """ Converts the document to a string. """ buff = u"" for child in self.content.iter(): if child.tag in [self.qn('text:p'), self.qn('text:h')]: buff += self.text_to_string(child) + "\n" # remove last newline char if buff: buff = buff[:-1] return buff def text_to_string(self, element): buff = u"" if element.text is not None: buff += element.text for child in element: if child.tag == self.qn('text:tab'): buff += "\t" if child.tail is not None: buff += child.tail elif child.tag == self.qn('text:s'): buff += u" " if child.get(self.qn('text:c')) is not None: buff += u" " * (int(child.get(self.qn('text:c'))) - 1) if child.tail is not None: buff += child.tail else: buff += self.text_to_string(child) if element.tail is not None: buff += element.tail return buff
deanmalmgren/textract
textract/cli.py
get_parser
python
def get_parser(): # initialize the parser parser = argparse.ArgumentParser( description=( 'Command line tool for extracting text from any document. ' ) % locals(), ) # define the command line options here parser.add_argument( 'filename', help='Filename to extract text.', ).completer = argcomplete.completers.FilesCompleter parser.add_argument( '-e', '--encoding', type=str, default=DEFAULT_ENCODING, choices=_get_available_encodings(), help='Specify the encoding of the output.', ) parser.add_argument( '--extension', type=str, default=None, choices=_get_available_extensions(), help='Specify the extension of the file.', ) parser.add_argument( '-m', '--method', default='', help='Specify a method of extraction for formats that support it', ) parser.add_argument( '-o', '--output', type=FileType('wb'), default='-', help='Output raw text in this file', ) parser.add_argument( '-O', '--option', type=str, action=AddToNamespaceAction, help=( 'Add arbitrary options to various parsers of the form ' 'KEYWORD=VALUE. A full list of available KEYWORD options is ' 'available at http://bit.ly/textract-options' ), ) parser.add_argument( '-v', '--version', action='version', version='%(prog)s '+VERSION, ) # enable autocompletion with argcomplete argcomplete.autocomplete(parser) return parser
Initialize the parser for the command line interface and bind the autocompletion functionality
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/cli.py#L45-L93
[ "def _get_available_extensions():\n \"\"\"Get a list of available file extensions to make it easy for\n tab-completion and exception handling.\n \"\"\"\n extensions = []\n\n # from filenames\n parsers_dir = os.path.join(os.path.dirname(__file__))\n glob_filename = os.path.join(parsers_dir, \"*\" + _FILENAME_SUFFIX + \".py\")\n ext_re = re.compile(glob_filename.replace('*', \"(?P<ext>\\w+)\"))\n for filename in glob.glob(glob_filename):\n ext_match = ext_re.match(filename)\n ext = ext_match.groups()[0]\n extensions.append(ext)\n extensions.append('.' + ext)\n\n # from relevant synonyms (don't use the '' synonym)\n for ext in EXTENSION_SYNONYMS.keys():\n if ext:\n extensions.append(ext)\n extensions.append(ext.replace('.', '', 1))\n extensions.sort()\n return extensions\n", "def _get_available_encodings():\n \"\"\"Get a list of the available encodings to make it easy to\n tab-complete the command line interface.\n\n Inspiration from http://stackoverflow.com/a/3824405/564709\n \"\"\"\n available_encodings = set(encodings.aliases.aliases.values())\n paths = [os.path.dirname(encodings.__file__)]\n for importer, modname, ispkg in pkgutil.walk_packages(path=paths):\n available_encodings.add(modname)\n available_encodings = list(available_encodings)\n available_encodings.sort()\n return available_encodings\n" ]
""" Use argparse to handle command-line arguments. """ import argparse import encodings import os import pkgutil import sys import six import re import glob import argcomplete from . import VERSION from .parsers import DEFAULT_ENCODING, _get_available_extensions class AddToNamespaceAction(argparse.Action): """This adds KEY,VALUE arbitrary pairs to the argparse.Namespace object """ def __call__(self, parser, namespace, values, option_string=None): key, val = values.strip().split('=') if hasattr(namespace, key): parser.error(( 'Duplicate specification of the key "%(key)s" with --option.' ) % locals()) setattr(namespace, key, val) # Fix FileType to honor 'b' flag, see: https://bugs.python.org/issue14156 class FileType(argparse.FileType): def __call__(self, string): if string == '-' and six.PY3: if 'r' in self._mode: string = sys.stdin.fileno() elif 'w' in self._mode: string = sys.stdout.fileno() return super(FileType, self).__call__(string) # This function is necessary to enable autodocumentation of the script # output def _get_available_encodings(): """Get a list of the available encodings to make it easy to tab-complete the command line interface. Inspiration from http://stackoverflow.com/a/3824405/564709 """ available_encodings = set(encodings.aliases.aliases.values()) paths = [os.path.dirname(encodings.__file__)] for importer, modname, ispkg in pkgutil.walk_packages(path=paths): available_encodings.add(modname) available_encodings = list(available_encodings) available_encodings.sort() return available_encodings
deanmalmgren/textract
textract/cli.py
_get_available_encodings
python
def _get_available_encodings(): available_encodings = set(encodings.aliases.aliases.values()) paths = [os.path.dirname(encodings.__file__)] for importer, modname, ispkg in pkgutil.walk_packages(path=paths): available_encodings.add(modname) available_encodings = list(available_encodings) available_encodings.sort() return available_encodings
Get a list of the available encodings to make it easy to tab-complete the command line interface. Inspiration from http://stackoverflow.com/a/3824405/564709
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/cli.py#L96-L108
null
""" Use argparse to handle command-line arguments. """ import argparse import encodings import os import pkgutil import sys import six import re import glob import argcomplete from . import VERSION from .parsers import DEFAULT_ENCODING, _get_available_extensions class AddToNamespaceAction(argparse.Action): """This adds KEY,VALUE arbitrary pairs to the argparse.Namespace object """ def __call__(self, parser, namespace, values, option_string=None): key, val = values.strip().split('=') if hasattr(namespace, key): parser.error(( 'Duplicate specification of the key "%(key)s" with --option.' ) % locals()) setattr(namespace, key, val) # Fix FileType to honor 'b' flag, see: https://bugs.python.org/issue14156 class FileType(argparse.FileType): def __call__(self, string): if string == '-' and six.PY3: if 'r' in self._mode: string = sys.stdin.fileno() elif 'w' in self._mode: string = sys.stdout.fileno() return super(FileType, self).__call__(string) # This function is necessary to enable autodocumentation of the script # output def get_parser(): """Initialize the parser for the command line interface and bind the autocompletion functionality""" # initialize the parser parser = argparse.ArgumentParser( description=( 'Command line tool for extracting text from any document. ' ) % locals(), ) # define the command line options here parser.add_argument( 'filename', help='Filename to extract text.', ).completer = argcomplete.completers.FilesCompleter parser.add_argument( '-e', '--encoding', type=str, default=DEFAULT_ENCODING, choices=_get_available_encodings(), help='Specify the encoding of the output.', ) parser.add_argument( '--extension', type=str, default=None, choices=_get_available_extensions(), help='Specify the extension of the file.', ) parser.add_argument( '-m', '--method', default='', help='Specify a method of extraction for formats that support it', ) parser.add_argument( '-o', '--output', type=FileType('wb'), default='-', help='Output raw text in this file', ) parser.add_argument( '-O', '--option', type=str, action=AddToNamespaceAction, help=( 'Add arbitrary options to various parsers of the form ' 'KEYWORD=VALUE. A full list of available KEYWORD options is ' 'available at http://bit.ly/textract-options' ), ) parser.add_argument( '-v', '--version', action='version', version='%(prog)s '+VERSION, ) # enable autocompletion with argcomplete argcomplete.autocomplete(parser) return parser
deanmalmgren/textract
textract/parsers/json_parser.py
Parser.get_text
python
def get_text(self, deserialized_json): if isinstance(deserialized_json, dict): result = '' for key in sorted(deserialized_json): result += self.get_text(deserialized_json[key]) + ' ' return result if isinstance(deserialized_json, list): result = '' for item in deserialized_json: result += self.get_text(item) + ' ' return result if isinstance(deserialized_json, six.string_types): return deserialized_json else: return ''
Recursively get text from subcomponents of a deserialized json. To enforce the same order on the documents, make sure to read keys of deserialized_json in a consistent (alphabetical) order.
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/json_parser.py#L18-L38
[ "def get_text(self, deserialized_json):\n \"\"\"Recursively get text from subcomponents of a deserialized json. To\n enforce the same order on the documents, make sure to read keys of\n deserialized_json in a consistent (alphabetical) order.\n \"\"\"\n if isinstance(deserialized_json, dict):\n result = ''\n for key in sorted(deserialized_json):\n result += self.get_text(deserialized_json[key]) + ' '\n return result\n\n if isinstance(deserialized_json, list):\n result = ''\n for item in deserialized_json:\n result += self.get_text(item) + ' '\n return result\n\n if isinstance(deserialized_json, six.string_types):\n return deserialized_json\n else:\n return ''\n" ]
class Parser(BaseParser): """Extract all of the string values of a json file (no keys as those are, in some sense, markup). This is useful for parsing content from mongodb dumps, for example. """ def extract(self, filename, **kwargs): with open(filename, 'r') as raw: deserialized_json = json.load(raw) return self.get_text(deserialized_json)
deanmalmgren/textract
textract/parsers/pdf_parser.py
Parser.extract_pdftotext
python
def extract_pdftotext(self, filename, **kwargs): if 'layout' in kwargs: args = ['pdftotext', '-layout', filename, '-'] else: args = ['pdftotext', filename, '-'] stdout, _ = self.run(args) return stdout
Extract text from pdfs using the pdftotext command line utility.
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/pdf_parser.py#L37-L44
[ "def run(self, args):\n \"\"\"Run ``command`` and return the subsequent ``stdout`` and ``stderr``\n as a tuple. If the command is not successful, this raises a\n :exc:`textract.exceptions.ShellError`.\n \"\"\"\n\n # run a subprocess and put the stdout and stderr on the pipe object\n try:\n pipe = subprocess.Popen(\n args,\n stdout=subprocess.PIPE, stderr=subprocess.PIPE,\n )\n except OSError as e:\n if e.errno == errno.ENOENT:\n # File not found.\n # This is equivalent to getting exitcode 127 from sh\n raise exceptions.ShellError(\n ' '.join(args), 127, '', '',\n )\n\n # pipe.wait() ends up hanging on large files. using\n # pipe.communicate appears to avoid this issue\n stdout, stderr = pipe.communicate()\n\n # if pipe is busted, raise an error (unlike Fabric)\n if pipe.returncode != 0:\n raise exceptions.ShellError(\n ' '.join(args), pipe.returncode, stdout, stderr,\n )\n\n return stdout, stderr\n" ]
class Parser(ShellParser): """Extract text from pdf files using either the ``pdftotext`` method (default) or the ``pdfminer`` method. """ def extract(self, filename, method='', **kwargs): if method == '' or method == 'pdftotext': try: return self.extract_pdftotext(filename, **kwargs) except ShellError as ex: # If pdftotext isn't installed and the pdftotext method # wasn't specified, then gracefully fallback to using # pdfminer instead. if method == '' and ex.is_not_installed(): return self.extract_pdfminer(filename, **kwargs) else: raise ex elif method == 'pdfminer': return self.extract_pdfminer(filename, **kwargs) elif method == 'tesseract': return self.extract_tesseract(filename, **kwargs) else: raise UnknownMethod(method) def extract_pdfminer(self, filename, **kwargs): """Extract text from pdfs using pdfminer.""" stdout, _ = self.run(['pdf2txt.py', filename]) return stdout def extract_tesseract(self, filename, **kwargs): """Extract text from pdfs using tesseract (per-page OCR).""" temp_dir = mkdtemp() base = os.path.join(temp_dir, 'conv') contents = [] try: stdout, _ = self.run(['pdftoppm', filename, base]) for page in sorted(os.listdir(temp_dir)): page_path = os.path.join(temp_dir, page) page_content = TesseractParser().extract(page_path, **kwargs) contents.append(page_content) return six.b('').join(contents) finally: shutil.rmtree(temp_dir)
deanmalmgren/textract
textract/parsers/pdf_parser.py
Parser.extract_pdfminer
python
def extract_pdfminer(self, filename, **kwargs): stdout, _ = self.run(['pdf2txt.py', filename]) return stdout
Extract text from pdfs using pdfminer.
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/pdf_parser.py#L46-L49
[ "def run(self, args):\n \"\"\"Run ``command`` and return the subsequent ``stdout`` and ``stderr``\n as a tuple. If the command is not successful, this raises a\n :exc:`textract.exceptions.ShellError`.\n \"\"\"\n\n # run a subprocess and put the stdout and stderr on the pipe object\n try:\n pipe = subprocess.Popen(\n args,\n stdout=subprocess.PIPE, stderr=subprocess.PIPE,\n )\n except OSError as e:\n if e.errno == errno.ENOENT:\n # File not found.\n # This is equivalent to getting exitcode 127 from sh\n raise exceptions.ShellError(\n ' '.join(args), 127, '', '',\n )\n\n # pipe.wait() ends up hanging on large files. using\n # pipe.communicate appears to avoid this issue\n stdout, stderr = pipe.communicate()\n\n # if pipe is busted, raise an error (unlike Fabric)\n if pipe.returncode != 0:\n raise exceptions.ShellError(\n ' '.join(args), pipe.returncode, stdout, stderr,\n )\n\n return stdout, stderr\n" ]
class Parser(ShellParser): """Extract text from pdf files using either the ``pdftotext`` method (default) or the ``pdfminer`` method. """ def extract(self, filename, method='', **kwargs): if method == '' or method == 'pdftotext': try: return self.extract_pdftotext(filename, **kwargs) except ShellError as ex: # If pdftotext isn't installed and the pdftotext method # wasn't specified, then gracefully fallback to using # pdfminer instead. if method == '' and ex.is_not_installed(): return self.extract_pdfminer(filename, **kwargs) else: raise ex elif method == 'pdfminer': return self.extract_pdfminer(filename, **kwargs) elif method == 'tesseract': return self.extract_tesseract(filename, **kwargs) else: raise UnknownMethod(method) def extract_pdftotext(self, filename, **kwargs): """Extract text from pdfs using the pdftotext command line utility.""" if 'layout' in kwargs: args = ['pdftotext', '-layout', filename, '-'] else: args = ['pdftotext', filename, '-'] stdout, _ = self.run(args) return stdout def extract_tesseract(self, filename, **kwargs): """Extract text from pdfs using tesseract (per-page OCR).""" temp_dir = mkdtemp() base = os.path.join(temp_dir, 'conv') contents = [] try: stdout, _ = self.run(['pdftoppm', filename, base]) for page in sorted(os.listdir(temp_dir)): page_path = os.path.join(temp_dir, page) page_content = TesseractParser().extract(page_path, **kwargs) contents.append(page_content) return six.b('').join(contents) finally: shutil.rmtree(temp_dir)
deanmalmgren/textract
textract/parsers/pdf_parser.py
Parser.extract_tesseract
python
def extract_tesseract(self, filename, **kwargs): temp_dir = mkdtemp() base = os.path.join(temp_dir, 'conv') contents = [] try: stdout, _ = self.run(['pdftoppm', filename, base]) for page in sorted(os.listdir(temp_dir)): page_path = os.path.join(temp_dir, page) page_content = TesseractParser().extract(page_path, **kwargs) contents.append(page_content) return six.b('').join(contents) finally: shutil.rmtree(temp_dir)
Extract text from pdfs using tesseract (per-page OCR).
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/pdf_parser.py#L51-L65
[ "def extract(self, filename, **kwargs):\n\n # if language given as argument, specify language for tesseract to use\n if 'language' in kwargs:\n args = ['tesseract', filename, 'stdout', '-l', kwargs['language']]\n else:\n args = ['tesseract', filename, 'stdout']\n\n stdout, _ = self.run(args)\n return stdout\n", "def run(self, args):\n \"\"\"Run ``command`` and return the subsequent ``stdout`` and ``stderr``\n as a tuple. If the command is not successful, this raises a\n :exc:`textract.exceptions.ShellError`.\n \"\"\"\n\n # run a subprocess and put the stdout and stderr on the pipe object\n try:\n pipe = subprocess.Popen(\n args,\n stdout=subprocess.PIPE, stderr=subprocess.PIPE,\n )\n except OSError as e:\n if e.errno == errno.ENOENT:\n # File not found.\n # This is equivalent to getting exitcode 127 from sh\n raise exceptions.ShellError(\n ' '.join(args), 127, '', '',\n )\n\n # pipe.wait() ends up hanging on large files. using\n # pipe.communicate appears to avoid this issue\n stdout, stderr = pipe.communicate()\n\n # if pipe is busted, raise an error (unlike Fabric)\n if pipe.returncode != 0:\n raise exceptions.ShellError(\n ' '.join(args), pipe.returncode, stdout, stderr,\n )\n\n return stdout, stderr\n" ]
class Parser(ShellParser): """Extract text from pdf files using either the ``pdftotext`` method (default) or the ``pdfminer`` method. """ def extract(self, filename, method='', **kwargs): if method == '' or method == 'pdftotext': try: return self.extract_pdftotext(filename, **kwargs) except ShellError as ex: # If pdftotext isn't installed and the pdftotext method # wasn't specified, then gracefully fallback to using # pdfminer instead. if method == '' and ex.is_not_installed(): return self.extract_pdfminer(filename, **kwargs) else: raise ex elif method == 'pdfminer': return self.extract_pdfminer(filename, **kwargs) elif method == 'tesseract': return self.extract_tesseract(filename, **kwargs) else: raise UnknownMethod(method) def extract_pdftotext(self, filename, **kwargs): """Extract text from pdfs using the pdftotext command line utility.""" if 'layout' in kwargs: args = ['pdftotext', '-layout', filename, '-'] else: args = ['pdftotext', filename, '-'] stdout, _ = self.run(args) return stdout def extract_pdfminer(self, filename, **kwargs): """Extract text from pdfs using pdfminer.""" stdout, _ = self.run(['pdf2txt.py', filename]) return stdout
deanmalmgren/textract
textract/parsers/audio.py
Parser.convert_to_wav
python
def convert_to_wav(self, filename): temp_filename = '{0}.wav'.format(self.temp_filename()) self.run(['sox', '-G', '-c', '1', filename, temp_filename]) return temp_filename
Uses sox cmdline tool, to convert audio file to .wav Note: for testing, use - http://www.text2speech.org/, with American Male 2 for best results
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/audio.py#L56-L66
[ "def run(self, args):\n \"\"\"Run ``command`` and return the subsequent ``stdout`` and ``stderr``\n as a tuple. If the command is not successful, this raises a\n :exc:`textract.exceptions.ShellError`.\n \"\"\"\n\n # run a subprocess and put the stdout and stderr on the pipe object\n try:\n pipe = subprocess.Popen(\n args,\n stdout=subprocess.PIPE, stderr=subprocess.PIPE,\n )\n except OSError as e:\n if e.errno == errno.ENOENT:\n # File not found.\n # This is equivalent to getting exitcode 127 from sh\n raise exceptions.ShellError(\n ' '.join(args), 127, '', '',\n )\n\n # pipe.wait() ends up hanging on large files. using\n # pipe.communicate appears to avoid this issue\n stdout, stderr = pipe.communicate()\n\n # if pipe is busted, raise an error (unlike Fabric)\n if pipe.returncode != 0:\n raise exceptions.ShellError(\n ' '.join(args), pipe.returncode, stdout, stderr,\n )\n\n return stdout, stderr\n", "def temp_filename(self):\n \"\"\"Return a unique tempfile name.\n \"\"\"\n # TODO: it would be nice to get this to behave more like a\n # context so we can make sure these temporary files are\n # removed, regardless of whether an error occurs or the\n # program is terminated.\n handle, filename = tempfile.mkstemp()\n os.close(handle)\n return filename\n" ]
class Parser(ShellParser): """ Extract text (i.e. speech) from an audio file, using SpeechRecognition. Since SpeechRecognition expects a .wav file, with 1 channel, the audio file has to be converted, via sox, if not compliant Note: for testing, use - http://www2.research.att.com/~ttsweb/tts/demo.php, with Rich (US English) for best results """ def extract(self, filename, method='', **kwargs): speech = '' # convert to wav, if not already .wav base, ext = os.path.splitext(filename) if ext != '.wav': temp_filename = self.convert_to_wav(filename) try: speech = self.extract(temp_filename, method, **kwargs) finally: # make sure temp_file is deleted os.remove(temp_filename) else: r = sr.Recognizer() with sr.WavFile(filename) as source: audio = r.record(source) try: if method == 'google' or method == '': speech = r.recognize_google(audio) elif method == 'sphinx': speech = r.recognize_sphinx(audio) else: raise UnknownMethod(method) except LookupError: # audio is not understandable speech = '' except sr.UnknownValueError: speech = '' except sr.RequestError as e: speech = '' # add a newline, to make output cleaner speech += '\n' return speech
deanmalmgren/textract
textract/parsers/html_parser.py
Parser._visible
python
def _visible(self, element): if element.name in self._disallowed_names: return False elif re.match(u'<!--.*-->', six.text_type(element.extract())): return False return True
Used to filter text elements that have invisible text on the page.
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/html_parser.py#L27-L34
null
class Parser(BaseParser): """Extract text from html file using beautifulsoup4. Filter text to only show the visible parts of the page. Insipration from `here <http://stackoverflow.com/a/1983219/564709>`_. """ _disallowed_names = [ 'style', 'script', '[document]', 'head', 'title', 'html', 'meta', 'link', 'body', ] _inline_tags = [ 'b', 'big', 'i', 'small', 'tt', 'abbr', 'acronym', 'cite', 'code', 'dfn', 'em', 'kbd', 'strong', 'samp', 'var', 'a', 'bdo', 'br', 'img', 'map', 'object', 'q', 'script', 'span', 'sub', 'sup', 'button', 'input', 'label', 'select', 'textarea', ] def _inline(self, element): """Used to check whether given element can be treated as inline element (without new line after). """ if element.name in self._inline_tags: return True return False def _find_any_text(self, tag): """Looks for any possible text within given tag. """ text = '' if tag is not None: text = six.text_type(tag) text = re.sub(r'(<[^>]+>)', '', text) text = re.sub(r'\s', ' ', text) text = text.strip() return text def _parse_tables(self, soup): """Returns array containing basic informations about tables for ASCII replacement (look: _replace_tables()). """ tables = [] for t in soup.find_all('table'): t_dict = {'width': 0, 'table': t, 'trs': [], 'col_width': {}} trs = t.find_all('tr') if len(trs) > 0: for tr in trs: tr_dict = [] tds = tr.find_all('th') + tr.find_all('td') if len(tds) > 0: for i, td in enumerate(tds): td_text = self._find_any_text(td) length = len(td_text) if i in t_dict['col_width']: t_dict['col_width'][i] = max( length, t_dict['col_width'][i] ) else: t_dict['col_width'][i] = length tr_dict.append({ 'text': td_text, 'colspan': int(td.get('colspan', 1)), }) t_dict['trs'].append(tr_dict) for col in t_dict['col_width']: t_dict['width'] += t_dict['col_width'][col] tables.append(t_dict) return tables def _replace_tables(self, soup, v_separator=' | ', h_separator='-'): """Replaces <table> elements with its ASCII equivalent. """ tables = self._parse_tables(soup) v_sep_len = len(v_separator) v_left_sep = v_separator.lstrip() for t in tables: html = '' trs = t['trs'] h_length = 1 + (v_sep_len * len(t['col_width'])) + t['width'] head_foot = (h_separator * h_length) + "\n" html += head_foot for tr in trs: html += v_left_sep for i, td in enumerate(tr): text = td['text'] col_width = t['col_width'][i] + v_sep_len if td['colspan'] > 1: for j in range(td['colspan']-1): j = j + 1 if (i+j) < len(t['col_width']): col_width += t['col_width'][i+j] + v_sep_len html += ('%' + str(col_width) + 's') % (text + v_separator) html += "\n" html += head_foot new_table = soup.new_tag('div') new_table.string = html t['table'].replace_with(new_table) return soup def _join_inlines(self, soup): """Unwraps inline elements defined in self._inline_tags. """ elements = soup.find_all(True) for elem in elements: if self._inline(elem): elem.unwrap() return soup def extract(self, filename, **kwargs): with open(filename, "rb") as stream: soup = BeautifulSoup(stream, 'lxml') # Convert tables to ASCII ones soup = self._replace_tables(soup) # Join inline elements soup = self._join_inlines(soup) # Make HTML html = '' elements = soup.find_all(True) elements = [el for el in filter(self._visible, elements)] for elem in elements: string = elem.string if string is None: string = self._find_any_text(elem) string = string.strip() if len(string) > 0: html += "\n" + string + "\n" return html
deanmalmgren/textract
textract/parsers/html_parser.py
Parser._find_any_text
python
def _find_any_text(self, tag): text = '' if tag is not None: text = six.text_type(tag) text = re.sub(r'(<[^>]+>)', '', text) text = re.sub(r'\s', ' ', text) text = text.strip() return text
Looks for any possible text within given tag.
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/html_parser.py#L44-L53
null
class Parser(BaseParser): """Extract text from html file using beautifulsoup4. Filter text to only show the visible parts of the page. Insipration from `here <http://stackoverflow.com/a/1983219/564709>`_. """ _disallowed_names = [ 'style', 'script', '[document]', 'head', 'title', 'html', 'meta', 'link', 'body', ] _inline_tags = [ 'b', 'big', 'i', 'small', 'tt', 'abbr', 'acronym', 'cite', 'code', 'dfn', 'em', 'kbd', 'strong', 'samp', 'var', 'a', 'bdo', 'br', 'img', 'map', 'object', 'q', 'script', 'span', 'sub', 'sup', 'button', 'input', 'label', 'select', 'textarea', ] def _visible(self, element): """Used to filter text elements that have invisible text on the page. """ if element.name in self._disallowed_names: return False elif re.match(u'<!--.*-->', six.text_type(element.extract())): return False return True def _inline(self, element): """Used to check whether given element can be treated as inline element (without new line after). """ if element.name in self._inline_tags: return True return False def _parse_tables(self, soup): """Returns array containing basic informations about tables for ASCII replacement (look: _replace_tables()). """ tables = [] for t in soup.find_all('table'): t_dict = {'width': 0, 'table': t, 'trs': [], 'col_width': {}} trs = t.find_all('tr') if len(trs) > 0: for tr in trs: tr_dict = [] tds = tr.find_all('th') + tr.find_all('td') if len(tds) > 0: for i, td in enumerate(tds): td_text = self._find_any_text(td) length = len(td_text) if i in t_dict['col_width']: t_dict['col_width'][i] = max( length, t_dict['col_width'][i] ) else: t_dict['col_width'][i] = length tr_dict.append({ 'text': td_text, 'colspan': int(td.get('colspan', 1)), }) t_dict['trs'].append(tr_dict) for col in t_dict['col_width']: t_dict['width'] += t_dict['col_width'][col] tables.append(t_dict) return tables def _replace_tables(self, soup, v_separator=' | ', h_separator='-'): """Replaces <table> elements with its ASCII equivalent. """ tables = self._parse_tables(soup) v_sep_len = len(v_separator) v_left_sep = v_separator.lstrip() for t in tables: html = '' trs = t['trs'] h_length = 1 + (v_sep_len * len(t['col_width'])) + t['width'] head_foot = (h_separator * h_length) + "\n" html += head_foot for tr in trs: html += v_left_sep for i, td in enumerate(tr): text = td['text'] col_width = t['col_width'][i] + v_sep_len if td['colspan'] > 1: for j in range(td['colspan']-1): j = j + 1 if (i+j) < len(t['col_width']): col_width += t['col_width'][i+j] + v_sep_len html += ('%' + str(col_width) + 's') % (text + v_separator) html += "\n" html += head_foot new_table = soup.new_tag('div') new_table.string = html t['table'].replace_with(new_table) return soup def _join_inlines(self, soup): """Unwraps inline elements defined in self._inline_tags. """ elements = soup.find_all(True) for elem in elements: if self._inline(elem): elem.unwrap() return soup def extract(self, filename, **kwargs): with open(filename, "rb") as stream: soup = BeautifulSoup(stream, 'lxml') # Convert tables to ASCII ones soup = self._replace_tables(soup) # Join inline elements soup = self._join_inlines(soup) # Make HTML html = '' elements = soup.find_all(True) elements = [el for el in filter(self._visible, elements)] for elem in elements: string = elem.string if string is None: string = self._find_any_text(elem) string = string.strip() if len(string) > 0: html += "\n" + string + "\n" return html
deanmalmgren/textract
textract/parsers/html_parser.py
Parser._parse_tables
python
def _parse_tables(self, soup): tables = [] for t in soup.find_all('table'): t_dict = {'width': 0, 'table': t, 'trs': [], 'col_width': {}} trs = t.find_all('tr') if len(trs) > 0: for tr in trs: tr_dict = [] tds = tr.find_all('th') + tr.find_all('td') if len(tds) > 0: for i, td in enumerate(tds): td_text = self._find_any_text(td) length = len(td_text) if i in t_dict['col_width']: t_dict['col_width'][i] = max( length, t_dict['col_width'][i] ) else: t_dict['col_width'][i] = length tr_dict.append({ 'text': td_text, 'colspan': int(td.get('colspan', 1)), }) t_dict['trs'].append(tr_dict) for col in t_dict['col_width']: t_dict['width'] += t_dict['col_width'][col] tables.append(t_dict) return tables
Returns array containing basic informations about tables for ASCII replacement (look: _replace_tables()).
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/html_parser.py#L55-L86
null
class Parser(BaseParser): """Extract text from html file using beautifulsoup4. Filter text to only show the visible parts of the page. Insipration from `here <http://stackoverflow.com/a/1983219/564709>`_. """ _disallowed_names = [ 'style', 'script', '[document]', 'head', 'title', 'html', 'meta', 'link', 'body', ] _inline_tags = [ 'b', 'big', 'i', 'small', 'tt', 'abbr', 'acronym', 'cite', 'code', 'dfn', 'em', 'kbd', 'strong', 'samp', 'var', 'a', 'bdo', 'br', 'img', 'map', 'object', 'q', 'script', 'span', 'sub', 'sup', 'button', 'input', 'label', 'select', 'textarea', ] def _visible(self, element): """Used to filter text elements that have invisible text on the page. """ if element.name in self._disallowed_names: return False elif re.match(u'<!--.*-->', six.text_type(element.extract())): return False return True def _inline(self, element): """Used to check whether given element can be treated as inline element (without new line after). """ if element.name in self._inline_tags: return True return False def _find_any_text(self, tag): """Looks for any possible text within given tag. """ text = '' if tag is not None: text = six.text_type(tag) text = re.sub(r'(<[^>]+>)', '', text) text = re.sub(r'\s', ' ', text) text = text.strip() return text def _replace_tables(self, soup, v_separator=' | ', h_separator='-'): """Replaces <table> elements with its ASCII equivalent. """ tables = self._parse_tables(soup) v_sep_len = len(v_separator) v_left_sep = v_separator.lstrip() for t in tables: html = '' trs = t['trs'] h_length = 1 + (v_sep_len * len(t['col_width'])) + t['width'] head_foot = (h_separator * h_length) + "\n" html += head_foot for tr in trs: html += v_left_sep for i, td in enumerate(tr): text = td['text'] col_width = t['col_width'][i] + v_sep_len if td['colspan'] > 1: for j in range(td['colspan']-1): j = j + 1 if (i+j) < len(t['col_width']): col_width += t['col_width'][i+j] + v_sep_len html += ('%' + str(col_width) + 's') % (text + v_separator) html += "\n" html += head_foot new_table = soup.new_tag('div') new_table.string = html t['table'].replace_with(new_table) return soup def _join_inlines(self, soup): """Unwraps inline elements defined in self._inline_tags. """ elements = soup.find_all(True) for elem in elements: if self._inline(elem): elem.unwrap() return soup def extract(self, filename, **kwargs): with open(filename, "rb") as stream: soup = BeautifulSoup(stream, 'lxml') # Convert tables to ASCII ones soup = self._replace_tables(soup) # Join inline elements soup = self._join_inlines(soup) # Make HTML html = '' elements = soup.find_all(True) elements = [el for el in filter(self._visible, elements)] for elem in elements: string = elem.string if string is None: string = self._find_any_text(elem) string = string.strip() if len(string) > 0: html += "\n" + string + "\n" return html
deanmalmgren/textract
textract/parsers/html_parser.py
Parser._replace_tables
python
def _replace_tables(self, soup, v_separator=' | ', h_separator='-'): tables = self._parse_tables(soup) v_sep_len = len(v_separator) v_left_sep = v_separator.lstrip() for t in tables: html = '' trs = t['trs'] h_length = 1 + (v_sep_len * len(t['col_width'])) + t['width'] head_foot = (h_separator * h_length) + "\n" html += head_foot for tr in trs: html += v_left_sep for i, td in enumerate(tr): text = td['text'] col_width = t['col_width'][i] + v_sep_len if td['colspan'] > 1: for j in range(td['colspan']-1): j = j + 1 if (i+j) < len(t['col_width']): col_width += t['col_width'][i+j] + v_sep_len html += ('%' + str(col_width) + 's') % (text + v_separator) html += "\n" html += head_foot new_table = soup.new_tag('div') new_table.string = html t['table'].replace_with(new_table) return soup
Replaces <table> elements with its ASCII equivalent.
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/html_parser.py#L88-L116
[ "def _parse_tables(self, soup):\n \"\"\"Returns array containing basic informations about tables for ASCII\n replacement (look: _replace_tables()).\n \"\"\"\n tables = []\n for t in soup.find_all('table'):\n t_dict = {'width': 0, 'table': t, 'trs': [], 'col_width': {}}\n trs = t.find_all('tr')\n if len(trs) > 0:\n for tr in trs:\n tr_dict = []\n tds = tr.find_all('th') + tr.find_all('td')\n if len(tds) > 0:\n for i, td in enumerate(tds):\n td_text = self._find_any_text(td)\n length = len(td_text)\n if i in t_dict['col_width']:\n t_dict['col_width'][i] = max(\n length,\n t_dict['col_width'][i]\n )\n else:\n t_dict['col_width'][i] = length\n tr_dict.append({\n 'text': td_text,\n 'colspan': int(td.get('colspan', 1)),\n })\n t_dict['trs'].append(tr_dict)\n for col in t_dict['col_width']:\n t_dict['width'] += t_dict['col_width'][col]\n tables.append(t_dict)\n return tables\n" ]
class Parser(BaseParser): """Extract text from html file using beautifulsoup4. Filter text to only show the visible parts of the page. Insipration from `here <http://stackoverflow.com/a/1983219/564709>`_. """ _disallowed_names = [ 'style', 'script', '[document]', 'head', 'title', 'html', 'meta', 'link', 'body', ] _inline_tags = [ 'b', 'big', 'i', 'small', 'tt', 'abbr', 'acronym', 'cite', 'code', 'dfn', 'em', 'kbd', 'strong', 'samp', 'var', 'a', 'bdo', 'br', 'img', 'map', 'object', 'q', 'script', 'span', 'sub', 'sup', 'button', 'input', 'label', 'select', 'textarea', ] def _visible(self, element): """Used to filter text elements that have invisible text on the page. """ if element.name in self._disallowed_names: return False elif re.match(u'<!--.*-->', six.text_type(element.extract())): return False return True def _inline(self, element): """Used to check whether given element can be treated as inline element (without new line after). """ if element.name in self._inline_tags: return True return False def _find_any_text(self, tag): """Looks for any possible text within given tag. """ text = '' if tag is not None: text = six.text_type(tag) text = re.sub(r'(<[^>]+>)', '', text) text = re.sub(r'\s', ' ', text) text = text.strip() return text def _parse_tables(self, soup): """Returns array containing basic informations about tables for ASCII replacement (look: _replace_tables()). """ tables = [] for t in soup.find_all('table'): t_dict = {'width': 0, 'table': t, 'trs': [], 'col_width': {}} trs = t.find_all('tr') if len(trs) > 0: for tr in trs: tr_dict = [] tds = tr.find_all('th') + tr.find_all('td') if len(tds) > 0: for i, td in enumerate(tds): td_text = self._find_any_text(td) length = len(td_text) if i in t_dict['col_width']: t_dict['col_width'][i] = max( length, t_dict['col_width'][i] ) else: t_dict['col_width'][i] = length tr_dict.append({ 'text': td_text, 'colspan': int(td.get('colspan', 1)), }) t_dict['trs'].append(tr_dict) for col in t_dict['col_width']: t_dict['width'] += t_dict['col_width'][col] tables.append(t_dict) return tables def _join_inlines(self, soup): """Unwraps inline elements defined in self._inline_tags. """ elements = soup.find_all(True) for elem in elements: if self._inline(elem): elem.unwrap() return soup def extract(self, filename, **kwargs): with open(filename, "rb") as stream: soup = BeautifulSoup(stream, 'lxml') # Convert tables to ASCII ones soup = self._replace_tables(soup) # Join inline elements soup = self._join_inlines(soup) # Make HTML html = '' elements = soup.find_all(True) elements = [el for el in filter(self._visible, elements)] for elem in elements: string = elem.string if string is None: string = self._find_any_text(elem) string = string.strip() if len(string) > 0: html += "\n" + string + "\n" return html
deanmalmgren/textract
textract/parsers/html_parser.py
Parser._join_inlines
python
def _join_inlines(self, soup): elements = soup.find_all(True) for elem in elements: if self._inline(elem): elem.unwrap() return soup
Unwraps inline elements defined in self._inline_tags.
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/html_parser.py#L118-L125
[ "def _inline(self, element):\n \"\"\"Used to check whether given element can be treated as inline\n element (without new line after).\n \"\"\"\n if element.name in self._inline_tags:\n return True\n return False\n" ]
class Parser(BaseParser): """Extract text from html file using beautifulsoup4. Filter text to only show the visible parts of the page. Insipration from `here <http://stackoverflow.com/a/1983219/564709>`_. """ _disallowed_names = [ 'style', 'script', '[document]', 'head', 'title', 'html', 'meta', 'link', 'body', ] _inline_tags = [ 'b', 'big', 'i', 'small', 'tt', 'abbr', 'acronym', 'cite', 'code', 'dfn', 'em', 'kbd', 'strong', 'samp', 'var', 'a', 'bdo', 'br', 'img', 'map', 'object', 'q', 'script', 'span', 'sub', 'sup', 'button', 'input', 'label', 'select', 'textarea', ] def _visible(self, element): """Used to filter text elements that have invisible text on the page. """ if element.name in self._disallowed_names: return False elif re.match(u'<!--.*-->', six.text_type(element.extract())): return False return True def _inline(self, element): """Used to check whether given element can be treated as inline element (without new line after). """ if element.name in self._inline_tags: return True return False def _find_any_text(self, tag): """Looks for any possible text within given tag. """ text = '' if tag is not None: text = six.text_type(tag) text = re.sub(r'(<[^>]+>)', '', text) text = re.sub(r'\s', ' ', text) text = text.strip() return text def _parse_tables(self, soup): """Returns array containing basic informations about tables for ASCII replacement (look: _replace_tables()). """ tables = [] for t in soup.find_all('table'): t_dict = {'width': 0, 'table': t, 'trs': [], 'col_width': {}} trs = t.find_all('tr') if len(trs) > 0: for tr in trs: tr_dict = [] tds = tr.find_all('th') + tr.find_all('td') if len(tds) > 0: for i, td in enumerate(tds): td_text = self._find_any_text(td) length = len(td_text) if i in t_dict['col_width']: t_dict['col_width'][i] = max( length, t_dict['col_width'][i] ) else: t_dict['col_width'][i] = length tr_dict.append({ 'text': td_text, 'colspan': int(td.get('colspan', 1)), }) t_dict['trs'].append(tr_dict) for col in t_dict['col_width']: t_dict['width'] += t_dict['col_width'][col] tables.append(t_dict) return tables def _replace_tables(self, soup, v_separator=' | ', h_separator='-'): """Replaces <table> elements with its ASCII equivalent. """ tables = self._parse_tables(soup) v_sep_len = len(v_separator) v_left_sep = v_separator.lstrip() for t in tables: html = '' trs = t['trs'] h_length = 1 + (v_sep_len * len(t['col_width'])) + t['width'] head_foot = (h_separator * h_length) + "\n" html += head_foot for tr in trs: html += v_left_sep for i, td in enumerate(tr): text = td['text'] col_width = t['col_width'][i] + v_sep_len if td['colspan'] > 1: for j in range(td['colspan']-1): j = j + 1 if (i+j) < len(t['col_width']): col_width += t['col_width'][i+j] + v_sep_len html += ('%' + str(col_width) + 's') % (text + v_separator) html += "\n" html += head_foot new_table = soup.new_tag('div') new_table.string = html t['table'].replace_with(new_table) return soup def extract(self, filename, **kwargs): with open(filename, "rb") as stream: soup = BeautifulSoup(stream, 'lxml') # Convert tables to ASCII ones soup = self._replace_tables(soup) # Join inline elements soup = self._join_inlines(soup) # Make HTML html = '' elements = soup.find_all(True) elements = [el for el in filter(self._visible, elements)] for elem in elements: string = elem.string if string is None: string = self._find_any_text(elem) string = string.strip() if len(string) > 0: html += "\n" + string + "\n" return html
deanmalmgren/textract
textract/parsers/utils.py
BaseParser.process
python
def process(self, filename, encoding, **kwargs): # make a "unicode sandwich" to handle dealing with unknown # input byte strings and converting them to a predictable # output encoding # http://nedbatchelder.com/text/unipain/unipain.html#35 byte_string = self.extract(filename, **kwargs) unicode_string = self.decode(byte_string) return self.encode(unicode_string, encoding)
Process ``filename`` and encode byte-string with ``encoding``. This method is called by :func:`textract.parsers.process` and wraps the :meth:`.BaseParser.extract` method in `a delicious unicode sandwich <http://nedbatchelder.com/text/unipain.html>`_.
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/utils.py#L35-L48
[ "def extract(self, filename, **kwargs):\n \"\"\"This method must be overwritten by child classes to extract raw\n text from a filename. This method can return either a\n byte-encoded string or unicode.\n \"\"\"\n raise NotImplementedError('must be overwritten by child classes')\n", "def encode(self, text, encoding):\n \"\"\"Encode the ``text`` in ``encoding`` byte-encoding. This ignores\n code points that can't be encoded in byte-strings.\n \"\"\"\n return text.encode(encoding, 'ignore')\n", "def decode(self, text):\n \"\"\"Decode ``text`` using the `chardet\n <https://github.com/chardet/chardet>`_ package.\n \"\"\"\n # only decode byte strings into unicode if it hasn't already\n # been done by a subclass\n if isinstance(text, six.text_type):\n return text\n\n # empty text? nothing to decode\n if not text:\n return u''\n\n # use chardet to automatically detect the encoding text\n result = chardet.detect(text)\n return text.decode(result['encoding'])\n" ]
class BaseParser(object): """The :class:`.BaseParser` abstracts out some common functionality that is used across all document Parsers. In particular, it has the responsibility of handling all unicode and byte-encoding. """ def extract(self, filename, **kwargs): """This method must be overwritten by child classes to extract raw text from a filename. This method can return either a byte-encoded string or unicode. """ raise NotImplementedError('must be overwritten by child classes') def encode(self, text, encoding): """Encode the ``text`` in ``encoding`` byte-encoding. This ignores code points that can't be encoded in byte-strings. """ return text.encode(encoding, 'ignore') def decode(self, text): """Decode ``text`` using the `chardet <https://github.com/chardet/chardet>`_ package. """ # only decode byte strings into unicode if it hasn't already # been done by a subclass if isinstance(text, six.text_type): return text # empty text? nothing to decode if not text: return u'' # use chardet to automatically detect the encoding text result = chardet.detect(text) return text.decode(result['encoding'])
deanmalmgren/textract
textract/parsers/utils.py
BaseParser.decode
python
def decode(self, text): # only decode byte strings into unicode if it hasn't already # been done by a subclass if isinstance(text, six.text_type): return text # empty text? nothing to decode if not text: return u'' # use chardet to automatically detect the encoding text result = chardet.detect(text) return text.decode(result['encoding'])
Decode ``text`` using the `chardet <https://github.com/chardet/chardet>`_ package.
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/utils.py#L50-L65
null
class BaseParser(object): """The :class:`.BaseParser` abstracts out some common functionality that is used across all document Parsers. In particular, it has the responsibility of handling all unicode and byte-encoding. """ def extract(self, filename, **kwargs): """This method must be overwritten by child classes to extract raw text from a filename. This method can return either a byte-encoded string or unicode. """ raise NotImplementedError('must be overwritten by child classes') def encode(self, text, encoding): """Encode the ``text`` in ``encoding`` byte-encoding. This ignores code points that can't be encoded in byte-strings. """ return text.encode(encoding, 'ignore') def process(self, filename, encoding, **kwargs): """Process ``filename`` and encode byte-string with ``encoding``. This method is called by :func:`textract.parsers.process` and wraps the :meth:`.BaseParser.extract` method in `a delicious unicode sandwich <http://nedbatchelder.com/text/unipain.html>`_. """ # make a "unicode sandwich" to handle dealing with unknown # input byte strings and converting them to a predictable # output encoding # http://nedbatchelder.com/text/unipain/unipain.html#35 byte_string = self.extract(filename, **kwargs) unicode_string = self.decode(byte_string) return self.encode(unicode_string, encoding)
deanmalmgren/textract
textract/parsers/utils.py
ShellParser.run
python
def run(self, args): # run a subprocess and put the stdout and stderr on the pipe object try: pipe = subprocess.Popen( args, stdout=subprocess.PIPE, stderr=subprocess.PIPE, ) except OSError as e: if e.errno == errno.ENOENT: # File not found. # This is equivalent to getting exitcode 127 from sh raise exceptions.ShellError( ' '.join(args), 127, '', '', ) # pipe.wait() ends up hanging on large files. using # pipe.communicate appears to avoid this issue stdout, stderr = pipe.communicate() # if pipe is busted, raise an error (unlike Fabric) if pipe.returncode != 0: raise exceptions.ShellError( ' '.join(args), pipe.returncode, stdout, stderr, ) return stdout, stderr
Run ``command`` and return the subsequent ``stdout`` and ``stderr`` as a tuple. If the command is not successful, this raises a :exc:`textract.exceptions.ShellError`.
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/utils.py#L74-L104
null
class ShellParser(BaseParser): """The :class:`.ShellParser` extends the :class:`.BaseParser` to make it easy to run external programs from the command line with `Fabric <http://www.fabfile.org/>`_-like behavior. """ def temp_filename(self): """Return a unique tempfile name. """ # TODO: it would be nice to get this to behave more like a # context so we can make sure these temporary files are # removed, regardless of whether an error occurs or the # program is terminated. handle, filename = tempfile.mkstemp() os.close(handle) return filename
deanmalmgren/textract
textract/parsers/utils.py
ShellParser.temp_filename
python
def temp_filename(self): # TODO: it would be nice to get this to behave more like a # context so we can make sure these temporary files are # removed, regardless of whether an error occurs or the # program is terminated. handle, filename = tempfile.mkstemp() os.close(handle) return filename
Return a unique tempfile name.
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/utils.py#L106-L115
null
class ShellParser(BaseParser): """The :class:`.ShellParser` extends the :class:`.BaseParser` to make it easy to run external programs from the command line with `Fabric <http://www.fabfile.org/>`_-like behavior. """ def run(self, args): """Run ``command`` and return the subsequent ``stdout`` and ``stderr`` as a tuple. If the command is not successful, this raises a :exc:`textract.exceptions.ShellError`. """ # run a subprocess and put the stdout and stderr on the pipe object try: pipe = subprocess.Popen( args, stdout=subprocess.PIPE, stderr=subprocess.PIPE, ) except OSError as e: if e.errno == errno.ENOENT: # File not found. # This is equivalent to getting exitcode 127 from sh raise exceptions.ShellError( ' '.join(args), 127, '', '', ) # pipe.wait() ends up hanging on large files. using # pipe.communicate appears to avoid this issue stdout, stderr = pipe.communicate() # if pipe is busted, raise an error (unlike Fabric) if pipe.returncode != 0: raise exceptions.ShellError( ' '.join(args), pipe.returncode, stdout, stderr, ) return stdout, stderr
deanmalmgren/textract
textract/parsers/__init__.py
process
python
def process(filename, encoding=DEFAULT_ENCODING, extension=None, **kwargs): # make sure the filename exists if not os.path.exists(filename): raise exceptions.MissingFileError(filename) # get the filename extension, which is something like .docx for # example, and import the module dynamically using importlib. This # is a relative import so the name of the package is necessary # normally, file extension will be extracted from the file name # if the file name has no extension, then the user can pass the # extension as an argument if extension: ext = extension # check if the extension has the leading . if not ext.startswith('.'): ext = '.' + ext ext = ext.lower() else: _, ext = os.path.splitext(filename) ext = ext.lower() # check the EXTENSION_SYNONYMS dictionary ext = EXTENSION_SYNONYMS.get(ext, ext) # to avoid conflicts with packages that are installed globally # (e.g. python's json module), all extension parser modules have # the _parser extension rel_module = ext + _FILENAME_SUFFIX # If we can't import the module, the file extension isn't currently # supported try: filetype_module = importlib.import_module( rel_module, 'textract.parsers' ) except ImportError: raise exceptions.ExtensionNotSupported(ext) # do the extraction parser = filetype_module.Parser() return parser.process(filename, encoding, **kwargs)
This is the core function used for extracting text. It routes the ``filename`` to the appropriate parser and returns the extracted text as a byte-string encoded with ``encoding``.
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/__init__.py#L31-L77
null
""" Route the request to the appropriate parser based on file type. """ import os import importlib import glob import re from .. import exceptions # Dictionary structure for synonymous file extension types EXTENSION_SYNONYMS = { ".jpeg": ".jpg", ".tff": ".tiff", ".tif": ".tiff", ".htm": ".html", "": ".txt", ".log": ".txt", } # default encoding that is returned by the process method. specify it # here so the default is used on both the process function and also by # the command line interface DEFAULT_ENCODING = 'utf_8' # filename format _FILENAME_SUFFIX = '_parser' def _get_available_extensions(): """Get a list of available file extensions to make it easy for tab-completion and exception handling. """ extensions = [] # from filenames parsers_dir = os.path.join(os.path.dirname(__file__)) glob_filename = os.path.join(parsers_dir, "*" + _FILENAME_SUFFIX + ".py") ext_re = re.compile(glob_filename.replace('*', "(?P<ext>\w+)")) for filename in glob.glob(glob_filename): ext_match = ext_re.match(filename) ext = ext_match.groups()[0] extensions.append(ext) extensions.append('.' + ext) # from relevant synonyms (don't use the '' synonym) for ext in EXTENSION_SYNONYMS.keys(): if ext: extensions.append(ext) extensions.append(ext.replace('.', '', 1)) extensions.sort() return extensions
deanmalmgren/textract
textract/parsers/__init__.py
_get_available_extensions
python
def _get_available_extensions(): extensions = [] # from filenames parsers_dir = os.path.join(os.path.dirname(__file__)) glob_filename = os.path.join(parsers_dir, "*" + _FILENAME_SUFFIX + ".py") ext_re = re.compile(glob_filename.replace('*', "(?P<ext>\w+)")) for filename in glob.glob(glob_filename): ext_match = ext_re.match(filename) ext = ext_match.groups()[0] extensions.append(ext) extensions.append('.' + ext) # from relevant synonyms (don't use the '' synonym) for ext in EXTENSION_SYNONYMS.keys(): if ext: extensions.append(ext) extensions.append(ext.replace('.', '', 1)) extensions.sort() return extensions
Get a list of available file extensions to make it easy for tab-completion and exception handling.
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/textract/parsers/__init__.py#L80-L102
null
""" Route the request to the appropriate parser based on file type. """ import os import importlib import glob import re from .. import exceptions # Dictionary structure for synonymous file extension types EXTENSION_SYNONYMS = { ".jpeg": ".jpg", ".tff": ".tiff", ".tif": ".tiff", ".htm": ".html", "": ".txt", ".log": ".txt", } # default encoding that is returned by the process method. specify it # here so the default is used on both the process function and also by # the command line interface DEFAULT_ENCODING = 'utf_8' # filename format _FILENAME_SUFFIX = '_parser' def process(filename, encoding=DEFAULT_ENCODING, extension=None, **kwargs): """This is the core function used for extracting text. It routes the ``filename`` to the appropriate parser and returns the extracted text as a byte-string encoded with ``encoding``. """ # make sure the filename exists if not os.path.exists(filename): raise exceptions.MissingFileError(filename) # get the filename extension, which is something like .docx for # example, and import the module dynamically using importlib. This # is a relative import so the name of the package is necessary # normally, file extension will be extracted from the file name # if the file name has no extension, then the user can pass the # extension as an argument if extension: ext = extension # check if the extension has the leading . if not ext.startswith('.'): ext = '.' + ext ext = ext.lower() else: _, ext = os.path.splitext(filename) ext = ext.lower() # check the EXTENSION_SYNONYMS dictionary ext = EXTENSION_SYNONYMS.get(ext, ext) # to avoid conflicts with packages that are installed globally # (e.g. python's json module), all extension parser modules have # the _parser extension rel_module = ext + _FILENAME_SUFFIX # If we can't import the module, the file extension isn't currently # supported try: filetype_module = importlib.import_module( rel_module, 'textract.parsers' ) except ImportError: raise exceptions.ExtensionNotSupported(ext) # do the extraction parser = filetype_module.Parser() return parser.process(filename, encoding, **kwargs)
deanmalmgren/textract
setup.py
parse_requirements
python
def parse_requirements(requirements_filename): dependencies, dependency_links = [], [] requirements_dir = os.path.dirname(requirements_filename) with open(requirements_filename, 'r') as stream: for line in stream: line = line.strip() if line.startswith("-r"): filename = os.path.join(requirements_dir, line[2:].strip()) _dependencies, _dependency_links = parse_requirements(filename) dependencies.extend(_dependencies) dependency_links.extend(_dependency_links) elif line.startswith("http"): dependency_links.append(line) else: package = line.split('#')[0] if package: dependencies.append(package) return dependencies, dependency_links
read in the dependencies from the requirements files
train
https://github.com/deanmalmgren/textract/blob/117ea191d93d80321e4bf01f23cc1ac54d69a075/setup.py#L18-L37
null
import glob import os import sys from setuptools import setup import textract # get all of the scripts scripts = glob.glob("bin/*") # read in the description from README with open("README.rst") as stream: long_description = stream.read() github_url = 'https://github.com/deanmalmgren/textract' requirements_filename = os.path.join("requirements", "python") dependencies, dependency_links = parse_requirements(requirements_filename) setup( name=textract.__name__, version="1.6.1", description="extract text from any document. no muss. no fuss.", long_description=long_description, url=github_url, download_url="%s/archives/master" % github_url, author='Dean Malmgren', author_email='dean.malmgren@datascopeanalytics.com', license='MIT', scripts=scripts, packages=[ 'textract', 'textract.parsers', ], install_requires=dependencies, dependency_links=dependency_links, zip_safe=False, )
PyCQA/pyflakes
pyflakes/reporter.py
Reporter.syntaxError
python
def syntaxError(self, filename, msg, lineno, offset, text): line = text.splitlines()[-1] if offset is not None: if sys.version_info < (3, 8): offset = offset - (len(text) - len(line)) + 1 self._stderr.write('%s:%d:%d: %s\n' % (filename, lineno, offset, msg)) else: self._stderr.write('%s:%d: %s\n' % (filename, lineno, msg)) self._stderr.write(line) self._stderr.write('\n') if offset is not None: self._stderr.write(re.sub(r'\S', ' ', line[:offset - 1]) + "^\n")
There was a syntax error in C{filename}. @param filename: The path to the file with the syntax error. @ptype filename: C{unicode} @param msg: An explanation of the syntax error. @ptype msg: C{unicode} @param lineno: The line number where the syntax error occurred. @ptype lineno: C{int} @param offset: The column on which the syntax error occurred, or None. @ptype offset: C{int} @param text: The source code containing the syntax error. @ptype text: C{unicode}
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/reporter.py#L39-L66
null
class Reporter(object): """ Formats the results of pyflakes checks to users. """ def __init__(self, warningStream, errorStream): """ Construct a L{Reporter}. @param warningStream: A file-like object where warnings will be written to. The stream's C{write} method must accept unicode. C{sys.stdout} is a good value. @param errorStream: A file-like object where error output will be written to. The stream's C{write} method must accept unicode. C{sys.stderr} is a good value. """ self._stdout = warningStream self._stderr = errorStream def unexpectedError(self, filename, msg): """ An unexpected error occurred trying to process C{filename}. @param filename: The path to a file that we could not process. @ptype filename: C{unicode} @param msg: A message explaining the problem. @ptype msg: C{unicode} """ self._stderr.write("%s: %s\n" % (filename, msg)) def flake(self, message): """ pyflakes found something wrong with the code. @param: A L{pyflakes.messages.Message}. """ self._stdout.write(str(message)) self._stdout.write('\n')
PyCQA/pyflakes
pyflakes/reporter.py
Reporter.flake
python
def flake(self, message): self._stdout.write(str(message)) self._stdout.write('\n')
pyflakes found something wrong with the code. @param: A L{pyflakes.messages.Message}.
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/reporter.py#L68-L75
null
class Reporter(object): """ Formats the results of pyflakes checks to users. """ def __init__(self, warningStream, errorStream): """ Construct a L{Reporter}. @param warningStream: A file-like object where warnings will be written to. The stream's C{write} method must accept unicode. C{sys.stdout} is a good value. @param errorStream: A file-like object where error output will be written to. The stream's C{write} method must accept unicode. C{sys.stderr} is a good value. """ self._stdout = warningStream self._stderr = errorStream def unexpectedError(self, filename, msg): """ An unexpected error occurred trying to process C{filename}. @param filename: The path to a file that we could not process. @ptype filename: C{unicode} @param msg: A message explaining the problem. @ptype msg: C{unicode} """ self._stderr.write("%s: %s\n" % (filename, msg)) def syntaxError(self, filename, msg, lineno, offset, text): """ There was a syntax error in C{filename}. @param filename: The path to the file with the syntax error. @ptype filename: C{unicode} @param msg: An explanation of the syntax error. @ptype msg: C{unicode} @param lineno: The line number where the syntax error occurred. @ptype lineno: C{int} @param offset: The column on which the syntax error occurred, or None. @ptype offset: C{int} @param text: The source code containing the syntax error. @ptype text: C{unicode} """ line = text.splitlines()[-1] if offset is not None: if sys.version_info < (3, 8): offset = offset - (len(text) - len(line)) + 1 self._stderr.write('%s:%d:%d: %s\n' % (filename, lineno, offset, msg)) else: self._stderr.write('%s:%d: %s\n' % (filename, lineno, msg)) self._stderr.write(line) self._stderr.write('\n') if offset is not None: self._stderr.write(re.sub(r'\S', ' ', line[:offset - 1]) + "^\n")
PyCQA/pyflakes
pyflakes/checker.py
counter
python
def counter(items): results = {} for item in items: results[item] = results.get(item, 0) + 1 return results
Simplest required implementation of collections.Counter. Required as 2.6 does not have Counter in collections.
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/checker.py#L103-L111
null
""" Main module. Implement the central Checker class. Also, it models the Bindings and Scopes. """ import __future__ import ast import bisect import collections import doctest import functools import os import re import sys import tokenize from pyflakes import messages PY2 = sys.version_info < (3, 0) PY35_PLUS = sys.version_info >= (3, 5) # Python 3.5 and above PY36_PLUS = sys.version_info >= (3, 6) # Python 3.6 and above PY38_PLUS = sys.version_info >= (3, 8) try: sys.pypy_version_info PYPY = True except AttributeError: PYPY = False builtin_vars = dir(__import__('__builtin__' if PY2 else 'builtins')) if PY2: tokenize_tokenize = tokenize.generate_tokens else: tokenize_tokenize = tokenize.tokenize if PY2: def getNodeType(node_class): # workaround str.upper() which is locale-dependent return str(unicode(node_class.__name__).upper()) def get_raise_argument(node): return node.type else: def getNodeType(node_class): return node_class.__name__.upper() def get_raise_argument(node): return node.exc # Silence `pyflakes` from reporting `undefined name 'unicode'` in Python 3. unicode = str # Python >= 3.3 uses ast.Try instead of (ast.TryExcept + ast.TryFinally) if PY2: def getAlternatives(n): if isinstance(n, (ast.If, ast.TryFinally)): return [n.body] if isinstance(n, ast.TryExcept): return [n.body + n.orelse] + [[hdl] for hdl in n.handlers] else: def getAlternatives(n): if isinstance(n, ast.If): return [n.body] if isinstance(n, ast.Try): return [n.body + n.orelse] + [[hdl] for hdl in n.handlers] if PY35_PLUS: FOR_TYPES = (ast.For, ast.AsyncFor) LOOP_TYPES = (ast.While, ast.For, ast.AsyncFor) else: FOR_TYPES = (ast.For,) LOOP_TYPES = (ast.While, ast.For) # https://github.com/python/typed_ast/blob/55420396/ast27/Parser/tokenizer.c#L102-L104 TYPE_COMMENT_RE = re.compile(r'^#\s*type:\s*') # https://github.com/python/typed_ast/blob/55420396/ast27/Parser/tokenizer.c#L1400 TYPE_IGNORE_RE = re.compile(TYPE_COMMENT_RE.pattern + r'ignore\s*(#|$)') # https://github.com/python/typed_ast/blob/55420396/ast27/Grammar/Grammar#L147 TYPE_FUNC_RE = re.compile(r'^(\(.*?\))\s*->\s*(.*)$') class _FieldsOrder(dict): """Fix order of AST node fields.""" def _get_fields(self, node_class): # handle iter before target, and generators before element fields = node_class._fields if 'iter' in fields: key_first = 'iter'.find elif 'generators' in fields: key_first = 'generators'.find else: key_first = 'value'.find return tuple(sorted(fields, key=key_first, reverse=True)) def __missing__(self, node_class): self[node_class] = fields = self._get_fields(node_class) return fields def iter_child_nodes(node, omit=None, _fields_order=_FieldsOrder()): """ Yield all direct child nodes of *node*, that is, all fields that are nodes and all items of fields that are lists of nodes. :param node: AST node to be iterated upon :param omit: String or tuple of strings denoting the attributes of the node to be omitted from further parsing :param _fields_order: Order of AST node fields """ for name in _fields_order[node.__class__]: if omit and name in omit: continue field = getattr(node, name, None) if isinstance(field, ast.AST): yield field elif isinstance(field, list): for item in field: yield item def convert_to_value(item): if isinstance(item, ast.Str): return item.s elif hasattr(ast, 'Bytes') and isinstance(item, ast.Bytes): return item.s elif isinstance(item, ast.Tuple): return tuple(convert_to_value(i) for i in item.elts) elif isinstance(item, ast.Num): return item.n elif isinstance(item, ast.Name): result = VariableKey(item=item) constants_lookup = { 'True': True, 'False': False, 'None': None, } return constants_lookup.get( result.name, result, ) elif (not PY2) and isinstance(item, ast.NameConstant): # None, True, False are nameconstants in python3, but names in 2 return item.value else: return UnhandledKeyType() def is_notimplemented_name_node(node): return isinstance(node, ast.Name) and getNodeName(node) == 'NotImplemented' class Binding(object): """ Represents the binding of a value to a name. The checker uses this to keep track of which names have been bound and which names have not. See L{Assignment} for a special type of binding that is checked with stricter rules. @ivar used: pair of (L{Scope}, node) indicating the scope and the node that this binding was last used. """ def __init__(self, name, source): self.name = name self.source = source self.used = False def __str__(self): return self.name def __repr__(self): return '<%s object %r from line %r at 0x%x>' % (self.__class__.__name__, self.name, self.source.lineno, id(self)) def redefines(self, other): return isinstance(other, Definition) and self.name == other.name class Definition(Binding): """ A binding that defines a function or a class. """ class Builtin(Definition): """A definition created for all Python builtins.""" def __init__(self, name): super(Builtin, self).__init__(name, None) def __repr__(self): return '<%s object %r at 0x%x>' % (self.__class__.__name__, self.name, id(self)) class UnhandledKeyType(object): """ A dictionary key of a type that we cannot or do not check for duplicates. """ class VariableKey(object): """ A dictionary key which is a variable. @ivar item: The variable AST object. """ def __init__(self, item): self.name = item.id def __eq__(self, compare): return ( compare.__class__ == self.__class__ and compare.name == self.name ) def __hash__(self): return hash(self.name) class Importation(Definition): """ A binding created by an import statement. @ivar fullName: The complete name given to the import statement, possibly including multiple dotted components. @type fullName: C{str} """ def __init__(self, name, source, full_name=None): self.fullName = full_name or name self.redefined = [] super(Importation, self).__init__(name, source) def redefines(self, other): if isinstance(other, SubmoduleImportation): # See note in SubmoduleImportation about RedefinedWhileUnused return self.fullName == other.fullName return isinstance(other, Definition) and self.name == other.name def _has_alias(self): """Return whether importation needs an as clause.""" return not self.fullName.split('.')[-1] == self.name @property def source_statement(self): """Generate a source statement equivalent to the import.""" if self._has_alias(): return 'import %s as %s' % (self.fullName, self.name) else: return 'import %s' % self.fullName def __str__(self): """Return import full name with alias.""" if self._has_alias(): return self.fullName + ' as ' + self.name else: return self.fullName class SubmoduleImportation(Importation): """ A binding created by a submodule import statement. A submodule import is a special case where the root module is implicitly imported, without an 'as' clause, and the submodule is also imported. Python does not restrict which attributes of the root module may be used. This class is only used when the submodule import is without an 'as' clause. pyflakes handles this case by registering the root module name in the scope, allowing any attribute of the root module to be accessed. RedefinedWhileUnused is suppressed in `redefines` unless the submodule name is also the same, to avoid false positives. """ def __init__(self, name, source): # A dot should only appear in the name when it is a submodule import assert '.' in name and (not source or isinstance(source, ast.Import)) package_name = name.split('.')[0] super(SubmoduleImportation, self).__init__(package_name, source) self.fullName = name def redefines(self, other): if isinstance(other, Importation): return self.fullName == other.fullName return super(SubmoduleImportation, self).redefines(other) def __str__(self): return self.fullName @property def source_statement(self): return 'import ' + self.fullName class ImportationFrom(Importation): def __init__(self, name, source, module, real_name=None): self.module = module self.real_name = real_name or name if module.endswith('.'): full_name = module + self.real_name else: full_name = module + '.' + self.real_name super(ImportationFrom, self).__init__(name, source, full_name) def __str__(self): """Return import full name with alias.""" if self.real_name != self.name: return self.fullName + ' as ' + self.name else: return self.fullName @property def source_statement(self): if self.real_name != self.name: return 'from %s import %s as %s' % (self.module, self.real_name, self.name) else: return 'from %s import %s' % (self.module, self.name) class StarImportation(Importation): """A binding created by a 'from x import *' statement.""" def __init__(self, name, source): super(StarImportation, self).__init__('*', source) # Each star importation needs a unique name, and # may not be the module name otherwise it will be deemed imported self.name = name + '.*' self.fullName = name @property def source_statement(self): return 'from ' + self.fullName + ' import *' def __str__(self): # When the module ends with a ., avoid the ambiguous '..*' if self.fullName.endswith('.'): return self.source_statement else: return self.name class FutureImportation(ImportationFrom): """ A binding created by a from `__future__` import statement. `__future__` imports are implicitly used. """ def __init__(self, name, source, scope): super(FutureImportation, self).__init__(name, source, '__future__') self.used = (scope, source) class Argument(Binding): """ Represents binding a name as an argument. """ class Assignment(Binding): """ Represents binding a name with an explicit assignment. The checker will raise warnings for any Assignment that isn't used. Also, the checker does not consider assignments in tuple/list unpacking to be Assignments, rather it treats them as simple Bindings. """ class FunctionDefinition(Definition): pass class ClassDefinition(Definition): pass class ExportBinding(Binding): """ A binding created by an C{__all__} assignment. If the names in the list can be determined statically, they will be treated as names for export and additional checking applied to them. The only recognized C{__all__} assignment via list concatenation is in the following format: __all__ = ['a'] + ['b'] + ['c'] Names which are imported and not otherwise used but appear in the value of C{__all__} will not have an unused import warning reported for them. """ def __init__(self, name, source, scope): if '__all__' in scope and isinstance(source, ast.AugAssign): self.names = list(scope['__all__'].names) else: self.names = [] def _add_to_names(container): for node in container.elts: if isinstance(node, ast.Str): self.names.append(node.s) if isinstance(source.value, (ast.List, ast.Tuple)): _add_to_names(source.value) # If concatenating lists elif isinstance(source.value, ast.BinOp): currentValue = source.value while isinstance(currentValue.right, ast.List): left = currentValue.left right = currentValue.right _add_to_names(right) # If more lists are being added if isinstance(left, ast.BinOp): currentValue = left # If just two lists are being added elif isinstance(left, ast.List): _add_to_names(left) # All lists accounted for - done break # If not list concatenation else: break super(ExportBinding, self).__init__(name, source) class Scope(dict): importStarred = False # set to True when import * is found def __repr__(self): scope_cls = self.__class__.__name__ return '<%s at 0x%x %s>' % (scope_cls, id(self), dict.__repr__(self)) class ClassScope(Scope): pass class FunctionScope(Scope): """ I represent a name scope for a function. @ivar globals: Names declared 'global' in this function. """ usesLocals = False alwaysUsed = {'__tracebackhide__', '__traceback_info__', '__traceback_supplement__'} def __init__(self): super(FunctionScope, self).__init__() # Simplify: manage the special locals as globals self.globals = self.alwaysUsed.copy() self.returnValue = None # First non-empty return self.isGenerator = False # Detect a generator def unusedAssignments(self): """ Return a generator for the assignments which have not been used. """ for name, binding in self.items(): if (not binding.used and name != '_' and # see issue #202 name not in self.globals and not self.usesLocals and isinstance(binding, Assignment)): yield name, binding class GeneratorScope(Scope): pass class ModuleScope(Scope): """Scope for a module.""" _futures_allowed = True _annotations_future_enabled = False class DoctestScope(ModuleScope): """Scope for a doctest.""" class DummyNode(object): """Used in place of an `ast.AST` to set error message positions""" def __init__(self, lineno, col_offset): self.lineno = lineno self.col_offset = col_offset # Globally defined names which are not attributes of the builtins module, or # are only present on some platforms. _MAGIC_GLOBALS = ['__file__', '__builtins__', 'WindowsError'] # module scope annotation will store in `__annotations__`, see also PEP 526. if PY36_PLUS: _MAGIC_GLOBALS.append('__annotations__') def getNodeName(node): # Returns node.id, or node.name, or None if hasattr(node, 'id'): # One of the many nodes with an id return node.id if hasattr(node, 'name'): # an ExceptHandler node return node.name def is_typing_overload(value, scope_stack): def name_is_typing_overload(name): # type: (str) -> bool for scope in reversed(scope_stack): if name in scope: return ( isinstance(scope[name], ImportationFrom) and scope[name].fullName == 'typing.overload' ) else: return False def is_typing_overload_decorator(node): return ( ( isinstance(node, ast.Name) and name_is_typing_overload(node.id) ) or ( isinstance(node, ast.Attribute) and isinstance(node.value, ast.Name) and node.value.id == 'typing' and node.attr == 'overload' ) ) return ( isinstance(value.source, ast.FunctionDef) and any( is_typing_overload_decorator(dec) for dec in value.source.decorator_list ) ) def make_tokens(code): # PY3: tokenize.tokenize requires readline of bytes if not isinstance(code, bytes): code = code.encode('UTF-8') lines = iter(code.splitlines(True)) # next(lines, b'') is to prevent an error in pypy3 return tuple(tokenize_tokenize(lambda: next(lines, b''))) class _TypeableVisitor(ast.NodeVisitor): """Collect the line number and nodes which are deemed typeable by PEP 484 https://www.python.org/dev/peps/pep-0484/#type-comments """ def __init__(self): self.typeable_lines = [] # type: List[int] self.typeable_nodes = {} # type: Dict[int, ast.AST] def _typeable(self, node): # if there is more than one typeable thing on a line last one wins self.typeable_lines.append(node.lineno) self.typeable_nodes[node.lineno] = node self.generic_visit(node) visit_Assign = visit_For = visit_FunctionDef = visit_With = _typeable visit_AsyncFor = visit_AsyncFunctionDef = visit_AsyncWith = _typeable def _collect_type_comments(tree, tokens): visitor = _TypeableVisitor() visitor.visit(tree) type_comments = collections.defaultdict(list) for tp, text, start, _, _ in tokens: if ( tp != tokenize.COMMENT or # skip non comments not TYPE_COMMENT_RE.match(text) or # skip non-type comments TYPE_IGNORE_RE.match(text) # skip ignores ): continue # search for the typeable node at or before the line number of the # type comment. # if the bisection insertion point is before any nodes this is an # invalid type comment which is ignored. lineno, _ = start idx = bisect.bisect_right(visitor.typeable_lines, lineno) if idx == 0: continue node = visitor.typeable_nodes[visitor.typeable_lines[idx - 1]] type_comments[node].append((start, text)) return type_comments class Checker(object): """ I check the cleanliness and sanity of Python code. @ivar _deferredFunctions: Tracking list used by L{deferFunction}. Elements of the list are two-tuples. The first element is the callable passed to L{deferFunction}. The second element is a copy of the scope stack at the time L{deferFunction} was called. @ivar _deferredAssignments: Similar to C{_deferredFunctions}, but for callables which are deferred assignment checks. """ _ast_node_scope = { ast.Module: ModuleScope, ast.ClassDef: ClassScope, ast.FunctionDef: FunctionScope, ast.Lambda: FunctionScope, ast.ListComp: GeneratorScope, ast.SetComp: GeneratorScope, ast.GeneratorExp: GeneratorScope, ast.DictComp: GeneratorScope, } if PY35_PLUS: _ast_node_scope[ast.AsyncFunctionDef] = FunctionScope, nodeDepth = 0 offset = None traceTree = False builtIns = set(builtin_vars).union(_MAGIC_GLOBALS) _customBuiltIns = os.environ.get('PYFLAKES_BUILTINS') if _customBuiltIns: builtIns.update(_customBuiltIns.split(',')) del _customBuiltIns # TODO: file_tokens= is required to perform checks on type comments, # eventually make this a required positional argument. For now it # is defaulted to `()` for api compatibility. def __init__(self, tree, filename='(none)', builtins=None, withDoctest='PYFLAKES_DOCTEST' in os.environ, file_tokens=()): self._nodeHandlers = {} self._deferredFunctions = [] self._deferredAssignments = [] self.deadScopes = [] self.messages = [] self.filename = filename if builtins: self.builtIns = self.builtIns.union(builtins) self.withDoctest = withDoctest try: self.scopeStack = [Checker._ast_node_scope[type(tree)]()] except KeyError: raise RuntimeError('No scope implemented for the node %r' % tree) self.exceptHandlers = [()] self.root = tree self._type_comments = _collect_type_comments(tree, file_tokens) for builtin in self.builtIns: self.addBinding(None, Builtin(builtin)) self.handleChildren(tree) self.runDeferred(self._deferredFunctions) # Set _deferredFunctions to None so that deferFunction will fail # noisily if called after we've run through the deferred functions. self._deferredFunctions = None self.runDeferred(self._deferredAssignments) # Set _deferredAssignments to None so that deferAssignment will fail # noisily if called after we've run through the deferred assignments. self._deferredAssignments = None del self.scopeStack[1:] self.popScope() self.checkDeadScopes() def deferFunction(self, callable): """ Schedule a function handler to be called just before completion. This is used for handling function bodies, which must be deferred because code later in the file might modify the global scope. When `callable` is called, the scope at the time this is called will be restored, however it will contain any new bindings added to it. """ self._deferredFunctions.append((callable, self.scopeStack[:], self.offset)) def deferAssignment(self, callable): """ Schedule an assignment handler to be called just after deferred function handlers. """ self._deferredAssignments.append((callable, self.scopeStack[:], self.offset)) def runDeferred(self, deferred): """ Run the callables in C{deferred} using their associated scope stack. """ for handler, scope, offset in deferred: self.scopeStack = scope self.offset = offset handler() def _in_doctest(self): return (len(self.scopeStack) >= 2 and isinstance(self.scopeStack[1], DoctestScope)) @property def futuresAllowed(self): if not all(isinstance(scope, ModuleScope) for scope in self.scopeStack): return False return self.scope._futures_allowed @futuresAllowed.setter def futuresAllowed(self, value): assert value is False if isinstance(self.scope, ModuleScope): self.scope._futures_allowed = False @property def annotationsFutureEnabled(self): scope = self.scopeStack[0] if not isinstance(scope, ModuleScope): return False return scope._annotations_future_enabled @annotationsFutureEnabled.setter def annotationsFutureEnabled(self, value): assert value is True assert isinstance(self.scope, ModuleScope) self.scope._annotations_future_enabled = True @property def scope(self): return self.scopeStack[-1] def popScope(self): self.deadScopes.append(self.scopeStack.pop()) def checkDeadScopes(self): """ Look at scopes which have been fully examined and report names in them which were imported but unused. """ for scope in self.deadScopes: # imports in classes are public members if isinstance(scope, ClassScope): continue all_binding = scope.get('__all__') if all_binding and not isinstance(all_binding, ExportBinding): all_binding = None if all_binding: all_names = set(all_binding.names) undefined = all_names.difference(scope) else: all_names = undefined = [] if undefined: if not scope.importStarred and \ os.path.basename(self.filename) != '__init__.py': # Look for possible mistakes in the export list for name in undefined: self.report(messages.UndefinedExport, scope['__all__'].source, name) # mark all import '*' as used by the undefined in __all__ if scope.importStarred: from_list = [] for binding in scope.values(): if isinstance(binding, StarImportation): binding.used = all_binding from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) for name in undefined: self.report(messages.ImportStarUsage, scope['__all__'].source, name, from_list) # Look for imported names that aren't used. for value in scope.values(): if isinstance(value, Importation): used = value.used or value.name in all_names if not used: messg = messages.UnusedImport self.report(messg, value.source, str(value)) for node in value.redefined: if isinstance(self.getParent(node), FOR_TYPES): messg = messages.ImportShadowedByLoopVar elif used: continue else: messg = messages.RedefinedWhileUnused self.report(messg, node, value.name, value.source) def pushScope(self, scopeClass=FunctionScope): self.scopeStack.append(scopeClass()) def report(self, messageClass, *args, **kwargs): self.messages.append(messageClass(self.filename, *args, **kwargs)) def getParent(self, node): # Lookup the first parent which is not Tuple, List or Starred while True: node = node.parent if not hasattr(node, 'elts') and not hasattr(node, 'ctx'): return node def getCommonAncestor(self, lnode, rnode, stop): if stop in (lnode, rnode) or not (hasattr(lnode, 'parent') and hasattr(rnode, 'parent')): return None if lnode is rnode: return lnode if (lnode.depth > rnode.depth): return self.getCommonAncestor(lnode.parent, rnode, stop) if (lnode.depth < rnode.depth): return self.getCommonAncestor(lnode, rnode.parent, stop) return self.getCommonAncestor(lnode.parent, rnode.parent, stop) def descendantOf(self, node, ancestors, stop): for a in ancestors: if self.getCommonAncestor(node, a, stop): return True return False def _getAncestor(self, node, ancestor_type): parent = node while True: if parent is self.root: return None parent = self.getParent(parent) if isinstance(parent, ancestor_type): return parent def getScopeNode(self, node): return self._getAncestor(node, tuple(Checker._ast_node_scope.keys())) def differentForks(self, lnode, rnode): """True, if lnode and rnode are located on different forks of IF/TRY""" ancestor = self.getCommonAncestor(lnode, rnode, self.root) parts = getAlternatives(ancestor) if parts: for items in parts: if self.descendantOf(lnode, items, ancestor) ^ \ self.descendantOf(rnode, items, ancestor): return True return False def addBinding(self, node, value): """ Called when a binding is altered. - `node` is the statement responsible for the change - `value` is the new value, a Binding instance """ # assert value.source in (node, node.parent): for scope in self.scopeStack[::-1]: if value.name in scope: break existing = scope.get(value.name) if (existing and not isinstance(existing, Builtin) and not self.differentForks(node, existing.source)): parent_stmt = self.getParent(value.source) if isinstance(existing, Importation) and isinstance(parent_stmt, FOR_TYPES): self.report(messages.ImportShadowedByLoopVar, node, value.name, existing.source) elif scope is self.scope: if (isinstance(parent_stmt, ast.comprehension) and not isinstance(self.getParent(existing.source), (FOR_TYPES, ast.comprehension))): self.report(messages.RedefinedInListComp, node, value.name, existing.source) elif not existing.used and value.redefines(existing): if value.name != '_' or isinstance(existing, Importation): if not is_typing_overload(existing, self.scopeStack): self.report(messages.RedefinedWhileUnused, node, value.name, existing.source) elif isinstance(existing, Importation) and value.redefines(existing): existing.redefined.append(node) if value.name in self.scope: # then assume the rebound name is used as a global or within a loop value.used = self.scope[value.name].used self.scope[value.name] = value def getNodeHandler(self, node_class): try: return self._nodeHandlers[node_class] except KeyError: nodeType = getNodeType(node_class) self._nodeHandlers[node_class] = handler = getattr(self, nodeType) return handler def handleNodeLoad(self, node): name = getNodeName(node) if not name: return in_generators = None importStarred = None # try enclosing function scopes and global scope for scope in self.scopeStack[-1::-1]: if isinstance(scope, ClassScope): if not PY2 and name == '__class__': return elif in_generators is False: # only generators used in a class scope can access the # names of the class. this is skipped during the first # iteration continue if (name == 'print' and isinstance(scope.get(name, None), Builtin)): parent = self.getParent(node) if (isinstance(parent, ast.BinOp) and isinstance(parent.op, ast.RShift)): self.report(messages.InvalidPrintSyntax, node) try: scope[name].used = (self.scope, node) # if the name of SubImportation is same as # alias of other Importation and the alias # is used, SubImportation also should be marked as used. n = scope[name] if isinstance(n, Importation) and n._has_alias(): try: scope[n.fullName].used = (self.scope, node) except KeyError: pass except KeyError: pass else: return importStarred = importStarred or scope.importStarred if in_generators is not False: in_generators = isinstance(scope, GeneratorScope) if importStarred: from_list = [] for scope in self.scopeStack[-1::-1]: for binding in scope.values(): if isinstance(binding, StarImportation): # mark '*' imports as used for each scope binding.used = (self.scope, node) from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) self.report(messages.ImportStarUsage, node, name, from_list) return if name == '__path__' and os.path.basename(self.filename) == '__init__.py': # the special name __path__ is valid only in packages return if name == '__module__' and isinstance(self.scope, ClassScope): return # protected with a NameError handler? if 'NameError' not in self.exceptHandlers[-1]: self.report(messages.UndefinedName, node, name) def handleNodeStore(self, node): name = getNodeName(node) if not name: return # if the name hasn't already been defined in the current scope if isinstance(self.scope, FunctionScope) and name not in self.scope: # for each function or module scope above us for scope in self.scopeStack[:-1]: if not isinstance(scope, (FunctionScope, ModuleScope)): continue # if the name was defined in that scope, and the name has # been accessed already in the current scope, and hasn't # been declared global used = name in scope and scope[name].used if used and used[0] is self.scope and name not in self.scope.globals: # then it's probably a mistake self.report(messages.UndefinedLocal, scope[name].used[1], name, scope[name].source) break parent_stmt = self.getParent(node) if isinstance(parent_stmt, (FOR_TYPES, ast.comprehension)) or ( parent_stmt != node.parent and not self.isLiteralTupleUnpacking(parent_stmt)): binding = Binding(name, node) elif name == '__all__' and isinstance(self.scope, ModuleScope): binding = ExportBinding(name, node.parent, self.scope) elif isinstance(getattr(node, 'ctx', None), ast.Param): binding = Argument(name, self.getScopeNode(node)) else: binding = Assignment(name, node) self.addBinding(node, binding) def handleNodeDelete(self, node): def on_conditional_branch(): """ Return `True` if node is part of a conditional body. """ current = getattr(node, 'parent', None) while current: if isinstance(current, (ast.If, ast.While, ast.IfExp)): return True current = getattr(current, 'parent', None) return False name = getNodeName(node) if not name: return if on_conditional_branch(): # We cannot predict if this conditional branch is going to # be executed. return if isinstance(self.scope, FunctionScope) and name in self.scope.globals: self.scope.globals.remove(name) else: try: del self.scope[name] except KeyError: self.report(messages.UndefinedName, node, name) def _handle_type_comments(self, node): for (lineno, col_offset), comment in self._type_comments.get(node, ()): comment = comment.split(':', 1)[1].strip() func_match = TYPE_FUNC_RE.match(comment) if func_match: parts = ( func_match.group(1).replace('*', ''), func_match.group(2).strip(), ) else: parts = (comment,) for part in parts: if PY2: part = part.replace('...', 'Ellipsis') self.deferFunction(functools.partial( self.handleStringAnnotation, part, DummyNode(lineno, col_offset), lineno, col_offset, messages.CommentAnnotationSyntaxError, )) def handleChildren(self, tree, omit=None): self._handle_type_comments(tree) for node in iter_child_nodes(tree, omit=omit): self.handleNode(node, tree) def isLiteralTupleUnpacking(self, node): if isinstance(node, ast.Assign): for child in node.targets + [node.value]: if not hasattr(child, 'elts'): return False return True def isDocstring(self, node): """ Determine if the given node is a docstring, as long as it is at the correct place in the node tree. """ return isinstance(node, ast.Str) or (isinstance(node, ast.Expr) and isinstance(node.value, ast.Str)) def getDocstring(self, node): if isinstance(node, ast.Expr): node = node.value if not isinstance(node, ast.Str): return (None, None) if PYPY or PY38_PLUS: doctest_lineno = node.lineno - 1 else: # Computed incorrectly if the docstring has backslash doctest_lineno = node.lineno - node.s.count('\n') - 1 return (node.s, doctest_lineno) def handleNode(self, node, parent): if node is None: return if self.offset and getattr(node, 'lineno', None) is not None: node.lineno += self.offset[0] node.col_offset += self.offset[1] if self.traceTree: print(' ' * self.nodeDepth + node.__class__.__name__) if self.futuresAllowed and not (isinstance(node, ast.ImportFrom) or self.isDocstring(node)): self.futuresAllowed = False self.nodeDepth += 1 node.depth = self.nodeDepth node.parent = parent try: handler = self.getNodeHandler(node.__class__) handler(node) finally: self.nodeDepth -= 1 if self.traceTree: print(' ' * self.nodeDepth + 'end ' + node.__class__.__name__) _getDoctestExamples = doctest.DocTestParser().get_examples def handleDoctests(self, node): try: if hasattr(node, 'docstring'): docstring = node.docstring # This is just a reasonable guess. In Python 3.7, docstrings no # longer have line numbers associated with them. This will be # incorrect if there are empty lines between the beginning # of the function and the docstring. node_lineno = node.lineno if hasattr(node, 'args'): node_lineno = max([node_lineno] + [arg.lineno for arg in node.args.args]) else: (docstring, node_lineno) = self.getDocstring(node.body[0]) examples = docstring and self._getDoctestExamples(docstring) except (ValueError, IndexError): # e.g. line 6 of the docstring for <string> has inconsistent # leading whitespace: ... return if not examples: return # Place doctest in module scope saved_stack = self.scopeStack self.scopeStack = [self.scopeStack[0]] node_offset = self.offset or (0, 0) self.pushScope(DoctestScope) self.addBinding(None, Builtin('_')) for example in examples: try: tree = ast.parse(example.source, "<doctest>") except SyntaxError: e = sys.exc_info()[1] if PYPY: e.offset += 1 position = (node_lineno + example.lineno + e.lineno, example.indent + 4 + (e.offset or 0)) self.report(messages.DoctestSyntaxError, node, position) else: self.offset = (node_offset[0] + node_lineno + example.lineno, node_offset[1] + example.indent + 4) self.handleChildren(tree) self.offset = node_offset self.popScope() self.scopeStack = saved_stack def handleStringAnnotation(self, s, node, ref_lineno, ref_col_offset, err): try: tree = ast.parse(s) except SyntaxError: self.report(err, node, s) return body = tree.body if len(body) != 1 or not isinstance(body[0], ast.Expr): self.report(err, node, s) return parsed_annotation = tree.body[0].value for descendant in ast.walk(parsed_annotation): if ( 'lineno' in descendant._attributes and 'col_offset' in descendant._attributes ): descendant.lineno = ref_lineno descendant.col_offset = ref_col_offset self.handleNode(parsed_annotation, node) def handleAnnotation(self, annotation, node): if isinstance(annotation, ast.Str): # Defer handling forward annotation. self.deferFunction(functools.partial( self.handleStringAnnotation, annotation.s, node, annotation.lineno, annotation.col_offset, messages.ForwardAnnotationSyntaxError, )) elif self.annotationsFutureEnabled: self.deferFunction(lambda: self.handleNode(annotation, node)) else: self.handleNode(annotation, node) def ignore(self, node): pass # "stmt" type nodes DELETE = PRINT = FOR = ASYNCFOR = WHILE = IF = WITH = WITHITEM = \ ASYNCWITH = ASYNCWITHITEM = TRYFINALLY = EXEC = \ EXPR = ASSIGN = handleChildren PASS = ignore # "expr" type nodes BOOLOP = BINOP = UNARYOP = IFEXP = SET = \ CALL = REPR = ATTRIBUTE = SUBSCRIPT = \ STARRED = NAMECONSTANT = handleChildren NUM = STR = BYTES = ELLIPSIS = CONSTANT = ignore # "slice" type nodes SLICE = EXTSLICE = INDEX = handleChildren # expression contexts are node instances too, though being constants LOAD = STORE = DEL = AUGLOAD = AUGSTORE = PARAM = ignore # same for operators AND = OR = ADD = SUB = MULT = DIV = MOD = POW = LSHIFT = RSHIFT = \ BITOR = BITXOR = BITAND = FLOORDIV = INVERT = NOT = UADD = USUB = \ EQ = NOTEQ = LT = LTE = GT = GTE = IS = ISNOT = IN = NOTIN = \ MATMULT = ignore def RAISE(self, node): self.handleChildren(node) arg = get_raise_argument(node) if isinstance(arg, ast.Call): if is_notimplemented_name_node(arg.func): # Handle "raise NotImplemented(...)" self.report(messages.RaiseNotImplemented, node) elif is_notimplemented_name_node(arg): # Handle "raise NotImplemented" self.report(messages.RaiseNotImplemented, node) # additional node types COMPREHENSION = KEYWORD = FORMATTEDVALUE = JOINEDSTR = handleChildren def DICT(self, node): # Complain if there are duplicate keys with different values # If they have the same value it's not going to cause potentially # unexpected behaviour so we'll not complain. keys = [ convert_to_value(key) for key in node.keys ] key_counts = counter(keys) duplicate_keys = [ key for key, count in key_counts.items() if count > 1 ] for key in duplicate_keys: key_indices = [i for i, i_key in enumerate(keys) if i_key == key] values = counter( convert_to_value(node.values[index]) for index in key_indices ) if any(count == 1 for value, count in values.items()): for key_index in key_indices: key_node = node.keys[key_index] if isinstance(key, VariableKey): self.report(messages.MultiValueRepeatedKeyVariable, key_node, key.name) else: self.report( messages.MultiValueRepeatedKeyLiteral, key_node, key, ) self.handleChildren(node) def ASSERT(self, node): if isinstance(node.test, ast.Tuple) and node.test.elts != []: self.report(messages.AssertTuple, node) self.handleChildren(node) def GLOBAL(self, node): """ Keep track of globals declarations. """ global_scope_index = 1 if self._in_doctest() else 0 global_scope = self.scopeStack[global_scope_index] # Ignore 'global' statement in global scope. if self.scope is not global_scope: # One 'global' statement can bind multiple (comma-delimited) names. for node_name in node.names: node_value = Assignment(node_name, node) # Remove UndefinedName messages already reported for this name. # TODO: if the global is not used in this scope, it does not # become a globally defined name. See test_unused_global. self.messages = [ m for m in self.messages if not isinstance(m, messages.UndefinedName) or m.message_args[0] != node_name] # Bind name to global scope if it doesn't exist already. global_scope.setdefault(node_name, node_value) # Bind name to non-global scopes, but as already "used". node_value.used = (global_scope, node) for scope in self.scopeStack[global_scope_index + 1:]: scope[node_name] = node_value NONLOCAL = GLOBAL def GENERATOREXP(self, node): self.pushScope(GeneratorScope) self.handleChildren(node) self.popScope() LISTCOMP = handleChildren if PY2 else GENERATOREXP DICTCOMP = SETCOMP = GENERATOREXP def NAME(self, node): """ Handle occurrence of Name (which can be a load/store/delete access.) """ # Locate the name in locals / function / globals scopes. if isinstance(node.ctx, (ast.Load, ast.AugLoad)): self.handleNodeLoad(node) if (node.id == 'locals' and isinstance(self.scope, FunctionScope) and isinstance(node.parent, ast.Call)): # we are doing locals() call in current scope self.scope.usesLocals = True elif isinstance(node.ctx, (ast.Store, ast.AugStore, ast.Param)): self.handleNodeStore(node) elif isinstance(node.ctx, ast.Del): self.handleNodeDelete(node) else: # Unknown context raise RuntimeError("Got impossible expression context: %r" % (node.ctx,)) def CONTINUE(self, node): # Walk the tree up until we see a loop (OK), a function or class # definition (not OK), for 'continue', a finally block (not OK), or # the top module scope (not OK) n = node while hasattr(n, 'parent'): n, n_child = n.parent, n if isinstance(n, LOOP_TYPES): # Doesn't apply unless it's in the loop itself if n_child not in n.orelse: return if isinstance(n, (ast.FunctionDef, ast.ClassDef)): break # Handle Try/TryFinally difference in Python < and >= 3.3 if hasattr(n, 'finalbody') and isinstance(node, ast.Continue): if n_child in n.finalbody: self.report(messages.ContinueInFinally, node) return if isinstance(node, ast.Continue): self.report(messages.ContinueOutsideLoop, node) else: # ast.Break self.report(messages.BreakOutsideLoop, node) BREAK = CONTINUE def RETURN(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.ReturnOutsideFunction, node) return if ( node.value and hasattr(self.scope, 'returnValue') and not self.scope.returnValue ): self.scope.returnValue = node.value self.handleNode(node.value, node) def YIELD(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.YieldOutsideFunction, node) return self.scope.isGenerator = True self.handleNode(node.value, node) AWAIT = YIELDFROM = YIELD def FUNCTIONDEF(self, node): for deco in node.decorator_list: self.handleNode(deco, node) self.LAMBDA(node) self.addBinding(node, FunctionDefinition(node.name, node)) # doctest does not process doctest within a doctest, # or in nested functions. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) ASYNCFUNCTIONDEF = FUNCTIONDEF def LAMBDA(self, node): args = [] annotations = [] if PY2: def addArgs(arglist): for arg in arglist: if isinstance(arg, ast.Tuple): addArgs(arg.elts) else: args.append(arg.id) addArgs(node.args.args) defaults = node.args.defaults else: for arg in node.args.args + node.args.kwonlyargs: args.append(arg.arg) annotations.append(arg.annotation) defaults = node.args.defaults + node.args.kw_defaults # Only for Python3 FunctionDefs is_py3_func = hasattr(node, 'returns') for arg_name in ('vararg', 'kwarg'): wildcard = getattr(node.args, arg_name) if not wildcard: continue args.append(wildcard if PY2 else wildcard.arg) if is_py3_func: if PY2: # Python 2.7 argannotation = arg_name + 'annotation' annotations.append(getattr(node.args, argannotation)) else: # Python >= 3.4 annotations.append(wildcard.annotation) if is_py3_func: annotations.append(node.returns) if len(set(args)) < len(args): for (idx, arg) in enumerate(args): if arg in args[:idx]: self.report(messages.DuplicateArgument, node, arg) for annotation in annotations: self.handleAnnotation(annotation, node) for default in defaults: self.handleNode(default, node) def runFunction(): self.pushScope() self.handleChildren(node, omit='decorator_list') def checkUnusedAssignments(): """ Check to see if any assignments have not been used. """ for name, binding in self.scope.unusedAssignments(): self.report(messages.UnusedVariable, binding.source, name) self.deferAssignment(checkUnusedAssignments) if PY2: def checkReturnWithArgumentInsideGenerator(): """ Check to see if there is any return statement with arguments but the function is a generator. """ if self.scope.isGenerator and self.scope.returnValue: self.report(messages.ReturnWithArgsInsideGenerator, self.scope.returnValue) self.deferAssignment(checkReturnWithArgumentInsideGenerator) self.popScope() self.deferFunction(runFunction) def ARGUMENTS(self, node): self.handleChildren(node, omit=('defaults', 'kw_defaults')) if PY2: scope_node = self.getScopeNode(node) if node.vararg: self.addBinding(node, Argument(node.vararg, scope_node)) if node.kwarg: self.addBinding(node, Argument(node.kwarg, scope_node)) def ARG(self, node): self.addBinding(node, Argument(node.arg, self.getScopeNode(node))) def CLASSDEF(self, node): """ Check names used in a class definition, including its decorators, base classes, and the body of its definition. Additionally, add its name to the current scope. """ for deco in node.decorator_list: self.handleNode(deco, node) for baseNode in node.bases: self.handleNode(baseNode, node) if not PY2: for keywordNode in node.keywords: self.handleNode(keywordNode, node) self.pushScope(ClassScope) # doctest does not process doctest within a doctest # classes within classes are processed. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) for stmt in node.body: self.handleNode(stmt, node) self.popScope() self.addBinding(node, ClassDefinition(node.name, node)) def AUGASSIGN(self, node): self.handleNodeLoad(node.target) self.handleNode(node.value, node) self.handleNode(node.target, node) def TUPLE(self, node): if not PY2 and isinstance(node.ctx, ast.Store): # Python 3 advanced tuple unpacking: a, *b, c = d. # Only one starred expression is allowed, and no more than 1<<8 # assignments are allowed before a stared expression. There is # also a limit of 1<<24 expressions after the starred expression, # which is impossible to test due to memory restrictions, but we # add it here anyway has_starred = False star_loc = -1 for i, n in enumerate(node.elts): if isinstance(n, ast.Starred): if has_starred: self.report(messages.TwoStarredExpressions, node) # The SyntaxError doesn't distinguish two from more # than two. break has_starred = True star_loc = i if star_loc >= 1 << 8 or len(node.elts) - star_loc - 1 >= 1 << 24: self.report(messages.TooManyExpressionsInStarredAssignment, node) self.handleChildren(node) LIST = TUPLE def IMPORT(self, node): for alias in node.names: if '.' in alias.name and not alias.asname: importation = SubmoduleImportation(alias.name, node) else: name = alias.asname or alias.name importation = Importation(name, node, alias.name) self.addBinding(node, importation) def IMPORTFROM(self, node): if node.module == '__future__': if not self.futuresAllowed: self.report(messages.LateFutureImport, node, [n.name for n in node.names]) else: self.futuresAllowed = False module = ('.' * node.level) + (node.module or '') for alias in node.names: name = alias.asname or alias.name if node.module == '__future__': importation = FutureImportation(name, node, self.scope) if alias.name not in __future__.all_feature_names: self.report(messages.FutureFeatureNotDefined, node, alias.name) if alias.name == 'annotations': self.annotationsFutureEnabled = True elif alias.name == '*': # Only Python 2, local import * is a SyntaxWarning if not PY2 and not isinstance(self.scope, ModuleScope): self.report(messages.ImportStarNotPermitted, node, module) continue self.scope.importStarred = True self.report(messages.ImportStarUsed, node, module) importation = StarImportation(module, node) else: importation = ImportationFrom(name, node, module, alias.name) self.addBinding(node, importation) def TRY(self, node): handler_names = [] # List the exception handlers for i, handler in enumerate(node.handlers): if isinstance(handler.type, ast.Tuple): for exc_type in handler.type.elts: handler_names.append(getNodeName(exc_type)) elif handler.type: handler_names.append(getNodeName(handler.type)) if handler.type is None and i < len(node.handlers) - 1: self.report(messages.DefaultExceptNotLast, handler) # Memorize the except handlers and process the body self.exceptHandlers.append(handler_names) for child in node.body: self.handleNode(child, node) self.exceptHandlers.pop() # Process the other nodes: "except:", "else:", "finally:" self.handleChildren(node, omit='body') TRYEXCEPT = TRY def EXCEPTHANDLER(self, node): if PY2 or node.name is None: self.handleChildren(node) return # If the name already exists in the scope, modify state of existing # binding. if node.name in self.scope: self.handleNodeStore(node) # 3.x: the name of the exception, which is not a Name node, but a # simple string, creates a local that is only bound within the scope of # the except: block. As such, temporarily remove the existing binding # to more accurately determine if the name is used in the except: # block. try: prev_definition = self.scope.pop(node.name) except KeyError: prev_definition = None self.handleNodeStore(node) self.handleChildren(node) # See discussion on https://github.com/PyCQA/pyflakes/pull/59 # We're removing the local name since it's being unbound after leaving # the except: block and it's always unbound if the except: block is # never entered. This will cause an "undefined name" error raised if # the checked code tries to use the name afterwards. # # Unless it's been removed already. Then do nothing. try: binding = self.scope.pop(node.name) except KeyError: pass else: if not binding.used: self.report(messages.UnusedVariable, node, node.name) # Restore. if prev_definition: self.scope[node.name] = prev_definition def ANNASSIGN(self, node): if node.value: # Only bind the *targets* if the assignment has a value. # Otherwise it's not really ast.Store and shouldn't silence # UndefinedLocal warnings. self.handleNode(node.target, node) self.handleAnnotation(node.annotation, node) if node.value: # If the assignment has value, handle the *value* now. self.handleNode(node.value, node) def COMPARE(self, node): literals = (ast.Str, ast.Num) if not PY2: literals += (ast.Bytes,) left = node.left for op, right in zip(node.ops, node.comparators): if (isinstance(op, (ast.Is, ast.IsNot)) and (isinstance(left, literals) or isinstance(right, literals))): self.report(messages.IsLiteral, node) left = right self.handleChildren(node)
PyCQA/pyflakes
pyflakes/checker.py
iter_child_nodes
python
def iter_child_nodes(node, omit=None, _fields_order=_FieldsOrder()): for name in _fields_order[node.__class__]: if omit and name in omit: continue field = getattr(node, name, None) if isinstance(field, ast.AST): yield field elif isinstance(field, list): for item in field: yield item
Yield all direct child nodes of *node*, that is, all fields that are nodes and all items of fields that are lists of nodes. :param node: AST node to be iterated upon :param omit: String or tuple of strings denoting the attributes of the node to be omitted from further parsing :param _fields_order: Order of AST node fields
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/checker.py#L114-L133
null
""" Main module. Implement the central Checker class. Also, it models the Bindings and Scopes. """ import __future__ import ast import bisect import collections import doctest import functools import os import re import sys import tokenize from pyflakes import messages PY2 = sys.version_info < (3, 0) PY35_PLUS = sys.version_info >= (3, 5) # Python 3.5 and above PY36_PLUS = sys.version_info >= (3, 6) # Python 3.6 and above PY38_PLUS = sys.version_info >= (3, 8) try: sys.pypy_version_info PYPY = True except AttributeError: PYPY = False builtin_vars = dir(__import__('__builtin__' if PY2 else 'builtins')) if PY2: tokenize_tokenize = tokenize.generate_tokens else: tokenize_tokenize = tokenize.tokenize if PY2: def getNodeType(node_class): # workaround str.upper() which is locale-dependent return str(unicode(node_class.__name__).upper()) def get_raise_argument(node): return node.type else: def getNodeType(node_class): return node_class.__name__.upper() def get_raise_argument(node): return node.exc # Silence `pyflakes` from reporting `undefined name 'unicode'` in Python 3. unicode = str # Python >= 3.3 uses ast.Try instead of (ast.TryExcept + ast.TryFinally) if PY2: def getAlternatives(n): if isinstance(n, (ast.If, ast.TryFinally)): return [n.body] if isinstance(n, ast.TryExcept): return [n.body + n.orelse] + [[hdl] for hdl in n.handlers] else: def getAlternatives(n): if isinstance(n, ast.If): return [n.body] if isinstance(n, ast.Try): return [n.body + n.orelse] + [[hdl] for hdl in n.handlers] if PY35_PLUS: FOR_TYPES = (ast.For, ast.AsyncFor) LOOP_TYPES = (ast.While, ast.For, ast.AsyncFor) else: FOR_TYPES = (ast.For,) LOOP_TYPES = (ast.While, ast.For) # https://github.com/python/typed_ast/blob/55420396/ast27/Parser/tokenizer.c#L102-L104 TYPE_COMMENT_RE = re.compile(r'^#\s*type:\s*') # https://github.com/python/typed_ast/blob/55420396/ast27/Parser/tokenizer.c#L1400 TYPE_IGNORE_RE = re.compile(TYPE_COMMENT_RE.pattern + r'ignore\s*(#|$)') # https://github.com/python/typed_ast/blob/55420396/ast27/Grammar/Grammar#L147 TYPE_FUNC_RE = re.compile(r'^(\(.*?\))\s*->\s*(.*)$') class _FieldsOrder(dict): """Fix order of AST node fields.""" def _get_fields(self, node_class): # handle iter before target, and generators before element fields = node_class._fields if 'iter' in fields: key_first = 'iter'.find elif 'generators' in fields: key_first = 'generators'.find else: key_first = 'value'.find return tuple(sorted(fields, key=key_first, reverse=True)) def __missing__(self, node_class): self[node_class] = fields = self._get_fields(node_class) return fields def counter(items): """ Simplest required implementation of collections.Counter. Required as 2.6 does not have Counter in collections. """ results = {} for item in items: results[item] = results.get(item, 0) + 1 return results def convert_to_value(item): if isinstance(item, ast.Str): return item.s elif hasattr(ast, 'Bytes') and isinstance(item, ast.Bytes): return item.s elif isinstance(item, ast.Tuple): return tuple(convert_to_value(i) for i in item.elts) elif isinstance(item, ast.Num): return item.n elif isinstance(item, ast.Name): result = VariableKey(item=item) constants_lookup = { 'True': True, 'False': False, 'None': None, } return constants_lookup.get( result.name, result, ) elif (not PY2) and isinstance(item, ast.NameConstant): # None, True, False are nameconstants in python3, but names in 2 return item.value else: return UnhandledKeyType() def is_notimplemented_name_node(node): return isinstance(node, ast.Name) and getNodeName(node) == 'NotImplemented' class Binding(object): """ Represents the binding of a value to a name. The checker uses this to keep track of which names have been bound and which names have not. See L{Assignment} for a special type of binding that is checked with stricter rules. @ivar used: pair of (L{Scope}, node) indicating the scope and the node that this binding was last used. """ def __init__(self, name, source): self.name = name self.source = source self.used = False def __str__(self): return self.name def __repr__(self): return '<%s object %r from line %r at 0x%x>' % (self.__class__.__name__, self.name, self.source.lineno, id(self)) def redefines(self, other): return isinstance(other, Definition) and self.name == other.name class Definition(Binding): """ A binding that defines a function or a class. """ class Builtin(Definition): """A definition created for all Python builtins.""" def __init__(self, name): super(Builtin, self).__init__(name, None) def __repr__(self): return '<%s object %r at 0x%x>' % (self.__class__.__name__, self.name, id(self)) class UnhandledKeyType(object): """ A dictionary key of a type that we cannot or do not check for duplicates. """ class VariableKey(object): """ A dictionary key which is a variable. @ivar item: The variable AST object. """ def __init__(self, item): self.name = item.id def __eq__(self, compare): return ( compare.__class__ == self.__class__ and compare.name == self.name ) def __hash__(self): return hash(self.name) class Importation(Definition): """ A binding created by an import statement. @ivar fullName: The complete name given to the import statement, possibly including multiple dotted components. @type fullName: C{str} """ def __init__(self, name, source, full_name=None): self.fullName = full_name or name self.redefined = [] super(Importation, self).__init__(name, source) def redefines(self, other): if isinstance(other, SubmoduleImportation): # See note in SubmoduleImportation about RedefinedWhileUnused return self.fullName == other.fullName return isinstance(other, Definition) and self.name == other.name def _has_alias(self): """Return whether importation needs an as clause.""" return not self.fullName.split('.')[-1] == self.name @property def source_statement(self): """Generate a source statement equivalent to the import.""" if self._has_alias(): return 'import %s as %s' % (self.fullName, self.name) else: return 'import %s' % self.fullName def __str__(self): """Return import full name with alias.""" if self._has_alias(): return self.fullName + ' as ' + self.name else: return self.fullName class SubmoduleImportation(Importation): """ A binding created by a submodule import statement. A submodule import is a special case where the root module is implicitly imported, without an 'as' clause, and the submodule is also imported. Python does not restrict which attributes of the root module may be used. This class is only used when the submodule import is without an 'as' clause. pyflakes handles this case by registering the root module name in the scope, allowing any attribute of the root module to be accessed. RedefinedWhileUnused is suppressed in `redefines` unless the submodule name is also the same, to avoid false positives. """ def __init__(self, name, source): # A dot should only appear in the name when it is a submodule import assert '.' in name and (not source or isinstance(source, ast.Import)) package_name = name.split('.')[0] super(SubmoduleImportation, self).__init__(package_name, source) self.fullName = name def redefines(self, other): if isinstance(other, Importation): return self.fullName == other.fullName return super(SubmoduleImportation, self).redefines(other) def __str__(self): return self.fullName @property def source_statement(self): return 'import ' + self.fullName class ImportationFrom(Importation): def __init__(self, name, source, module, real_name=None): self.module = module self.real_name = real_name or name if module.endswith('.'): full_name = module + self.real_name else: full_name = module + '.' + self.real_name super(ImportationFrom, self).__init__(name, source, full_name) def __str__(self): """Return import full name with alias.""" if self.real_name != self.name: return self.fullName + ' as ' + self.name else: return self.fullName @property def source_statement(self): if self.real_name != self.name: return 'from %s import %s as %s' % (self.module, self.real_name, self.name) else: return 'from %s import %s' % (self.module, self.name) class StarImportation(Importation): """A binding created by a 'from x import *' statement.""" def __init__(self, name, source): super(StarImportation, self).__init__('*', source) # Each star importation needs a unique name, and # may not be the module name otherwise it will be deemed imported self.name = name + '.*' self.fullName = name @property def source_statement(self): return 'from ' + self.fullName + ' import *' def __str__(self): # When the module ends with a ., avoid the ambiguous '..*' if self.fullName.endswith('.'): return self.source_statement else: return self.name class FutureImportation(ImportationFrom): """ A binding created by a from `__future__` import statement. `__future__` imports are implicitly used. """ def __init__(self, name, source, scope): super(FutureImportation, self).__init__(name, source, '__future__') self.used = (scope, source) class Argument(Binding): """ Represents binding a name as an argument. """ class Assignment(Binding): """ Represents binding a name with an explicit assignment. The checker will raise warnings for any Assignment that isn't used. Also, the checker does not consider assignments in tuple/list unpacking to be Assignments, rather it treats them as simple Bindings. """ class FunctionDefinition(Definition): pass class ClassDefinition(Definition): pass class ExportBinding(Binding): """ A binding created by an C{__all__} assignment. If the names in the list can be determined statically, they will be treated as names for export and additional checking applied to them. The only recognized C{__all__} assignment via list concatenation is in the following format: __all__ = ['a'] + ['b'] + ['c'] Names which are imported and not otherwise used but appear in the value of C{__all__} will not have an unused import warning reported for them. """ def __init__(self, name, source, scope): if '__all__' in scope and isinstance(source, ast.AugAssign): self.names = list(scope['__all__'].names) else: self.names = [] def _add_to_names(container): for node in container.elts: if isinstance(node, ast.Str): self.names.append(node.s) if isinstance(source.value, (ast.List, ast.Tuple)): _add_to_names(source.value) # If concatenating lists elif isinstance(source.value, ast.BinOp): currentValue = source.value while isinstance(currentValue.right, ast.List): left = currentValue.left right = currentValue.right _add_to_names(right) # If more lists are being added if isinstance(left, ast.BinOp): currentValue = left # If just two lists are being added elif isinstance(left, ast.List): _add_to_names(left) # All lists accounted for - done break # If not list concatenation else: break super(ExportBinding, self).__init__(name, source) class Scope(dict): importStarred = False # set to True when import * is found def __repr__(self): scope_cls = self.__class__.__name__ return '<%s at 0x%x %s>' % (scope_cls, id(self), dict.__repr__(self)) class ClassScope(Scope): pass class FunctionScope(Scope): """ I represent a name scope for a function. @ivar globals: Names declared 'global' in this function. """ usesLocals = False alwaysUsed = {'__tracebackhide__', '__traceback_info__', '__traceback_supplement__'} def __init__(self): super(FunctionScope, self).__init__() # Simplify: manage the special locals as globals self.globals = self.alwaysUsed.copy() self.returnValue = None # First non-empty return self.isGenerator = False # Detect a generator def unusedAssignments(self): """ Return a generator for the assignments which have not been used. """ for name, binding in self.items(): if (not binding.used and name != '_' and # see issue #202 name not in self.globals and not self.usesLocals and isinstance(binding, Assignment)): yield name, binding class GeneratorScope(Scope): pass class ModuleScope(Scope): """Scope for a module.""" _futures_allowed = True _annotations_future_enabled = False class DoctestScope(ModuleScope): """Scope for a doctest.""" class DummyNode(object): """Used in place of an `ast.AST` to set error message positions""" def __init__(self, lineno, col_offset): self.lineno = lineno self.col_offset = col_offset # Globally defined names which are not attributes of the builtins module, or # are only present on some platforms. _MAGIC_GLOBALS = ['__file__', '__builtins__', 'WindowsError'] # module scope annotation will store in `__annotations__`, see also PEP 526. if PY36_PLUS: _MAGIC_GLOBALS.append('__annotations__') def getNodeName(node): # Returns node.id, or node.name, or None if hasattr(node, 'id'): # One of the many nodes with an id return node.id if hasattr(node, 'name'): # an ExceptHandler node return node.name def is_typing_overload(value, scope_stack): def name_is_typing_overload(name): # type: (str) -> bool for scope in reversed(scope_stack): if name in scope: return ( isinstance(scope[name], ImportationFrom) and scope[name].fullName == 'typing.overload' ) else: return False def is_typing_overload_decorator(node): return ( ( isinstance(node, ast.Name) and name_is_typing_overload(node.id) ) or ( isinstance(node, ast.Attribute) and isinstance(node.value, ast.Name) and node.value.id == 'typing' and node.attr == 'overload' ) ) return ( isinstance(value.source, ast.FunctionDef) and any( is_typing_overload_decorator(dec) for dec in value.source.decorator_list ) ) def make_tokens(code): # PY3: tokenize.tokenize requires readline of bytes if not isinstance(code, bytes): code = code.encode('UTF-8') lines = iter(code.splitlines(True)) # next(lines, b'') is to prevent an error in pypy3 return tuple(tokenize_tokenize(lambda: next(lines, b''))) class _TypeableVisitor(ast.NodeVisitor): """Collect the line number and nodes which are deemed typeable by PEP 484 https://www.python.org/dev/peps/pep-0484/#type-comments """ def __init__(self): self.typeable_lines = [] # type: List[int] self.typeable_nodes = {} # type: Dict[int, ast.AST] def _typeable(self, node): # if there is more than one typeable thing on a line last one wins self.typeable_lines.append(node.lineno) self.typeable_nodes[node.lineno] = node self.generic_visit(node) visit_Assign = visit_For = visit_FunctionDef = visit_With = _typeable visit_AsyncFor = visit_AsyncFunctionDef = visit_AsyncWith = _typeable def _collect_type_comments(tree, tokens): visitor = _TypeableVisitor() visitor.visit(tree) type_comments = collections.defaultdict(list) for tp, text, start, _, _ in tokens: if ( tp != tokenize.COMMENT or # skip non comments not TYPE_COMMENT_RE.match(text) or # skip non-type comments TYPE_IGNORE_RE.match(text) # skip ignores ): continue # search for the typeable node at or before the line number of the # type comment. # if the bisection insertion point is before any nodes this is an # invalid type comment which is ignored. lineno, _ = start idx = bisect.bisect_right(visitor.typeable_lines, lineno) if idx == 0: continue node = visitor.typeable_nodes[visitor.typeable_lines[idx - 1]] type_comments[node].append((start, text)) return type_comments class Checker(object): """ I check the cleanliness and sanity of Python code. @ivar _deferredFunctions: Tracking list used by L{deferFunction}. Elements of the list are two-tuples. The first element is the callable passed to L{deferFunction}. The second element is a copy of the scope stack at the time L{deferFunction} was called. @ivar _deferredAssignments: Similar to C{_deferredFunctions}, but for callables which are deferred assignment checks. """ _ast_node_scope = { ast.Module: ModuleScope, ast.ClassDef: ClassScope, ast.FunctionDef: FunctionScope, ast.Lambda: FunctionScope, ast.ListComp: GeneratorScope, ast.SetComp: GeneratorScope, ast.GeneratorExp: GeneratorScope, ast.DictComp: GeneratorScope, } if PY35_PLUS: _ast_node_scope[ast.AsyncFunctionDef] = FunctionScope, nodeDepth = 0 offset = None traceTree = False builtIns = set(builtin_vars).union(_MAGIC_GLOBALS) _customBuiltIns = os.environ.get('PYFLAKES_BUILTINS') if _customBuiltIns: builtIns.update(_customBuiltIns.split(',')) del _customBuiltIns # TODO: file_tokens= is required to perform checks on type comments, # eventually make this a required positional argument. For now it # is defaulted to `()` for api compatibility. def __init__(self, tree, filename='(none)', builtins=None, withDoctest='PYFLAKES_DOCTEST' in os.environ, file_tokens=()): self._nodeHandlers = {} self._deferredFunctions = [] self._deferredAssignments = [] self.deadScopes = [] self.messages = [] self.filename = filename if builtins: self.builtIns = self.builtIns.union(builtins) self.withDoctest = withDoctest try: self.scopeStack = [Checker._ast_node_scope[type(tree)]()] except KeyError: raise RuntimeError('No scope implemented for the node %r' % tree) self.exceptHandlers = [()] self.root = tree self._type_comments = _collect_type_comments(tree, file_tokens) for builtin in self.builtIns: self.addBinding(None, Builtin(builtin)) self.handleChildren(tree) self.runDeferred(self._deferredFunctions) # Set _deferredFunctions to None so that deferFunction will fail # noisily if called after we've run through the deferred functions. self._deferredFunctions = None self.runDeferred(self._deferredAssignments) # Set _deferredAssignments to None so that deferAssignment will fail # noisily if called after we've run through the deferred assignments. self._deferredAssignments = None del self.scopeStack[1:] self.popScope() self.checkDeadScopes() def deferFunction(self, callable): """ Schedule a function handler to be called just before completion. This is used for handling function bodies, which must be deferred because code later in the file might modify the global scope. When `callable` is called, the scope at the time this is called will be restored, however it will contain any new bindings added to it. """ self._deferredFunctions.append((callable, self.scopeStack[:], self.offset)) def deferAssignment(self, callable): """ Schedule an assignment handler to be called just after deferred function handlers. """ self._deferredAssignments.append((callable, self.scopeStack[:], self.offset)) def runDeferred(self, deferred): """ Run the callables in C{deferred} using their associated scope stack. """ for handler, scope, offset in deferred: self.scopeStack = scope self.offset = offset handler() def _in_doctest(self): return (len(self.scopeStack) >= 2 and isinstance(self.scopeStack[1], DoctestScope)) @property def futuresAllowed(self): if not all(isinstance(scope, ModuleScope) for scope in self.scopeStack): return False return self.scope._futures_allowed @futuresAllowed.setter def futuresAllowed(self, value): assert value is False if isinstance(self.scope, ModuleScope): self.scope._futures_allowed = False @property def annotationsFutureEnabled(self): scope = self.scopeStack[0] if not isinstance(scope, ModuleScope): return False return scope._annotations_future_enabled @annotationsFutureEnabled.setter def annotationsFutureEnabled(self, value): assert value is True assert isinstance(self.scope, ModuleScope) self.scope._annotations_future_enabled = True @property def scope(self): return self.scopeStack[-1] def popScope(self): self.deadScopes.append(self.scopeStack.pop()) def checkDeadScopes(self): """ Look at scopes which have been fully examined and report names in them which were imported but unused. """ for scope in self.deadScopes: # imports in classes are public members if isinstance(scope, ClassScope): continue all_binding = scope.get('__all__') if all_binding and not isinstance(all_binding, ExportBinding): all_binding = None if all_binding: all_names = set(all_binding.names) undefined = all_names.difference(scope) else: all_names = undefined = [] if undefined: if not scope.importStarred and \ os.path.basename(self.filename) != '__init__.py': # Look for possible mistakes in the export list for name in undefined: self.report(messages.UndefinedExport, scope['__all__'].source, name) # mark all import '*' as used by the undefined in __all__ if scope.importStarred: from_list = [] for binding in scope.values(): if isinstance(binding, StarImportation): binding.used = all_binding from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) for name in undefined: self.report(messages.ImportStarUsage, scope['__all__'].source, name, from_list) # Look for imported names that aren't used. for value in scope.values(): if isinstance(value, Importation): used = value.used or value.name in all_names if not used: messg = messages.UnusedImport self.report(messg, value.source, str(value)) for node in value.redefined: if isinstance(self.getParent(node), FOR_TYPES): messg = messages.ImportShadowedByLoopVar elif used: continue else: messg = messages.RedefinedWhileUnused self.report(messg, node, value.name, value.source) def pushScope(self, scopeClass=FunctionScope): self.scopeStack.append(scopeClass()) def report(self, messageClass, *args, **kwargs): self.messages.append(messageClass(self.filename, *args, **kwargs)) def getParent(self, node): # Lookup the first parent which is not Tuple, List or Starred while True: node = node.parent if not hasattr(node, 'elts') and not hasattr(node, 'ctx'): return node def getCommonAncestor(self, lnode, rnode, stop): if stop in (lnode, rnode) or not (hasattr(lnode, 'parent') and hasattr(rnode, 'parent')): return None if lnode is rnode: return lnode if (lnode.depth > rnode.depth): return self.getCommonAncestor(lnode.parent, rnode, stop) if (lnode.depth < rnode.depth): return self.getCommonAncestor(lnode, rnode.parent, stop) return self.getCommonAncestor(lnode.parent, rnode.parent, stop) def descendantOf(self, node, ancestors, stop): for a in ancestors: if self.getCommonAncestor(node, a, stop): return True return False def _getAncestor(self, node, ancestor_type): parent = node while True: if parent is self.root: return None parent = self.getParent(parent) if isinstance(parent, ancestor_type): return parent def getScopeNode(self, node): return self._getAncestor(node, tuple(Checker._ast_node_scope.keys())) def differentForks(self, lnode, rnode): """True, if lnode and rnode are located on different forks of IF/TRY""" ancestor = self.getCommonAncestor(lnode, rnode, self.root) parts = getAlternatives(ancestor) if parts: for items in parts: if self.descendantOf(lnode, items, ancestor) ^ \ self.descendantOf(rnode, items, ancestor): return True return False def addBinding(self, node, value): """ Called when a binding is altered. - `node` is the statement responsible for the change - `value` is the new value, a Binding instance """ # assert value.source in (node, node.parent): for scope in self.scopeStack[::-1]: if value.name in scope: break existing = scope.get(value.name) if (existing and not isinstance(existing, Builtin) and not self.differentForks(node, existing.source)): parent_stmt = self.getParent(value.source) if isinstance(existing, Importation) and isinstance(parent_stmt, FOR_TYPES): self.report(messages.ImportShadowedByLoopVar, node, value.name, existing.source) elif scope is self.scope: if (isinstance(parent_stmt, ast.comprehension) and not isinstance(self.getParent(existing.source), (FOR_TYPES, ast.comprehension))): self.report(messages.RedefinedInListComp, node, value.name, existing.source) elif not existing.used and value.redefines(existing): if value.name != '_' or isinstance(existing, Importation): if not is_typing_overload(existing, self.scopeStack): self.report(messages.RedefinedWhileUnused, node, value.name, existing.source) elif isinstance(existing, Importation) and value.redefines(existing): existing.redefined.append(node) if value.name in self.scope: # then assume the rebound name is used as a global or within a loop value.used = self.scope[value.name].used self.scope[value.name] = value def getNodeHandler(self, node_class): try: return self._nodeHandlers[node_class] except KeyError: nodeType = getNodeType(node_class) self._nodeHandlers[node_class] = handler = getattr(self, nodeType) return handler def handleNodeLoad(self, node): name = getNodeName(node) if not name: return in_generators = None importStarred = None # try enclosing function scopes and global scope for scope in self.scopeStack[-1::-1]: if isinstance(scope, ClassScope): if not PY2 and name == '__class__': return elif in_generators is False: # only generators used in a class scope can access the # names of the class. this is skipped during the first # iteration continue if (name == 'print' and isinstance(scope.get(name, None), Builtin)): parent = self.getParent(node) if (isinstance(parent, ast.BinOp) and isinstance(parent.op, ast.RShift)): self.report(messages.InvalidPrintSyntax, node) try: scope[name].used = (self.scope, node) # if the name of SubImportation is same as # alias of other Importation and the alias # is used, SubImportation also should be marked as used. n = scope[name] if isinstance(n, Importation) and n._has_alias(): try: scope[n.fullName].used = (self.scope, node) except KeyError: pass except KeyError: pass else: return importStarred = importStarred or scope.importStarred if in_generators is not False: in_generators = isinstance(scope, GeneratorScope) if importStarred: from_list = [] for scope in self.scopeStack[-1::-1]: for binding in scope.values(): if isinstance(binding, StarImportation): # mark '*' imports as used for each scope binding.used = (self.scope, node) from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) self.report(messages.ImportStarUsage, node, name, from_list) return if name == '__path__' and os.path.basename(self.filename) == '__init__.py': # the special name __path__ is valid only in packages return if name == '__module__' and isinstance(self.scope, ClassScope): return # protected with a NameError handler? if 'NameError' not in self.exceptHandlers[-1]: self.report(messages.UndefinedName, node, name) def handleNodeStore(self, node): name = getNodeName(node) if not name: return # if the name hasn't already been defined in the current scope if isinstance(self.scope, FunctionScope) and name not in self.scope: # for each function or module scope above us for scope in self.scopeStack[:-1]: if not isinstance(scope, (FunctionScope, ModuleScope)): continue # if the name was defined in that scope, and the name has # been accessed already in the current scope, and hasn't # been declared global used = name in scope and scope[name].used if used and used[0] is self.scope and name not in self.scope.globals: # then it's probably a mistake self.report(messages.UndefinedLocal, scope[name].used[1], name, scope[name].source) break parent_stmt = self.getParent(node) if isinstance(parent_stmt, (FOR_TYPES, ast.comprehension)) or ( parent_stmt != node.parent and not self.isLiteralTupleUnpacking(parent_stmt)): binding = Binding(name, node) elif name == '__all__' and isinstance(self.scope, ModuleScope): binding = ExportBinding(name, node.parent, self.scope) elif isinstance(getattr(node, 'ctx', None), ast.Param): binding = Argument(name, self.getScopeNode(node)) else: binding = Assignment(name, node) self.addBinding(node, binding) def handleNodeDelete(self, node): def on_conditional_branch(): """ Return `True` if node is part of a conditional body. """ current = getattr(node, 'parent', None) while current: if isinstance(current, (ast.If, ast.While, ast.IfExp)): return True current = getattr(current, 'parent', None) return False name = getNodeName(node) if not name: return if on_conditional_branch(): # We cannot predict if this conditional branch is going to # be executed. return if isinstance(self.scope, FunctionScope) and name in self.scope.globals: self.scope.globals.remove(name) else: try: del self.scope[name] except KeyError: self.report(messages.UndefinedName, node, name) def _handle_type_comments(self, node): for (lineno, col_offset), comment in self._type_comments.get(node, ()): comment = comment.split(':', 1)[1].strip() func_match = TYPE_FUNC_RE.match(comment) if func_match: parts = ( func_match.group(1).replace('*', ''), func_match.group(2).strip(), ) else: parts = (comment,) for part in parts: if PY2: part = part.replace('...', 'Ellipsis') self.deferFunction(functools.partial( self.handleStringAnnotation, part, DummyNode(lineno, col_offset), lineno, col_offset, messages.CommentAnnotationSyntaxError, )) def handleChildren(self, tree, omit=None): self._handle_type_comments(tree) for node in iter_child_nodes(tree, omit=omit): self.handleNode(node, tree) def isLiteralTupleUnpacking(self, node): if isinstance(node, ast.Assign): for child in node.targets + [node.value]: if not hasattr(child, 'elts'): return False return True def isDocstring(self, node): """ Determine if the given node is a docstring, as long as it is at the correct place in the node tree. """ return isinstance(node, ast.Str) or (isinstance(node, ast.Expr) and isinstance(node.value, ast.Str)) def getDocstring(self, node): if isinstance(node, ast.Expr): node = node.value if not isinstance(node, ast.Str): return (None, None) if PYPY or PY38_PLUS: doctest_lineno = node.lineno - 1 else: # Computed incorrectly if the docstring has backslash doctest_lineno = node.lineno - node.s.count('\n') - 1 return (node.s, doctest_lineno) def handleNode(self, node, parent): if node is None: return if self.offset and getattr(node, 'lineno', None) is not None: node.lineno += self.offset[0] node.col_offset += self.offset[1] if self.traceTree: print(' ' * self.nodeDepth + node.__class__.__name__) if self.futuresAllowed and not (isinstance(node, ast.ImportFrom) or self.isDocstring(node)): self.futuresAllowed = False self.nodeDepth += 1 node.depth = self.nodeDepth node.parent = parent try: handler = self.getNodeHandler(node.__class__) handler(node) finally: self.nodeDepth -= 1 if self.traceTree: print(' ' * self.nodeDepth + 'end ' + node.__class__.__name__) _getDoctestExamples = doctest.DocTestParser().get_examples def handleDoctests(self, node): try: if hasattr(node, 'docstring'): docstring = node.docstring # This is just a reasonable guess. In Python 3.7, docstrings no # longer have line numbers associated with them. This will be # incorrect if there are empty lines between the beginning # of the function and the docstring. node_lineno = node.lineno if hasattr(node, 'args'): node_lineno = max([node_lineno] + [arg.lineno for arg in node.args.args]) else: (docstring, node_lineno) = self.getDocstring(node.body[0]) examples = docstring and self._getDoctestExamples(docstring) except (ValueError, IndexError): # e.g. line 6 of the docstring for <string> has inconsistent # leading whitespace: ... return if not examples: return # Place doctest in module scope saved_stack = self.scopeStack self.scopeStack = [self.scopeStack[0]] node_offset = self.offset or (0, 0) self.pushScope(DoctestScope) self.addBinding(None, Builtin('_')) for example in examples: try: tree = ast.parse(example.source, "<doctest>") except SyntaxError: e = sys.exc_info()[1] if PYPY: e.offset += 1 position = (node_lineno + example.lineno + e.lineno, example.indent + 4 + (e.offset or 0)) self.report(messages.DoctestSyntaxError, node, position) else: self.offset = (node_offset[0] + node_lineno + example.lineno, node_offset[1] + example.indent + 4) self.handleChildren(tree) self.offset = node_offset self.popScope() self.scopeStack = saved_stack def handleStringAnnotation(self, s, node, ref_lineno, ref_col_offset, err): try: tree = ast.parse(s) except SyntaxError: self.report(err, node, s) return body = tree.body if len(body) != 1 or not isinstance(body[0], ast.Expr): self.report(err, node, s) return parsed_annotation = tree.body[0].value for descendant in ast.walk(parsed_annotation): if ( 'lineno' in descendant._attributes and 'col_offset' in descendant._attributes ): descendant.lineno = ref_lineno descendant.col_offset = ref_col_offset self.handleNode(parsed_annotation, node) def handleAnnotation(self, annotation, node): if isinstance(annotation, ast.Str): # Defer handling forward annotation. self.deferFunction(functools.partial( self.handleStringAnnotation, annotation.s, node, annotation.lineno, annotation.col_offset, messages.ForwardAnnotationSyntaxError, )) elif self.annotationsFutureEnabled: self.deferFunction(lambda: self.handleNode(annotation, node)) else: self.handleNode(annotation, node) def ignore(self, node): pass # "stmt" type nodes DELETE = PRINT = FOR = ASYNCFOR = WHILE = IF = WITH = WITHITEM = \ ASYNCWITH = ASYNCWITHITEM = TRYFINALLY = EXEC = \ EXPR = ASSIGN = handleChildren PASS = ignore # "expr" type nodes BOOLOP = BINOP = UNARYOP = IFEXP = SET = \ CALL = REPR = ATTRIBUTE = SUBSCRIPT = \ STARRED = NAMECONSTANT = handleChildren NUM = STR = BYTES = ELLIPSIS = CONSTANT = ignore # "slice" type nodes SLICE = EXTSLICE = INDEX = handleChildren # expression contexts are node instances too, though being constants LOAD = STORE = DEL = AUGLOAD = AUGSTORE = PARAM = ignore # same for operators AND = OR = ADD = SUB = MULT = DIV = MOD = POW = LSHIFT = RSHIFT = \ BITOR = BITXOR = BITAND = FLOORDIV = INVERT = NOT = UADD = USUB = \ EQ = NOTEQ = LT = LTE = GT = GTE = IS = ISNOT = IN = NOTIN = \ MATMULT = ignore def RAISE(self, node): self.handleChildren(node) arg = get_raise_argument(node) if isinstance(arg, ast.Call): if is_notimplemented_name_node(arg.func): # Handle "raise NotImplemented(...)" self.report(messages.RaiseNotImplemented, node) elif is_notimplemented_name_node(arg): # Handle "raise NotImplemented" self.report(messages.RaiseNotImplemented, node) # additional node types COMPREHENSION = KEYWORD = FORMATTEDVALUE = JOINEDSTR = handleChildren def DICT(self, node): # Complain if there are duplicate keys with different values # If they have the same value it's not going to cause potentially # unexpected behaviour so we'll not complain. keys = [ convert_to_value(key) for key in node.keys ] key_counts = counter(keys) duplicate_keys = [ key for key, count in key_counts.items() if count > 1 ] for key in duplicate_keys: key_indices = [i for i, i_key in enumerate(keys) if i_key == key] values = counter( convert_to_value(node.values[index]) for index in key_indices ) if any(count == 1 for value, count in values.items()): for key_index in key_indices: key_node = node.keys[key_index] if isinstance(key, VariableKey): self.report(messages.MultiValueRepeatedKeyVariable, key_node, key.name) else: self.report( messages.MultiValueRepeatedKeyLiteral, key_node, key, ) self.handleChildren(node) def ASSERT(self, node): if isinstance(node.test, ast.Tuple) and node.test.elts != []: self.report(messages.AssertTuple, node) self.handleChildren(node) def GLOBAL(self, node): """ Keep track of globals declarations. """ global_scope_index = 1 if self._in_doctest() else 0 global_scope = self.scopeStack[global_scope_index] # Ignore 'global' statement in global scope. if self.scope is not global_scope: # One 'global' statement can bind multiple (comma-delimited) names. for node_name in node.names: node_value = Assignment(node_name, node) # Remove UndefinedName messages already reported for this name. # TODO: if the global is not used in this scope, it does not # become a globally defined name. See test_unused_global. self.messages = [ m for m in self.messages if not isinstance(m, messages.UndefinedName) or m.message_args[0] != node_name] # Bind name to global scope if it doesn't exist already. global_scope.setdefault(node_name, node_value) # Bind name to non-global scopes, but as already "used". node_value.used = (global_scope, node) for scope in self.scopeStack[global_scope_index + 1:]: scope[node_name] = node_value NONLOCAL = GLOBAL def GENERATOREXP(self, node): self.pushScope(GeneratorScope) self.handleChildren(node) self.popScope() LISTCOMP = handleChildren if PY2 else GENERATOREXP DICTCOMP = SETCOMP = GENERATOREXP def NAME(self, node): """ Handle occurrence of Name (which can be a load/store/delete access.) """ # Locate the name in locals / function / globals scopes. if isinstance(node.ctx, (ast.Load, ast.AugLoad)): self.handleNodeLoad(node) if (node.id == 'locals' and isinstance(self.scope, FunctionScope) and isinstance(node.parent, ast.Call)): # we are doing locals() call in current scope self.scope.usesLocals = True elif isinstance(node.ctx, (ast.Store, ast.AugStore, ast.Param)): self.handleNodeStore(node) elif isinstance(node.ctx, ast.Del): self.handleNodeDelete(node) else: # Unknown context raise RuntimeError("Got impossible expression context: %r" % (node.ctx,)) def CONTINUE(self, node): # Walk the tree up until we see a loop (OK), a function or class # definition (not OK), for 'continue', a finally block (not OK), or # the top module scope (not OK) n = node while hasattr(n, 'parent'): n, n_child = n.parent, n if isinstance(n, LOOP_TYPES): # Doesn't apply unless it's in the loop itself if n_child not in n.orelse: return if isinstance(n, (ast.FunctionDef, ast.ClassDef)): break # Handle Try/TryFinally difference in Python < and >= 3.3 if hasattr(n, 'finalbody') and isinstance(node, ast.Continue): if n_child in n.finalbody: self.report(messages.ContinueInFinally, node) return if isinstance(node, ast.Continue): self.report(messages.ContinueOutsideLoop, node) else: # ast.Break self.report(messages.BreakOutsideLoop, node) BREAK = CONTINUE def RETURN(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.ReturnOutsideFunction, node) return if ( node.value and hasattr(self.scope, 'returnValue') and not self.scope.returnValue ): self.scope.returnValue = node.value self.handleNode(node.value, node) def YIELD(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.YieldOutsideFunction, node) return self.scope.isGenerator = True self.handleNode(node.value, node) AWAIT = YIELDFROM = YIELD def FUNCTIONDEF(self, node): for deco in node.decorator_list: self.handleNode(deco, node) self.LAMBDA(node) self.addBinding(node, FunctionDefinition(node.name, node)) # doctest does not process doctest within a doctest, # or in nested functions. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) ASYNCFUNCTIONDEF = FUNCTIONDEF def LAMBDA(self, node): args = [] annotations = [] if PY2: def addArgs(arglist): for arg in arglist: if isinstance(arg, ast.Tuple): addArgs(arg.elts) else: args.append(arg.id) addArgs(node.args.args) defaults = node.args.defaults else: for arg in node.args.args + node.args.kwonlyargs: args.append(arg.arg) annotations.append(arg.annotation) defaults = node.args.defaults + node.args.kw_defaults # Only for Python3 FunctionDefs is_py3_func = hasattr(node, 'returns') for arg_name in ('vararg', 'kwarg'): wildcard = getattr(node.args, arg_name) if not wildcard: continue args.append(wildcard if PY2 else wildcard.arg) if is_py3_func: if PY2: # Python 2.7 argannotation = arg_name + 'annotation' annotations.append(getattr(node.args, argannotation)) else: # Python >= 3.4 annotations.append(wildcard.annotation) if is_py3_func: annotations.append(node.returns) if len(set(args)) < len(args): for (idx, arg) in enumerate(args): if arg in args[:idx]: self.report(messages.DuplicateArgument, node, arg) for annotation in annotations: self.handleAnnotation(annotation, node) for default in defaults: self.handleNode(default, node) def runFunction(): self.pushScope() self.handleChildren(node, omit='decorator_list') def checkUnusedAssignments(): """ Check to see if any assignments have not been used. """ for name, binding in self.scope.unusedAssignments(): self.report(messages.UnusedVariable, binding.source, name) self.deferAssignment(checkUnusedAssignments) if PY2: def checkReturnWithArgumentInsideGenerator(): """ Check to see if there is any return statement with arguments but the function is a generator. """ if self.scope.isGenerator and self.scope.returnValue: self.report(messages.ReturnWithArgsInsideGenerator, self.scope.returnValue) self.deferAssignment(checkReturnWithArgumentInsideGenerator) self.popScope() self.deferFunction(runFunction) def ARGUMENTS(self, node): self.handleChildren(node, omit=('defaults', 'kw_defaults')) if PY2: scope_node = self.getScopeNode(node) if node.vararg: self.addBinding(node, Argument(node.vararg, scope_node)) if node.kwarg: self.addBinding(node, Argument(node.kwarg, scope_node)) def ARG(self, node): self.addBinding(node, Argument(node.arg, self.getScopeNode(node))) def CLASSDEF(self, node): """ Check names used in a class definition, including its decorators, base classes, and the body of its definition. Additionally, add its name to the current scope. """ for deco in node.decorator_list: self.handleNode(deco, node) for baseNode in node.bases: self.handleNode(baseNode, node) if not PY2: for keywordNode in node.keywords: self.handleNode(keywordNode, node) self.pushScope(ClassScope) # doctest does not process doctest within a doctest # classes within classes are processed. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) for stmt in node.body: self.handleNode(stmt, node) self.popScope() self.addBinding(node, ClassDefinition(node.name, node)) def AUGASSIGN(self, node): self.handleNodeLoad(node.target) self.handleNode(node.value, node) self.handleNode(node.target, node) def TUPLE(self, node): if not PY2 and isinstance(node.ctx, ast.Store): # Python 3 advanced tuple unpacking: a, *b, c = d. # Only one starred expression is allowed, and no more than 1<<8 # assignments are allowed before a stared expression. There is # also a limit of 1<<24 expressions after the starred expression, # which is impossible to test due to memory restrictions, but we # add it here anyway has_starred = False star_loc = -1 for i, n in enumerate(node.elts): if isinstance(n, ast.Starred): if has_starred: self.report(messages.TwoStarredExpressions, node) # The SyntaxError doesn't distinguish two from more # than two. break has_starred = True star_loc = i if star_loc >= 1 << 8 or len(node.elts) - star_loc - 1 >= 1 << 24: self.report(messages.TooManyExpressionsInStarredAssignment, node) self.handleChildren(node) LIST = TUPLE def IMPORT(self, node): for alias in node.names: if '.' in alias.name and not alias.asname: importation = SubmoduleImportation(alias.name, node) else: name = alias.asname or alias.name importation = Importation(name, node, alias.name) self.addBinding(node, importation) def IMPORTFROM(self, node): if node.module == '__future__': if not self.futuresAllowed: self.report(messages.LateFutureImport, node, [n.name for n in node.names]) else: self.futuresAllowed = False module = ('.' * node.level) + (node.module or '') for alias in node.names: name = alias.asname or alias.name if node.module == '__future__': importation = FutureImportation(name, node, self.scope) if alias.name not in __future__.all_feature_names: self.report(messages.FutureFeatureNotDefined, node, alias.name) if alias.name == 'annotations': self.annotationsFutureEnabled = True elif alias.name == '*': # Only Python 2, local import * is a SyntaxWarning if not PY2 and not isinstance(self.scope, ModuleScope): self.report(messages.ImportStarNotPermitted, node, module) continue self.scope.importStarred = True self.report(messages.ImportStarUsed, node, module) importation = StarImportation(module, node) else: importation = ImportationFrom(name, node, module, alias.name) self.addBinding(node, importation) def TRY(self, node): handler_names = [] # List the exception handlers for i, handler in enumerate(node.handlers): if isinstance(handler.type, ast.Tuple): for exc_type in handler.type.elts: handler_names.append(getNodeName(exc_type)) elif handler.type: handler_names.append(getNodeName(handler.type)) if handler.type is None and i < len(node.handlers) - 1: self.report(messages.DefaultExceptNotLast, handler) # Memorize the except handlers and process the body self.exceptHandlers.append(handler_names) for child in node.body: self.handleNode(child, node) self.exceptHandlers.pop() # Process the other nodes: "except:", "else:", "finally:" self.handleChildren(node, omit='body') TRYEXCEPT = TRY def EXCEPTHANDLER(self, node): if PY2 or node.name is None: self.handleChildren(node) return # If the name already exists in the scope, modify state of existing # binding. if node.name in self.scope: self.handleNodeStore(node) # 3.x: the name of the exception, which is not a Name node, but a # simple string, creates a local that is only bound within the scope of # the except: block. As such, temporarily remove the existing binding # to more accurately determine if the name is used in the except: # block. try: prev_definition = self.scope.pop(node.name) except KeyError: prev_definition = None self.handleNodeStore(node) self.handleChildren(node) # See discussion on https://github.com/PyCQA/pyflakes/pull/59 # We're removing the local name since it's being unbound after leaving # the except: block and it's always unbound if the except: block is # never entered. This will cause an "undefined name" error raised if # the checked code tries to use the name afterwards. # # Unless it's been removed already. Then do nothing. try: binding = self.scope.pop(node.name) except KeyError: pass else: if not binding.used: self.report(messages.UnusedVariable, node, node.name) # Restore. if prev_definition: self.scope[node.name] = prev_definition def ANNASSIGN(self, node): if node.value: # Only bind the *targets* if the assignment has a value. # Otherwise it's not really ast.Store and shouldn't silence # UndefinedLocal warnings. self.handleNode(node.target, node) self.handleAnnotation(node.annotation, node) if node.value: # If the assignment has value, handle the *value* now. self.handleNode(node.value, node) def COMPARE(self, node): literals = (ast.Str, ast.Num) if not PY2: literals += (ast.Bytes,) left = node.left for op, right in zip(node.ops, node.comparators): if (isinstance(op, (ast.Is, ast.IsNot)) and (isinstance(left, literals) or isinstance(right, literals))): self.report(messages.IsLiteral, node) left = right self.handleChildren(node)
PyCQA/pyflakes
pyflakes/checker.py
Importation.source_statement
python
def source_statement(self): if self._has_alias(): return 'import %s as %s' % (self.fullName, self.name) else: return 'import %s' % self.fullName
Generate a source statement equivalent to the import.
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/checker.py#L265-L270
[ "def _has_alias(self):\n \"\"\"Return whether importation needs an as clause.\"\"\"\n return not self.fullName.split('.')[-1] == self.name\n" ]
class Importation(Definition): """ A binding created by an import statement. @ivar fullName: The complete name given to the import statement, possibly including multiple dotted components. @type fullName: C{str} """ def __init__(self, name, source, full_name=None): self.fullName = full_name or name self.redefined = [] super(Importation, self).__init__(name, source) def redefines(self, other): if isinstance(other, SubmoduleImportation): # See note in SubmoduleImportation about RedefinedWhileUnused return self.fullName == other.fullName return isinstance(other, Definition) and self.name == other.name def _has_alias(self): """Return whether importation needs an as clause.""" return not self.fullName.split('.')[-1] == self.name @property def __str__(self): """Return import full name with alias.""" if self._has_alias(): return self.fullName + ' as ' + self.name else: return self.fullName
PyCQA/pyflakes
pyflakes/checker.py
Checker.deferFunction
python
def deferFunction(self, callable): self._deferredFunctions.append((callable, self.scopeStack[:], self.offset))
Schedule a function handler to be called just before completion. This is used for handling function bodies, which must be deferred because code later in the file might modify the global scope. When `callable` is called, the scope at the time this is called will be restored, however it will contain any new bindings added to it.
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/checker.py#L694-L703
null
class Checker(object): """ I check the cleanliness and sanity of Python code. @ivar _deferredFunctions: Tracking list used by L{deferFunction}. Elements of the list are two-tuples. The first element is the callable passed to L{deferFunction}. The second element is a copy of the scope stack at the time L{deferFunction} was called. @ivar _deferredAssignments: Similar to C{_deferredFunctions}, but for callables which are deferred assignment checks. """ _ast_node_scope = { ast.Module: ModuleScope, ast.ClassDef: ClassScope, ast.FunctionDef: FunctionScope, ast.Lambda: FunctionScope, ast.ListComp: GeneratorScope, ast.SetComp: GeneratorScope, ast.GeneratorExp: GeneratorScope, ast.DictComp: GeneratorScope, } if PY35_PLUS: _ast_node_scope[ast.AsyncFunctionDef] = FunctionScope, nodeDepth = 0 offset = None traceTree = False builtIns = set(builtin_vars).union(_MAGIC_GLOBALS) _customBuiltIns = os.environ.get('PYFLAKES_BUILTINS') if _customBuiltIns: builtIns.update(_customBuiltIns.split(',')) del _customBuiltIns # TODO: file_tokens= is required to perform checks on type comments, # eventually make this a required positional argument. For now it # is defaulted to `()` for api compatibility. def __init__(self, tree, filename='(none)', builtins=None, withDoctest='PYFLAKES_DOCTEST' in os.environ, file_tokens=()): self._nodeHandlers = {} self._deferredFunctions = [] self._deferredAssignments = [] self.deadScopes = [] self.messages = [] self.filename = filename if builtins: self.builtIns = self.builtIns.union(builtins) self.withDoctest = withDoctest try: self.scopeStack = [Checker._ast_node_scope[type(tree)]()] except KeyError: raise RuntimeError('No scope implemented for the node %r' % tree) self.exceptHandlers = [()] self.root = tree self._type_comments = _collect_type_comments(tree, file_tokens) for builtin in self.builtIns: self.addBinding(None, Builtin(builtin)) self.handleChildren(tree) self.runDeferred(self._deferredFunctions) # Set _deferredFunctions to None so that deferFunction will fail # noisily if called after we've run through the deferred functions. self._deferredFunctions = None self.runDeferred(self._deferredAssignments) # Set _deferredAssignments to None so that deferAssignment will fail # noisily if called after we've run through the deferred assignments. self._deferredAssignments = None del self.scopeStack[1:] self.popScope() self.checkDeadScopes() def deferAssignment(self, callable): """ Schedule an assignment handler to be called just after deferred function handlers. """ self._deferredAssignments.append((callable, self.scopeStack[:], self.offset)) def runDeferred(self, deferred): """ Run the callables in C{deferred} using their associated scope stack. """ for handler, scope, offset in deferred: self.scopeStack = scope self.offset = offset handler() def _in_doctest(self): return (len(self.scopeStack) >= 2 and isinstance(self.scopeStack[1], DoctestScope)) @property def futuresAllowed(self): if not all(isinstance(scope, ModuleScope) for scope in self.scopeStack): return False return self.scope._futures_allowed @futuresAllowed.setter def futuresAllowed(self, value): assert value is False if isinstance(self.scope, ModuleScope): self.scope._futures_allowed = False @property def annotationsFutureEnabled(self): scope = self.scopeStack[0] if not isinstance(scope, ModuleScope): return False return scope._annotations_future_enabled @annotationsFutureEnabled.setter def annotationsFutureEnabled(self, value): assert value is True assert isinstance(self.scope, ModuleScope) self.scope._annotations_future_enabled = True @property def scope(self): return self.scopeStack[-1] def popScope(self): self.deadScopes.append(self.scopeStack.pop()) def checkDeadScopes(self): """ Look at scopes which have been fully examined and report names in them which were imported but unused. """ for scope in self.deadScopes: # imports in classes are public members if isinstance(scope, ClassScope): continue all_binding = scope.get('__all__') if all_binding and not isinstance(all_binding, ExportBinding): all_binding = None if all_binding: all_names = set(all_binding.names) undefined = all_names.difference(scope) else: all_names = undefined = [] if undefined: if not scope.importStarred and \ os.path.basename(self.filename) != '__init__.py': # Look for possible mistakes in the export list for name in undefined: self.report(messages.UndefinedExport, scope['__all__'].source, name) # mark all import '*' as used by the undefined in __all__ if scope.importStarred: from_list = [] for binding in scope.values(): if isinstance(binding, StarImportation): binding.used = all_binding from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) for name in undefined: self.report(messages.ImportStarUsage, scope['__all__'].source, name, from_list) # Look for imported names that aren't used. for value in scope.values(): if isinstance(value, Importation): used = value.used or value.name in all_names if not used: messg = messages.UnusedImport self.report(messg, value.source, str(value)) for node in value.redefined: if isinstance(self.getParent(node), FOR_TYPES): messg = messages.ImportShadowedByLoopVar elif used: continue else: messg = messages.RedefinedWhileUnused self.report(messg, node, value.name, value.source) def pushScope(self, scopeClass=FunctionScope): self.scopeStack.append(scopeClass()) def report(self, messageClass, *args, **kwargs): self.messages.append(messageClass(self.filename, *args, **kwargs)) def getParent(self, node): # Lookup the first parent which is not Tuple, List or Starred while True: node = node.parent if not hasattr(node, 'elts') and not hasattr(node, 'ctx'): return node def getCommonAncestor(self, lnode, rnode, stop): if stop in (lnode, rnode) or not (hasattr(lnode, 'parent') and hasattr(rnode, 'parent')): return None if lnode is rnode: return lnode if (lnode.depth > rnode.depth): return self.getCommonAncestor(lnode.parent, rnode, stop) if (lnode.depth < rnode.depth): return self.getCommonAncestor(lnode, rnode.parent, stop) return self.getCommonAncestor(lnode.parent, rnode.parent, stop) def descendantOf(self, node, ancestors, stop): for a in ancestors: if self.getCommonAncestor(node, a, stop): return True return False def _getAncestor(self, node, ancestor_type): parent = node while True: if parent is self.root: return None parent = self.getParent(parent) if isinstance(parent, ancestor_type): return parent def getScopeNode(self, node): return self._getAncestor(node, tuple(Checker._ast_node_scope.keys())) def differentForks(self, lnode, rnode): """True, if lnode and rnode are located on different forks of IF/TRY""" ancestor = self.getCommonAncestor(lnode, rnode, self.root) parts = getAlternatives(ancestor) if parts: for items in parts: if self.descendantOf(lnode, items, ancestor) ^ \ self.descendantOf(rnode, items, ancestor): return True return False def addBinding(self, node, value): """ Called when a binding is altered. - `node` is the statement responsible for the change - `value` is the new value, a Binding instance """ # assert value.source in (node, node.parent): for scope in self.scopeStack[::-1]: if value.name in scope: break existing = scope.get(value.name) if (existing and not isinstance(existing, Builtin) and not self.differentForks(node, existing.source)): parent_stmt = self.getParent(value.source) if isinstance(existing, Importation) and isinstance(parent_stmt, FOR_TYPES): self.report(messages.ImportShadowedByLoopVar, node, value.name, existing.source) elif scope is self.scope: if (isinstance(parent_stmt, ast.comprehension) and not isinstance(self.getParent(existing.source), (FOR_TYPES, ast.comprehension))): self.report(messages.RedefinedInListComp, node, value.name, existing.source) elif not existing.used and value.redefines(existing): if value.name != '_' or isinstance(existing, Importation): if not is_typing_overload(existing, self.scopeStack): self.report(messages.RedefinedWhileUnused, node, value.name, existing.source) elif isinstance(existing, Importation) and value.redefines(existing): existing.redefined.append(node) if value.name in self.scope: # then assume the rebound name is used as a global or within a loop value.used = self.scope[value.name].used self.scope[value.name] = value def getNodeHandler(self, node_class): try: return self._nodeHandlers[node_class] except KeyError: nodeType = getNodeType(node_class) self._nodeHandlers[node_class] = handler = getattr(self, nodeType) return handler def handleNodeLoad(self, node): name = getNodeName(node) if not name: return in_generators = None importStarred = None # try enclosing function scopes and global scope for scope in self.scopeStack[-1::-1]: if isinstance(scope, ClassScope): if not PY2 and name == '__class__': return elif in_generators is False: # only generators used in a class scope can access the # names of the class. this is skipped during the first # iteration continue if (name == 'print' and isinstance(scope.get(name, None), Builtin)): parent = self.getParent(node) if (isinstance(parent, ast.BinOp) and isinstance(parent.op, ast.RShift)): self.report(messages.InvalidPrintSyntax, node) try: scope[name].used = (self.scope, node) # if the name of SubImportation is same as # alias of other Importation and the alias # is used, SubImportation also should be marked as used. n = scope[name] if isinstance(n, Importation) and n._has_alias(): try: scope[n.fullName].used = (self.scope, node) except KeyError: pass except KeyError: pass else: return importStarred = importStarred or scope.importStarred if in_generators is not False: in_generators = isinstance(scope, GeneratorScope) if importStarred: from_list = [] for scope in self.scopeStack[-1::-1]: for binding in scope.values(): if isinstance(binding, StarImportation): # mark '*' imports as used for each scope binding.used = (self.scope, node) from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) self.report(messages.ImportStarUsage, node, name, from_list) return if name == '__path__' and os.path.basename(self.filename) == '__init__.py': # the special name __path__ is valid only in packages return if name == '__module__' and isinstance(self.scope, ClassScope): return # protected with a NameError handler? if 'NameError' not in self.exceptHandlers[-1]: self.report(messages.UndefinedName, node, name) def handleNodeStore(self, node): name = getNodeName(node) if not name: return # if the name hasn't already been defined in the current scope if isinstance(self.scope, FunctionScope) and name not in self.scope: # for each function or module scope above us for scope in self.scopeStack[:-1]: if not isinstance(scope, (FunctionScope, ModuleScope)): continue # if the name was defined in that scope, and the name has # been accessed already in the current scope, and hasn't # been declared global used = name in scope and scope[name].used if used and used[0] is self.scope and name not in self.scope.globals: # then it's probably a mistake self.report(messages.UndefinedLocal, scope[name].used[1], name, scope[name].source) break parent_stmt = self.getParent(node) if isinstance(parent_stmt, (FOR_TYPES, ast.comprehension)) or ( parent_stmt != node.parent and not self.isLiteralTupleUnpacking(parent_stmt)): binding = Binding(name, node) elif name == '__all__' and isinstance(self.scope, ModuleScope): binding = ExportBinding(name, node.parent, self.scope) elif isinstance(getattr(node, 'ctx', None), ast.Param): binding = Argument(name, self.getScopeNode(node)) else: binding = Assignment(name, node) self.addBinding(node, binding) def handleNodeDelete(self, node): def on_conditional_branch(): """ Return `True` if node is part of a conditional body. """ current = getattr(node, 'parent', None) while current: if isinstance(current, (ast.If, ast.While, ast.IfExp)): return True current = getattr(current, 'parent', None) return False name = getNodeName(node) if not name: return if on_conditional_branch(): # We cannot predict if this conditional branch is going to # be executed. return if isinstance(self.scope, FunctionScope) and name in self.scope.globals: self.scope.globals.remove(name) else: try: del self.scope[name] except KeyError: self.report(messages.UndefinedName, node, name) def _handle_type_comments(self, node): for (lineno, col_offset), comment in self._type_comments.get(node, ()): comment = comment.split(':', 1)[1].strip() func_match = TYPE_FUNC_RE.match(comment) if func_match: parts = ( func_match.group(1).replace('*', ''), func_match.group(2).strip(), ) else: parts = (comment,) for part in parts: if PY2: part = part.replace('...', 'Ellipsis') self.deferFunction(functools.partial( self.handleStringAnnotation, part, DummyNode(lineno, col_offset), lineno, col_offset, messages.CommentAnnotationSyntaxError, )) def handleChildren(self, tree, omit=None): self._handle_type_comments(tree) for node in iter_child_nodes(tree, omit=omit): self.handleNode(node, tree) def isLiteralTupleUnpacking(self, node): if isinstance(node, ast.Assign): for child in node.targets + [node.value]: if not hasattr(child, 'elts'): return False return True def isDocstring(self, node): """ Determine if the given node is a docstring, as long as it is at the correct place in the node tree. """ return isinstance(node, ast.Str) or (isinstance(node, ast.Expr) and isinstance(node.value, ast.Str)) def getDocstring(self, node): if isinstance(node, ast.Expr): node = node.value if not isinstance(node, ast.Str): return (None, None) if PYPY or PY38_PLUS: doctest_lineno = node.lineno - 1 else: # Computed incorrectly if the docstring has backslash doctest_lineno = node.lineno - node.s.count('\n') - 1 return (node.s, doctest_lineno) def handleNode(self, node, parent): if node is None: return if self.offset and getattr(node, 'lineno', None) is not None: node.lineno += self.offset[0] node.col_offset += self.offset[1] if self.traceTree: print(' ' * self.nodeDepth + node.__class__.__name__) if self.futuresAllowed and not (isinstance(node, ast.ImportFrom) or self.isDocstring(node)): self.futuresAllowed = False self.nodeDepth += 1 node.depth = self.nodeDepth node.parent = parent try: handler = self.getNodeHandler(node.__class__) handler(node) finally: self.nodeDepth -= 1 if self.traceTree: print(' ' * self.nodeDepth + 'end ' + node.__class__.__name__) _getDoctestExamples = doctest.DocTestParser().get_examples def handleDoctests(self, node): try: if hasattr(node, 'docstring'): docstring = node.docstring # This is just a reasonable guess. In Python 3.7, docstrings no # longer have line numbers associated with them. This will be # incorrect if there are empty lines between the beginning # of the function and the docstring. node_lineno = node.lineno if hasattr(node, 'args'): node_lineno = max([node_lineno] + [arg.lineno for arg in node.args.args]) else: (docstring, node_lineno) = self.getDocstring(node.body[0]) examples = docstring and self._getDoctestExamples(docstring) except (ValueError, IndexError): # e.g. line 6 of the docstring for <string> has inconsistent # leading whitespace: ... return if not examples: return # Place doctest in module scope saved_stack = self.scopeStack self.scopeStack = [self.scopeStack[0]] node_offset = self.offset or (0, 0) self.pushScope(DoctestScope) self.addBinding(None, Builtin('_')) for example in examples: try: tree = ast.parse(example.source, "<doctest>") except SyntaxError: e = sys.exc_info()[1] if PYPY: e.offset += 1 position = (node_lineno + example.lineno + e.lineno, example.indent + 4 + (e.offset or 0)) self.report(messages.DoctestSyntaxError, node, position) else: self.offset = (node_offset[0] + node_lineno + example.lineno, node_offset[1] + example.indent + 4) self.handleChildren(tree) self.offset = node_offset self.popScope() self.scopeStack = saved_stack def handleStringAnnotation(self, s, node, ref_lineno, ref_col_offset, err): try: tree = ast.parse(s) except SyntaxError: self.report(err, node, s) return body = tree.body if len(body) != 1 or not isinstance(body[0], ast.Expr): self.report(err, node, s) return parsed_annotation = tree.body[0].value for descendant in ast.walk(parsed_annotation): if ( 'lineno' in descendant._attributes and 'col_offset' in descendant._attributes ): descendant.lineno = ref_lineno descendant.col_offset = ref_col_offset self.handleNode(parsed_annotation, node) def handleAnnotation(self, annotation, node): if isinstance(annotation, ast.Str): # Defer handling forward annotation. self.deferFunction(functools.partial( self.handleStringAnnotation, annotation.s, node, annotation.lineno, annotation.col_offset, messages.ForwardAnnotationSyntaxError, )) elif self.annotationsFutureEnabled: self.deferFunction(lambda: self.handleNode(annotation, node)) else: self.handleNode(annotation, node) def ignore(self, node): pass # "stmt" type nodes DELETE = PRINT = FOR = ASYNCFOR = WHILE = IF = WITH = WITHITEM = \ ASYNCWITH = ASYNCWITHITEM = TRYFINALLY = EXEC = \ EXPR = ASSIGN = handleChildren PASS = ignore # "expr" type nodes BOOLOP = BINOP = UNARYOP = IFEXP = SET = \ CALL = REPR = ATTRIBUTE = SUBSCRIPT = \ STARRED = NAMECONSTANT = handleChildren NUM = STR = BYTES = ELLIPSIS = CONSTANT = ignore # "slice" type nodes SLICE = EXTSLICE = INDEX = handleChildren # expression contexts are node instances too, though being constants LOAD = STORE = DEL = AUGLOAD = AUGSTORE = PARAM = ignore # same for operators AND = OR = ADD = SUB = MULT = DIV = MOD = POW = LSHIFT = RSHIFT = \ BITOR = BITXOR = BITAND = FLOORDIV = INVERT = NOT = UADD = USUB = \ EQ = NOTEQ = LT = LTE = GT = GTE = IS = ISNOT = IN = NOTIN = \ MATMULT = ignore def RAISE(self, node): self.handleChildren(node) arg = get_raise_argument(node) if isinstance(arg, ast.Call): if is_notimplemented_name_node(arg.func): # Handle "raise NotImplemented(...)" self.report(messages.RaiseNotImplemented, node) elif is_notimplemented_name_node(arg): # Handle "raise NotImplemented" self.report(messages.RaiseNotImplemented, node) # additional node types COMPREHENSION = KEYWORD = FORMATTEDVALUE = JOINEDSTR = handleChildren def DICT(self, node): # Complain if there are duplicate keys with different values # If they have the same value it's not going to cause potentially # unexpected behaviour so we'll not complain. keys = [ convert_to_value(key) for key in node.keys ] key_counts = counter(keys) duplicate_keys = [ key for key, count in key_counts.items() if count > 1 ] for key in duplicate_keys: key_indices = [i for i, i_key in enumerate(keys) if i_key == key] values = counter( convert_to_value(node.values[index]) for index in key_indices ) if any(count == 1 for value, count in values.items()): for key_index in key_indices: key_node = node.keys[key_index] if isinstance(key, VariableKey): self.report(messages.MultiValueRepeatedKeyVariable, key_node, key.name) else: self.report( messages.MultiValueRepeatedKeyLiteral, key_node, key, ) self.handleChildren(node) def ASSERT(self, node): if isinstance(node.test, ast.Tuple) and node.test.elts != []: self.report(messages.AssertTuple, node) self.handleChildren(node) def GLOBAL(self, node): """ Keep track of globals declarations. """ global_scope_index = 1 if self._in_doctest() else 0 global_scope = self.scopeStack[global_scope_index] # Ignore 'global' statement in global scope. if self.scope is not global_scope: # One 'global' statement can bind multiple (comma-delimited) names. for node_name in node.names: node_value = Assignment(node_name, node) # Remove UndefinedName messages already reported for this name. # TODO: if the global is not used in this scope, it does not # become a globally defined name. See test_unused_global. self.messages = [ m for m in self.messages if not isinstance(m, messages.UndefinedName) or m.message_args[0] != node_name] # Bind name to global scope if it doesn't exist already. global_scope.setdefault(node_name, node_value) # Bind name to non-global scopes, but as already "used". node_value.used = (global_scope, node) for scope in self.scopeStack[global_scope_index + 1:]: scope[node_name] = node_value NONLOCAL = GLOBAL def GENERATOREXP(self, node): self.pushScope(GeneratorScope) self.handleChildren(node) self.popScope() LISTCOMP = handleChildren if PY2 else GENERATOREXP DICTCOMP = SETCOMP = GENERATOREXP def NAME(self, node): """ Handle occurrence of Name (which can be a load/store/delete access.) """ # Locate the name in locals / function / globals scopes. if isinstance(node.ctx, (ast.Load, ast.AugLoad)): self.handleNodeLoad(node) if (node.id == 'locals' and isinstance(self.scope, FunctionScope) and isinstance(node.parent, ast.Call)): # we are doing locals() call in current scope self.scope.usesLocals = True elif isinstance(node.ctx, (ast.Store, ast.AugStore, ast.Param)): self.handleNodeStore(node) elif isinstance(node.ctx, ast.Del): self.handleNodeDelete(node) else: # Unknown context raise RuntimeError("Got impossible expression context: %r" % (node.ctx,)) def CONTINUE(self, node): # Walk the tree up until we see a loop (OK), a function or class # definition (not OK), for 'continue', a finally block (not OK), or # the top module scope (not OK) n = node while hasattr(n, 'parent'): n, n_child = n.parent, n if isinstance(n, LOOP_TYPES): # Doesn't apply unless it's in the loop itself if n_child not in n.orelse: return if isinstance(n, (ast.FunctionDef, ast.ClassDef)): break # Handle Try/TryFinally difference in Python < and >= 3.3 if hasattr(n, 'finalbody') and isinstance(node, ast.Continue): if n_child in n.finalbody: self.report(messages.ContinueInFinally, node) return if isinstance(node, ast.Continue): self.report(messages.ContinueOutsideLoop, node) else: # ast.Break self.report(messages.BreakOutsideLoop, node) BREAK = CONTINUE def RETURN(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.ReturnOutsideFunction, node) return if ( node.value and hasattr(self.scope, 'returnValue') and not self.scope.returnValue ): self.scope.returnValue = node.value self.handleNode(node.value, node) def YIELD(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.YieldOutsideFunction, node) return self.scope.isGenerator = True self.handleNode(node.value, node) AWAIT = YIELDFROM = YIELD def FUNCTIONDEF(self, node): for deco in node.decorator_list: self.handleNode(deco, node) self.LAMBDA(node) self.addBinding(node, FunctionDefinition(node.name, node)) # doctest does not process doctest within a doctest, # or in nested functions. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) ASYNCFUNCTIONDEF = FUNCTIONDEF def LAMBDA(self, node): args = [] annotations = [] if PY2: def addArgs(arglist): for arg in arglist: if isinstance(arg, ast.Tuple): addArgs(arg.elts) else: args.append(arg.id) addArgs(node.args.args) defaults = node.args.defaults else: for arg in node.args.args + node.args.kwonlyargs: args.append(arg.arg) annotations.append(arg.annotation) defaults = node.args.defaults + node.args.kw_defaults # Only for Python3 FunctionDefs is_py3_func = hasattr(node, 'returns') for arg_name in ('vararg', 'kwarg'): wildcard = getattr(node.args, arg_name) if not wildcard: continue args.append(wildcard if PY2 else wildcard.arg) if is_py3_func: if PY2: # Python 2.7 argannotation = arg_name + 'annotation' annotations.append(getattr(node.args, argannotation)) else: # Python >= 3.4 annotations.append(wildcard.annotation) if is_py3_func: annotations.append(node.returns) if len(set(args)) < len(args): for (idx, arg) in enumerate(args): if arg in args[:idx]: self.report(messages.DuplicateArgument, node, arg) for annotation in annotations: self.handleAnnotation(annotation, node) for default in defaults: self.handleNode(default, node) def runFunction(): self.pushScope() self.handleChildren(node, omit='decorator_list') def checkUnusedAssignments(): """ Check to see if any assignments have not been used. """ for name, binding in self.scope.unusedAssignments(): self.report(messages.UnusedVariable, binding.source, name) self.deferAssignment(checkUnusedAssignments) if PY2: def checkReturnWithArgumentInsideGenerator(): """ Check to see if there is any return statement with arguments but the function is a generator. """ if self.scope.isGenerator and self.scope.returnValue: self.report(messages.ReturnWithArgsInsideGenerator, self.scope.returnValue) self.deferAssignment(checkReturnWithArgumentInsideGenerator) self.popScope() self.deferFunction(runFunction) def ARGUMENTS(self, node): self.handleChildren(node, omit=('defaults', 'kw_defaults')) if PY2: scope_node = self.getScopeNode(node) if node.vararg: self.addBinding(node, Argument(node.vararg, scope_node)) if node.kwarg: self.addBinding(node, Argument(node.kwarg, scope_node)) def ARG(self, node): self.addBinding(node, Argument(node.arg, self.getScopeNode(node))) def CLASSDEF(self, node): """ Check names used in a class definition, including its decorators, base classes, and the body of its definition. Additionally, add its name to the current scope. """ for deco in node.decorator_list: self.handleNode(deco, node) for baseNode in node.bases: self.handleNode(baseNode, node) if not PY2: for keywordNode in node.keywords: self.handleNode(keywordNode, node) self.pushScope(ClassScope) # doctest does not process doctest within a doctest # classes within classes are processed. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) for stmt in node.body: self.handleNode(stmt, node) self.popScope() self.addBinding(node, ClassDefinition(node.name, node)) def AUGASSIGN(self, node): self.handleNodeLoad(node.target) self.handleNode(node.value, node) self.handleNode(node.target, node) def TUPLE(self, node): if not PY2 and isinstance(node.ctx, ast.Store): # Python 3 advanced tuple unpacking: a, *b, c = d. # Only one starred expression is allowed, and no more than 1<<8 # assignments are allowed before a stared expression. There is # also a limit of 1<<24 expressions after the starred expression, # which is impossible to test due to memory restrictions, but we # add it here anyway has_starred = False star_loc = -1 for i, n in enumerate(node.elts): if isinstance(n, ast.Starred): if has_starred: self.report(messages.TwoStarredExpressions, node) # The SyntaxError doesn't distinguish two from more # than two. break has_starred = True star_loc = i if star_loc >= 1 << 8 or len(node.elts) - star_loc - 1 >= 1 << 24: self.report(messages.TooManyExpressionsInStarredAssignment, node) self.handleChildren(node) LIST = TUPLE def IMPORT(self, node): for alias in node.names: if '.' in alias.name and not alias.asname: importation = SubmoduleImportation(alias.name, node) else: name = alias.asname or alias.name importation = Importation(name, node, alias.name) self.addBinding(node, importation) def IMPORTFROM(self, node): if node.module == '__future__': if not self.futuresAllowed: self.report(messages.LateFutureImport, node, [n.name for n in node.names]) else: self.futuresAllowed = False module = ('.' * node.level) + (node.module or '') for alias in node.names: name = alias.asname or alias.name if node.module == '__future__': importation = FutureImportation(name, node, self.scope) if alias.name not in __future__.all_feature_names: self.report(messages.FutureFeatureNotDefined, node, alias.name) if alias.name == 'annotations': self.annotationsFutureEnabled = True elif alias.name == '*': # Only Python 2, local import * is a SyntaxWarning if not PY2 and not isinstance(self.scope, ModuleScope): self.report(messages.ImportStarNotPermitted, node, module) continue self.scope.importStarred = True self.report(messages.ImportStarUsed, node, module) importation = StarImportation(module, node) else: importation = ImportationFrom(name, node, module, alias.name) self.addBinding(node, importation) def TRY(self, node): handler_names = [] # List the exception handlers for i, handler in enumerate(node.handlers): if isinstance(handler.type, ast.Tuple): for exc_type in handler.type.elts: handler_names.append(getNodeName(exc_type)) elif handler.type: handler_names.append(getNodeName(handler.type)) if handler.type is None and i < len(node.handlers) - 1: self.report(messages.DefaultExceptNotLast, handler) # Memorize the except handlers and process the body self.exceptHandlers.append(handler_names) for child in node.body: self.handleNode(child, node) self.exceptHandlers.pop() # Process the other nodes: "except:", "else:", "finally:" self.handleChildren(node, omit='body') TRYEXCEPT = TRY def EXCEPTHANDLER(self, node): if PY2 or node.name is None: self.handleChildren(node) return # If the name already exists in the scope, modify state of existing # binding. if node.name in self.scope: self.handleNodeStore(node) # 3.x: the name of the exception, which is not a Name node, but a # simple string, creates a local that is only bound within the scope of # the except: block. As such, temporarily remove the existing binding # to more accurately determine if the name is used in the except: # block. try: prev_definition = self.scope.pop(node.name) except KeyError: prev_definition = None self.handleNodeStore(node) self.handleChildren(node) # See discussion on https://github.com/PyCQA/pyflakes/pull/59 # We're removing the local name since it's being unbound after leaving # the except: block and it's always unbound if the except: block is # never entered. This will cause an "undefined name" error raised if # the checked code tries to use the name afterwards. # # Unless it's been removed already. Then do nothing. try: binding = self.scope.pop(node.name) except KeyError: pass else: if not binding.used: self.report(messages.UnusedVariable, node, node.name) # Restore. if prev_definition: self.scope[node.name] = prev_definition def ANNASSIGN(self, node): if node.value: # Only bind the *targets* if the assignment has a value. # Otherwise it's not really ast.Store and shouldn't silence # UndefinedLocal warnings. self.handleNode(node.target, node) self.handleAnnotation(node.annotation, node) if node.value: # If the assignment has value, handle the *value* now. self.handleNode(node.value, node) def COMPARE(self, node): literals = (ast.Str, ast.Num) if not PY2: literals += (ast.Bytes,) left = node.left for op, right in zip(node.ops, node.comparators): if (isinstance(op, (ast.Is, ast.IsNot)) and (isinstance(left, literals) or isinstance(right, literals))): self.report(messages.IsLiteral, node) left = right self.handleChildren(node)
PyCQA/pyflakes
pyflakes/checker.py
Checker.deferAssignment
python
def deferAssignment(self, callable): self._deferredAssignments.append((callable, self.scopeStack[:], self.offset))
Schedule an assignment handler to be called just after deferred function handlers.
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/checker.py#L705-L710
null
class Checker(object): """ I check the cleanliness and sanity of Python code. @ivar _deferredFunctions: Tracking list used by L{deferFunction}. Elements of the list are two-tuples. The first element is the callable passed to L{deferFunction}. The second element is a copy of the scope stack at the time L{deferFunction} was called. @ivar _deferredAssignments: Similar to C{_deferredFunctions}, but for callables which are deferred assignment checks. """ _ast_node_scope = { ast.Module: ModuleScope, ast.ClassDef: ClassScope, ast.FunctionDef: FunctionScope, ast.Lambda: FunctionScope, ast.ListComp: GeneratorScope, ast.SetComp: GeneratorScope, ast.GeneratorExp: GeneratorScope, ast.DictComp: GeneratorScope, } if PY35_PLUS: _ast_node_scope[ast.AsyncFunctionDef] = FunctionScope, nodeDepth = 0 offset = None traceTree = False builtIns = set(builtin_vars).union(_MAGIC_GLOBALS) _customBuiltIns = os.environ.get('PYFLAKES_BUILTINS') if _customBuiltIns: builtIns.update(_customBuiltIns.split(',')) del _customBuiltIns # TODO: file_tokens= is required to perform checks on type comments, # eventually make this a required positional argument. For now it # is defaulted to `()` for api compatibility. def __init__(self, tree, filename='(none)', builtins=None, withDoctest='PYFLAKES_DOCTEST' in os.environ, file_tokens=()): self._nodeHandlers = {} self._deferredFunctions = [] self._deferredAssignments = [] self.deadScopes = [] self.messages = [] self.filename = filename if builtins: self.builtIns = self.builtIns.union(builtins) self.withDoctest = withDoctest try: self.scopeStack = [Checker._ast_node_scope[type(tree)]()] except KeyError: raise RuntimeError('No scope implemented for the node %r' % tree) self.exceptHandlers = [()] self.root = tree self._type_comments = _collect_type_comments(tree, file_tokens) for builtin in self.builtIns: self.addBinding(None, Builtin(builtin)) self.handleChildren(tree) self.runDeferred(self._deferredFunctions) # Set _deferredFunctions to None so that deferFunction will fail # noisily if called after we've run through the deferred functions. self._deferredFunctions = None self.runDeferred(self._deferredAssignments) # Set _deferredAssignments to None so that deferAssignment will fail # noisily if called after we've run through the deferred assignments. self._deferredAssignments = None del self.scopeStack[1:] self.popScope() self.checkDeadScopes() def deferFunction(self, callable): """ Schedule a function handler to be called just before completion. This is used for handling function bodies, which must be deferred because code later in the file might modify the global scope. When `callable` is called, the scope at the time this is called will be restored, however it will contain any new bindings added to it. """ self._deferredFunctions.append((callable, self.scopeStack[:], self.offset)) def runDeferred(self, deferred): """ Run the callables in C{deferred} using their associated scope stack. """ for handler, scope, offset in deferred: self.scopeStack = scope self.offset = offset handler() def _in_doctest(self): return (len(self.scopeStack) >= 2 and isinstance(self.scopeStack[1], DoctestScope)) @property def futuresAllowed(self): if not all(isinstance(scope, ModuleScope) for scope in self.scopeStack): return False return self.scope._futures_allowed @futuresAllowed.setter def futuresAllowed(self, value): assert value is False if isinstance(self.scope, ModuleScope): self.scope._futures_allowed = False @property def annotationsFutureEnabled(self): scope = self.scopeStack[0] if not isinstance(scope, ModuleScope): return False return scope._annotations_future_enabled @annotationsFutureEnabled.setter def annotationsFutureEnabled(self, value): assert value is True assert isinstance(self.scope, ModuleScope) self.scope._annotations_future_enabled = True @property def scope(self): return self.scopeStack[-1] def popScope(self): self.deadScopes.append(self.scopeStack.pop()) def checkDeadScopes(self): """ Look at scopes which have been fully examined and report names in them which were imported but unused. """ for scope in self.deadScopes: # imports in classes are public members if isinstance(scope, ClassScope): continue all_binding = scope.get('__all__') if all_binding and not isinstance(all_binding, ExportBinding): all_binding = None if all_binding: all_names = set(all_binding.names) undefined = all_names.difference(scope) else: all_names = undefined = [] if undefined: if not scope.importStarred and \ os.path.basename(self.filename) != '__init__.py': # Look for possible mistakes in the export list for name in undefined: self.report(messages.UndefinedExport, scope['__all__'].source, name) # mark all import '*' as used by the undefined in __all__ if scope.importStarred: from_list = [] for binding in scope.values(): if isinstance(binding, StarImportation): binding.used = all_binding from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) for name in undefined: self.report(messages.ImportStarUsage, scope['__all__'].source, name, from_list) # Look for imported names that aren't used. for value in scope.values(): if isinstance(value, Importation): used = value.used or value.name in all_names if not used: messg = messages.UnusedImport self.report(messg, value.source, str(value)) for node in value.redefined: if isinstance(self.getParent(node), FOR_TYPES): messg = messages.ImportShadowedByLoopVar elif used: continue else: messg = messages.RedefinedWhileUnused self.report(messg, node, value.name, value.source) def pushScope(self, scopeClass=FunctionScope): self.scopeStack.append(scopeClass()) def report(self, messageClass, *args, **kwargs): self.messages.append(messageClass(self.filename, *args, **kwargs)) def getParent(self, node): # Lookup the first parent which is not Tuple, List or Starred while True: node = node.parent if not hasattr(node, 'elts') and not hasattr(node, 'ctx'): return node def getCommonAncestor(self, lnode, rnode, stop): if stop in (lnode, rnode) or not (hasattr(lnode, 'parent') and hasattr(rnode, 'parent')): return None if lnode is rnode: return lnode if (lnode.depth > rnode.depth): return self.getCommonAncestor(lnode.parent, rnode, stop) if (lnode.depth < rnode.depth): return self.getCommonAncestor(lnode, rnode.parent, stop) return self.getCommonAncestor(lnode.parent, rnode.parent, stop) def descendantOf(self, node, ancestors, stop): for a in ancestors: if self.getCommonAncestor(node, a, stop): return True return False def _getAncestor(self, node, ancestor_type): parent = node while True: if parent is self.root: return None parent = self.getParent(parent) if isinstance(parent, ancestor_type): return parent def getScopeNode(self, node): return self._getAncestor(node, tuple(Checker._ast_node_scope.keys())) def differentForks(self, lnode, rnode): """True, if lnode and rnode are located on different forks of IF/TRY""" ancestor = self.getCommonAncestor(lnode, rnode, self.root) parts = getAlternatives(ancestor) if parts: for items in parts: if self.descendantOf(lnode, items, ancestor) ^ \ self.descendantOf(rnode, items, ancestor): return True return False def addBinding(self, node, value): """ Called when a binding is altered. - `node` is the statement responsible for the change - `value` is the new value, a Binding instance """ # assert value.source in (node, node.parent): for scope in self.scopeStack[::-1]: if value.name in scope: break existing = scope.get(value.name) if (existing and not isinstance(existing, Builtin) and not self.differentForks(node, existing.source)): parent_stmt = self.getParent(value.source) if isinstance(existing, Importation) and isinstance(parent_stmt, FOR_TYPES): self.report(messages.ImportShadowedByLoopVar, node, value.name, existing.source) elif scope is self.scope: if (isinstance(parent_stmt, ast.comprehension) and not isinstance(self.getParent(existing.source), (FOR_TYPES, ast.comprehension))): self.report(messages.RedefinedInListComp, node, value.name, existing.source) elif not existing.used and value.redefines(existing): if value.name != '_' or isinstance(existing, Importation): if not is_typing_overload(existing, self.scopeStack): self.report(messages.RedefinedWhileUnused, node, value.name, existing.source) elif isinstance(existing, Importation) and value.redefines(existing): existing.redefined.append(node) if value.name in self.scope: # then assume the rebound name is used as a global or within a loop value.used = self.scope[value.name].used self.scope[value.name] = value def getNodeHandler(self, node_class): try: return self._nodeHandlers[node_class] except KeyError: nodeType = getNodeType(node_class) self._nodeHandlers[node_class] = handler = getattr(self, nodeType) return handler def handleNodeLoad(self, node): name = getNodeName(node) if not name: return in_generators = None importStarred = None # try enclosing function scopes and global scope for scope in self.scopeStack[-1::-1]: if isinstance(scope, ClassScope): if not PY2 and name == '__class__': return elif in_generators is False: # only generators used in a class scope can access the # names of the class. this is skipped during the first # iteration continue if (name == 'print' and isinstance(scope.get(name, None), Builtin)): parent = self.getParent(node) if (isinstance(parent, ast.BinOp) and isinstance(parent.op, ast.RShift)): self.report(messages.InvalidPrintSyntax, node) try: scope[name].used = (self.scope, node) # if the name of SubImportation is same as # alias of other Importation and the alias # is used, SubImportation also should be marked as used. n = scope[name] if isinstance(n, Importation) and n._has_alias(): try: scope[n.fullName].used = (self.scope, node) except KeyError: pass except KeyError: pass else: return importStarred = importStarred or scope.importStarred if in_generators is not False: in_generators = isinstance(scope, GeneratorScope) if importStarred: from_list = [] for scope in self.scopeStack[-1::-1]: for binding in scope.values(): if isinstance(binding, StarImportation): # mark '*' imports as used for each scope binding.used = (self.scope, node) from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) self.report(messages.ImportStarUsage, node, name, from_list) return if name == '__path__' and os.path.basename(self.filename) == '__init__.py': # the special name __path__ is valid only in packages return if name == '__module__' and isinstance(self.scope, ClassScope): return # protected with a NameError handler? if 'NameError' not in self.exceptHandlers[-1]: self.report(messages.UndefinedName, node, name) def handleNodeStore(self, node): name = getNodeName(node) if not name: return # if the name hasn't already been defined in the current scope if isinstance(self.scope, FunctionScope) and name not in self.scope: # for each function or module scope above us for scope in self.scopeStack[:-1]: if not isinstance(scope, (FunctionScope, ModuleScope)): continue # if the name was defined in that scope, and the name has # been accessed already in the current scope, and hasn't # been declared global used = name in scope and scope[name].used if used and used[0] is self.scope and name not in self.scope.globals: # then it's probably a mistake self.report(messages.UndefinedLocal, scope[name].used[1], name, scope[name].source) break parent_stmt = self.getParent(node) if isinstance(parent_stmt, (FOR_TYPES, ast.comprehension)) or ( parent_stmt != node.parent and not self.isLiteralTupleUnpacking(parent_stmt)): binding = Binding(name, node) elif name == '__all__' and isinstance(self.scope, ModuleScope): binding = ExportBinding(name, node.parent, self.scope) elif isinstance(getattr(node, 'ctx', None), ast.Param): binding = Argument(name, self.getScopeNode(node)) else: binding = Assignment(name, node) self.addBinding(node, binding) def handleNodeDelete(self, node): def on_conditional_branch(): """ Return `True` if node is part of a conditional body. """ current = getattr(node, 'parent', None) while current: if isinstance(current, (ast.If, ast.While, ast.IfExp)): return True current = getattr(current, 'parent', None) return False name = getNodeName(node) if not name: return if on_conditional_branch(): # We cannot predict if this conditional branch is going to # be executed. return if isinstance(self.scope, FunctionScope) and name in self.scope.globals: self.scope.globals.remove(name) else: try: del self.scope[name] except KeyError: self.report(messages.UndefinedName, node, name) def _handle_type_comments(self, node): for (lineno, col_offset), comment in self._type_comments.get(node, ()): comment = comment.split(':', 1)[1].strip() func_match = TYPE_FUNC_RE.match(comment) if func_match: parts = ( func_match.group(1).replace('*', ''), func_match.group(2).strip(), ) else: parts = (comment,) for part in parts: if PY2: part = part.replace('...', 'Ellipsis') self.deferFunction(functools.partial( self.handleStringAnnotation, part, DummyNode(lineno, col_offset), lineno, col_offset, messages.CommentAnnotationSyntaxError, )) def handleChildren(self, tree, omit=None): self._handle_type_comments(tree) for node in iter_child_nodes(tree, omit=omit): self.handleNode(node, tree) def isLiteralTupleUnpacking(self, node): if isinstance(node, ast.Assign): for child in node.targets + [node.value]: if not hasattr(child, 'elts'): return False return True def isDocstring(self, node): """ Determine if the given node is a docstring, as long as it is at the correct place in the node tree. """ return isinstance(node, ast.Str) or (isinstance(node, ast.Expr) and isinstance(node.value, ast.Str)) def getDocstring(self, node): if isinstance(node, ast.Expr): node = node.value if not isinstance(node, ast.Str): return (None, None) if PYPY or PY38_PLUS: doctest_lineno = node.lineno - 1 else: # Computed incorrectly if the docstring has backslash doctest_lineno = node.lineno - node.s.count('\n') - 1 return (node.s, doctest_lineno) def handleNode(self, node, parent): if node is None: return if self.offset and getattr(node, 'lineno', None) is not None: node.lineno += self.offset[0] node.col_offset += self.offset[1] if self.traceTree: print(' ' * self.nodeDepth + node.__class__.__name__) if self.futuresAllowed and not (isinstance(node, ast.ImportFrom) or self.isDocstring(node)): self.futuresAllowed = False self.nodeDepth += 1 node.depth = self.nodeDepth node.parent = parent try: handler = self.getNodeHandler(node.__class__) handler(node) finally: self.nodeDepth -= 1 if self.traceTree: print(' ' * self.nodeDepth + 'end ' + node.__class__.__name__) _getDoctestExamples = doctest.DocTestParser().get_examples def handleDoctests(self, node): try: if hasattr(node, 'docstring'): docstring = node.docstring # This is just a reasonable guess. In Python 3.7, docstrings no # longer have line numbers associated with them. This will be # incorrect if there are empty lines between the beginning # of the function and the docstring. node_lineno = node.lineno if hasattr(node, 'args'): node_lineno = max([node_lineno] + [arg.lineno for arg in node.args.args]) else: (docstring, node_lineno) = self.getDocstring(node.body[0]) examples = docstring and self._getDoctestExamples(docstring) except (ValueError, IndexError): # e.g. line 6 of the docstring for <string> has inconsistent # leading whitespace: ... return if not examples: return # Place doctest in module scope saved_stack = self.scopeStack self.scopeStack = [self.scopeStack[0]] node_offset = self.offset or (0, 0) self.pushScope(DoctestScope) self.addBinding(None, Builtin('_')) for example in examples: try: tree = ast.parse(example.source, "<doctest>") except SyntaxError: e = sys.exc_info()[1] if PYPY: e.offset += 1 position = (node_lineno + example.lineno + e.lineno, example.indent + 4 + (e.offset or 0)) self.report(messages.DoctestSyntaxError, node, position) else: self.offset = (node_offset[0] + node_lineno + example.lineno, node_offset[1] + example.indent + 4) self.handleChildren(tree) self.offset = node_offset self.popScope() self.scopeStack = saved_stack def handleStringAnnotation(self, s, node, ref_lineno, ref_col_offset, err): try: tree = ast.parse(s) except SyntaxError: self.report(err, node, s) return body = tree.body if len(body) != 1 or not isinstance(body[0], ast.Expr): self.report(err, node, s) return parsed_annotation = tree.body[0].value for descendant in ast.walk(parsed_annotation): if ( 'lineno' in descendant._attributes and 'col_offset' in descendant._attributes ): descendant.lineno = ref_lineno descendant.col_offset = ref_col_offset self.handleNode(parsed_annotation, node) def handleAnnotation(self, annotation, node): if isinstance(annotation, ast.Str): # Defer handling forward annotation. self.deferFunction(functools.partial( self.handleStringAnnotation, annotation.s, node, annotation.lineno, annotation.col_offset, messages.ForwardAnnotationSyntaxError, )) elif self.annotationsFutureEnabled: self.deferFunction(lambda: self.handleNode(annotation, node)) else: self.handleNode(annotation, node) def ignore(self, node): pass # "stmt" type nodes DELETE = PRINT = FOR = ASYNCFOR = WHILE = IF = WITH = WITHITEM = \ ASYNCWITH = ASYNCWITHITEM = TRYFINALLY = EXEC = \ EXPR = ASSIGN = handleChildren PASS = ignore # "expr" type nodes BOOLOP = BINOP = UNARYOP = IFEXP = SET = \ CALL = REPR = ATTRIBUTE = SUBSCRIPT = \ STARRED = NAMECONSTANT = handleChildren NUM = STR = BYTES = ELLIPSIS = CONSTANT = ignore # "slice" type nodes SLICE = EXTSLICE = INDEX = handleChildren # expression contexts are node instances too, though being constants LOAD = STORE = DEL = AUGLOAD = AUGSTORE = PARAM = ignore # same for operators AND = OR = ADD = SUB = MULT = DIV = MOD = POW = LSHIFT = RSHIFT = \ BITOR = BITXOR = BITAND = FLOORDIV = INVERT = NOT = UADD = USUB = \ EQ = NOTEQ = LT = LTE = GT = GTE = IS = ISNOT = IN = NOTIN = \ MATMULT = ignore def RAISE(self, node): self.handleChildren(node) arg = get_raise_argument(node) if isinstance(arg, ast.Call): if is_notimplemented_name_node(arg.func): # Handle "raise NotImplemented(...)" self.report(messages.RaiseNotImplemented, node) elif is_notimplemented_name_node(arg): # Handle "raise NotImplemented" self.report(messages.RaiseNotImplemented, node) # additional node types COMPREHENSION = KEYWORD = FORMATTEDVALUE = JOINEDSTR = handleChildren def DICT(self, node): # Complain if there are duplicate keys with different values # If they have the same value it's not going to cause potentially # unexpected behaviour so we'll not complain. keys = [ convert_to_value(key) for key in node.keys ] key_counts = counter(keys) duplicate_keys = [ key for key, count in key_counts.items() if count > 1 ] for key in duplicate_keys: key_indices = [i for i, i_key in enumerate(keys) if i_key == key] values = counter( convert_to_value(node.values[index]) for index in key_indices ) if any(count == 1 for value, count in values.items()): for key_index in key_indices: key_node = node.keys[key_index] if isinstance(key, VariableKey): self.report(messages.MultiValueRepeatedKeyVariable, key_node, key.name) else: self.report( messages.MultiValueRepeatedKeyLiteral, key_node, key, ) self.handleChildren(node) def ASSERT(self, node): if isinstance(node.test, ast.Tuple) and node.test.elts != []: self.report(messages.AssertTuple, node) self.handleChildren(node) def GLOBAL(self, node): """ Keep track of globals declarations. """ global_scope_index = 1 if self._in_doctest() else 0 global_scope = self.scopeStack[global_scope_index] # Ignore 'global' statement in global scope. if self.scope is not global_scope: # One 'global' statement can bind multiple (comma-delimited) names. for node_name in node.names: node_value = Assignment(node_name, node) # Remove UndefinedName messages already reported for this name. # TODO: if the global is not used in this scope, it does not # become a globally defined name. See test_unused_global. self.messages = [ m for m in self.messages if not isinstance(m, messages.UndefinedName) or m.message_args[0] != node_name] # Bind name to global scope if it doesn't exist already. global_scope.setdefault(node_name, node_value) # Bind name to non-global scopes, but as already "used". node_value.used = (global_scope, node) for scope in self.scopeStack[global_scope_index + 1:]: scope[node_name] = node_value NONLOCAL = GLOBAL def GENERATOREXP(self, node): self.pushScope(GeneratorScope) self.handleChildren(node) self.popScope() LISTCOMP = handleChildren if PY2 else GENERATOREXP DICTCOMP = SETCOMP = GENERATOREXP def NAME(self, node): """ Handle occurrence of Name (which can be a load/store/delete access.) """ # Locate the name in locals / function / globals scopes. if isinstance(node.ctx, (ast.Load, ast.AugLoad)): self.handleNodeLoad(node) if (node.id == 'locals' and isinstance(self.scope, FunctionScope) and isinstance(node.parent, ast.Call)): # we are doing locals() call in current scope self.scope.usesLocals = True elif isinstance(node.ctx, (ast.Store, ast.AugStore, ast.Param)): self.handleNodeStore(node) elif isinstance(node.ctx, ast.Del): self.handleNodeDelete(node) else: # Unknown context raise RuntimeError("Got impossible expression context: %r" % (node.ctx,)) def CONTINUE(self, node): # Walk the tree up until we see a loop (OK), a function or class # definition (not OK), for 'continue', a finally block (not OK), or # the top module scope (not OK) n = node while hasattr(n, 'parent'): n, n_child = n.parent, n if isinstance(n, LOOP_TYPES): # Doesn't apply unless it's in the loop itself if n_child not in n.orelse: return if isinstance(n, (ast.FunctionDef, ast.ClassDef)): break # Handle Try/TryFinally difference in Python < and >= 3.3 if hasattr(n, 'finalbody') and isinstance(node, ast.Continue): if n_child in n.finalbody: self.report(messages.ContinueInFinally, node) return if isinstance(node, ast.Continue): self.report(messages.ContinueOutsideLoop, node) else: # ast.Break self.report(messages.BreakOutsideLoop, node) BREAK = CONTINUE def RETURN(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.ReturnOutsideFunction, node) return if ( node.value and hasattr(self.scope, 'returnValue') and not self.scope.returnValue ): self.scope.returnValue = node.value self.handleNode(node.value, node) def YIELD(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.YieldOutsideFunction, node) return self.scope.isGenerator = True self.handleNode(node.value, node) AWAIT = YIELDFROM = YIELD def FUNCTIONDEF(self, node): for deco in node.decorator_list: self.handleNode(deco, node) self.LAMBDA(node) self.addBinding(node, FunctionDefinition(node.name, node)) # doctest does not process doctest within a doctest, # or in nested functions. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) ASYNCFUNCTIONDEF = FUNCTIONDEF def LAMBDA(self, node): args = [] annotations = [] if PY2: def addArgs(arglist): for arg in arglist: if isinstance(arg, ast.Tuple): addArgs(arg.elts) else: args.append(arg.id) addArgs(node.args.args) defaults = node.args.defaults else: for arg in node.args.args + node.args.kwonlyargs: args.append(arg.arg) annotations.append(arg.annotation) defaults = node.args.defaults + node.args.kw_defaults # Only for Python3 FunctionDefs is_py3_func = hasattr(node, 'returns') for arg_name in ('vararg', 'kwarg'): wildcard = getattr(node.args, arg_name) if not wildcard: continue args.append(wildcard if PY2 else wildcard.arg) if is_py3_func: if PY2: # Python 2.7 argannotation = arg_name + 'annotation' annotations.append(getattr(node.args, argannotation)) else: # Python >= 3.4 annotations.append(wildcard.annotation) if is_py3_func: annotations.append(node.returns) if len(set(args)) < len(args): for (idx, arg) in enumerate(args): if arg in args[:idx]: self.report(messages.DuplicateArgument, node, arg) for annotation in annotations: self.handleAnnotation(annotation, node) for default in defaults: self.handleNode(default, node) def runFunction(): self.pushScope() self.handleChildren(node, omit='decorator_list') def checkUnusedAssignments(): """ Check to see if any assignments have not been used. """ for name, binding in self.scope.unusedAssignments(): self.report(messages.UnusedVariable, binding.source, name) self.deferAssignment(checkUnusedAssignments) if PY2: def checkReturnWithArgumentInsideGenerator(): """ Check to see if there is any return statement with arguments but the function is a generator. """ if self.scope.isGenerator and self.scope.returnValue: self.report(messages.ReturnWithArgsInsideGenerator, self.scope.returnValue) self.deferAssignment(checkReturnWithArgumentInsideGenerator) self.popScope() self.deferFunction(runFunction) def ARGUMENTS(self, node): self.handleChildren(node, omit=('defaults', 'kw_defaults')) if PY2: scope_node = self.getScopeNode(node) if node.vararg: self.addBinding(node, Argument(node.vararg, scope_node)) if node.kwarg: self.addBinding(node, Argument(node.kwarg, scope_node)) def ARG(self, node): self.addBinding(node, Argument(node.arg, self.getScopeNode(node))) def CLASSDEF(self, node): """ Check names used in a class definition, including its decorators, base classes, and the body of its definition. Additionally, add its name to the current scope. """ for deco in node.decorator_list: self.handleNode(deco, node) for baseNode in node.bases: self.handleNode(baseNode, node) if not PY2: for keywordNode in node.keywords: self.handleNode(keywordNode, node) self.pushScope(ClassScope) # doctest does not process doctest within a doctest # classes within classes are processed. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) for stmt in node.body: self.handleNode(stmt, node) self.popScope() self.addBinding(node, ClassDefinition(node.name, node)) def AUGASSIGN(self, node): self.handleNodeLoad(node.target) self.handleNode(node.value, node) self.handleNode(node.target, node) def TUPLE(self, node): if not PY2 and isinstance(node.ctx, ast.Store): # Python 3 advanced tuple unpacking: a, *b, c = d. # Only one starred expression is allowed, and no more than 1<<8 # assignments are allowed before a stared expression. There is # also a limit of 1<<24 expressions after the starred expression, # which is impossible to test due to memory restrictions, but we # add it here anyway has_starred = False star_loc = -1 for i, n in enumerate(node.elts): if isinstance(n, ast.Starred): if has_starred: self.report(messages.TwoStarredExpressions, node) # The SyntaxError doesn't distinguish two from more # than two. break has_starred = True star_loc = i if star_loc >= 1 << 8 or len(node.elts) - star_loc - 1 >= 1 << 24: self.report(messages.TooManyExpressionsInStarredAssignment, node) self.handleChildren(node) LIST = TUPLE def IMPORT(self, node): for alias in node.names: if '.' in alias.name and not alias.asname: importation = SubmoduleImportation(alias.name, node) else: name = alias.asname or alias.name importation = Importation(name, node, alias.name) self.addBinding(node, importation) def IMPORTFROM(self, node): if node.module == '__future__': if not self.futuresAllowed: self.report(messages.LateFutureImport, node, [n.name for n in node.names]) else: self.futuresAllowed = False module = ('.' * node.level) + (node.module or '') for alias in node.names: name = alias.asname or alias.name if node.module == '__future__': importation = FutureImportation(name, node, self.scope) if alias.name not in __future__.all_feature_names: self.report(messages.FutureFeatureNotDefined, node, alias.name) if alias.name == 'annotations': self.annotationsFutureEnabled = True elif alias.name == '*': # Only Python 2, local import * is a SyntaxWarning if not PY2 and not isinstance(self.scope, ModuleScope): self.report(messages.ImportStarNotPermitted, node, module) continue self.scope.importStarred = True self.report(messages.ImportStarUsed, node, module) importation = StarImportation(module, node) else: importation = ImportationFrom(name, node, module, alias.name) self.addBinding(node, importation) def TRY(self, node): handler_names = [] # List the exception handlers for i, handler in enumerate(node.handlers): if isinstance(handler.type, ast.Tuple): for exc_type in handler.type.elts: handler_names.append(getNodeName(exc_type)) elif handler.type: handler_names.append(getNodeName(handler.type)) if handler.type is None and i < len(node.handlers) - 1: self.report(messages.DefaultExceptNotLast, handler) # Memorize the except handlers and process the body self.exceptHandlers.append(handler_names) for child in node.body: self.handleNode(child, node) self.exceptHandlers.pop() # Process the other nodes: "except:", "else:", "finally:" self.handleChildren(node, omit='body') TRYEXCEPT = TRY def EXCEPTHANDLER(self, node): if PY2 or node.name is None: self.handleChildren(node) return # If the name already exists in the scope, modify state of existing # binding. if node.name in self.scope: self.handleNodeStore(node) # 3.x: the name of the exception, which is not a Name node, but a # simple string, creates a local that is only bound within the scope of # the except: block. As such, temporarily remove the existing binding # to more accurately determine if the name is used in the except: # block. try: prev_definition = self.scope.pop(node.name) except KeyError: prev_definition = None self.handleNodeStore(node) self.handleChildren(node) # See discussion on https://github.com/PyCQA/pyflakes/pull/59 # We're removing the local name since it's being unbound after leaving # the except: block and it's always unbound if the except: block is # never entered. This will cause an "undefined name" error raised if # the checked code tries to use the name afterwards. # # Unless it's been removed already. Then do nothing. try: binding = self.scope.pop(node.name) except KeyError: pass else: if not binding.used: self.report(messages.UnusedVariable, node, node.name) # Restore. if prev_definition: self.scope[node.name] = prev_definition def ANNASSIGN(self, node): if node.value: # Only bind the *targets* if the assignment has a value. # Otherwise it's not really ast.Store and shouldn't silence # UndefinedLocal warnings. self.handleNode(node.target, node) self.handleAnnotation(node.annotation, node) if node.value: # If the assignment has value, handle the *value* now. self.handleNode(node.value, node) def COMPARE(self, node): literals = (ast.Str, ast.Num) if not PY2: literals += (ast.Bytes,) left = node.left for op, right in zip(node.ops, node.comparators): if (isinstance(op, (ast.Is, ast.IsNot)) and (isinstance(left, literals) or isinstance(right, literals))): self.report(messages.IsLiteral, node) left = right self.handleChildren(node)
PyCQA/pyflakes
pyflakes/checker.py
Checker.runDeferred
python
def runDeferred(self, deferred): for handler, scope, offset in deferred: self.scopeStack = scope self.offset = offset handler()
Run the callables in C{deferred} using their associated scope stack.
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/checker.py#L712-L719
null
class Checker(object): """ I check the cleanliness and sanity of Python code. @ivar _deferredFunctions: Tracking list used by L{deferFunction}. Elements of the list are two-tuples. The first element is the callable passed to L{deferFunction}. The second element is a copy of the scope stack at the time L{deferFunction} was called. @ivar _deferredAssignments: Similar to C{_deferredFunctions}, but for callables which are deferred assignment checks. """ _ast_node_scope = { ast.Module: ModuleScope, ast.ClassDef: ClassScope, ast.FunctionDef: FunctionScope, ast.Lambda: FunctionScope, ast.ListComp: GeneratorScope, ast.SetComp: GeneratorScope, ast.GeneratorExp: GeneratorScope, ast.DictComp: GeneratorScope, } if PY35_PLUS: _ast_node_scope[ast.AsyncFunctionDef] = FunctionScope, nodeDepth = 0 offset = None traceTree = False builtIns = set(builtin_vars).union(_MAGIC_GLOBALS) _customBuiltIns = os.environ.get('PYFLAKES_BUILTINS') if _customBuiltIns: builtIns.update(_customBuiltIns.split(',')) del _customBuiltIns # TODO: file_tokens= is required to perform checks on type comments, # eventually make this a required positional argument. For now it # is defaulted to `()` for api compatibility. def __init__(self, tree, filename='(none)', builtins=None, withDoctest='PYFLAKES_DOCTEST' in os.environ, file_tokens=()): self._nodeHandlers = {} self._deferredFunctions = [] self._deferredAssignments = [] self.deadScopes = [] self.messages = [] self.filename = filename if builtins: self.builtIns = self.builtIns.union(builtins) self.withDoctest = withDoctest try: self.scopeStack = [Checker._ast_node_scope[type(tree)]()] except KeyError: raise RuntimeError('No scope implemented for the node %r' % tree) self.exceptHandlers = [()] self.root = tree self._type_comments = _collect_type_comments(tree, file_tokens) for builtin in self.builtIns: self.addBinding(None, Builtin(builtin)) self.handleChildren(tree) self.runDeferred(self._deferredFunctions) # Set _deferredFunctions to None so that deferFunction will fail # noisily if called after we've run through the deferred functions. self._deferredFunctions = None self.runDeferred(self._deferredAssignments) # Set _deferredAssignments to None so that deferAssignment will fail # noisily if called after we've run through the deferred assignments. self._deferredAssignments = None del self.scopeStack[1:] self.popScope() self.checkDeadScopes() def deferFunction(self, callable): """ Schedule a function handler to be called just before completion. This is used for handling function bodies, which must be deferred because code later in the file might modify the global scope. When `callable` is called, the scope at the time this is called will be restored, however it will contain any new bindings added to it. """ self._deferredFunctions.append((callable, self.scopeStack[:], self.offset)) def deferAssignment(self, callable): """ Schedule an assignment handler to be called just after deferred function handlers. """ self._deferredAssignments.append((callable, self.scopeStack[:], self.offset)) def _in_doctest(self): return (len(self.scopeStack) >= 2 and isinstance(self.scopeStack[1], DoctestScope)) @property def futuresAllowed(self): if not all(isinstance(scope, ModuleScope) for scope in self.scopeStack): return False return self.scope._futures_allowed @futuresAllowed.setter def futuresAllowed(self, value): assert value is False if isinstance(self.scope, ModuleScope): self.scope._futures_allowed = False @property def annotationsFutureEnabled(self): scope = self.scopeStack[0] if not isinstance(scope, ModuleScope): return False return scope._annotations_future_enabled @annotationsFutureEnabled.setter def annotationsFutureEnabled(self, value): assert value is True assert isinstance(self.scope, ModuleScope) self.scope._annotations_future_enabled = True @property def scope(self): return self.scopeStack[-1] def popScope(self): self.deadScopes.append(self.scopeStack.pop()) def checkDeadScopes(self): """ Look at scopes which have been fully examined and report names in them which were imported but unused. """ for scope in self.deadScopes: # imports in classes are public members if isinstance(scope, ClassScope): continue all_binding = scope.get('__all__') if all_binding and not isinstance(all_binding, ExportBinding): all_binding = None if all_binding: all_names = set(all_binding.names) undefined = all_names.difference(scope) else: all_names = undefined = [] if undefined: if not scope.importStarred and \ os.path.basename(self.filename) != '__init__.py': # Look for possible mistakes in the export list for name in undefined: self.report(messages.UndefinedExport, scope['__all__'].source, name) # mark all import '*' as used by the undefined in __all__ if scope.importStarred: from_list = [] for binding in scope.values(): if isinstance(binding, StarImportation): binding.used = all_binding from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) for name in undefined: self.report(messages.ImportStarUsage, scope['__all__'].source, name, from_list) # Look for imported names that aren't used. for value in scope.values(): if isinstance(value, Importation): used = value.used or value.name in all_names if not used: messg = messages.UnusedImport self.report(messg, value.source, str(value)) for node in value.redefined: if isinstance(self.getParent(node), FOR_TYPES): messg = messages.ImportShadowedByLoopVar elif used: continue else: messg = messages.RedefinedWhileUnused self.report(messg, node, value.name, value.source) def pushScope(self, scopeClass=FunctionScope): self.scopeStack.append(scopeClass()) def report(self, messageClass, *args, **kwargs): self.messages.append(messageClass(self.filename, *args, **kwargs)) def getParent(self, node): # Lookup the first parent which is not Tuple, List or Starred while True: node = node.parent if not hasattr(node, 'elts') and not hasattr(node, 'ctx'): return node def getCommonAncestor(self, lnode, rnode, stop): if stop in (lnode, rnode) or not (hasattr(lnode, 'parent') and hasattr(rnode, 'parent')): return None if lnode is rnode: return lnode if (lnode.depth > rnode.depth): return self.getCommonAncestor(lnode.parent, rnode, stop) if (lnode.depth < rnode.depth): return self.getCommonAncestor(lnode, rnode.parent, stop) return self.getCommonAncestor(lnode.parent, rnode.parent, stop) def descendantOf(self, node, ancestors, stop): for a in ancestors: if self.getCommonAncestor(node, a, stop): return True return False def _getAncestor(self, node, ancestor_type): parent = node while True: if parent is self.root: return None parent = self.getParent(parent) if isinstance(parent, ancestor_type): return parent def getScopeNode(self, node): return self._getAncestor(node, tuple(Checker._ast_node_scope.keys())) def differentForks(self, lnode, rnode): """True, if lnode and rnode are located on different forks of IF/TRY""" ancestor = self.getCommonAncestor(lnode, rnode, self.root) parts = getAlternatives(ancestor) if parts: for items in parts: if self.descendantOf(lnode, items, ancestor) ^ \ self.descendantOf(rnode, items, ancestor): return True return False def addBinding(self, node, value): """ Called when a binding is altered. - `node` is the statement responsible for the change - `value` is the new value, a Binding instance """ # assert value.source in (node, node.parent): for scope in self.scopeStack[::-1]: if value.name in scope: break existing = scope.get(value.name) if (existing and not isinstance(existing, Builtin) and not self.differentForks(node, existing.source)): parent_stmt = self.getParent(value.source) if isinstance(existing, Importation) and isinstance(parent_stmt, FOR_TYPES): self.report(messages.ImportShadowedByLoopVar, node, value.name, existing.source) elif scope is self.scope: if (isinstance(parent_stmt, ast.comprehension) and not isinstance(self.getParent(existing.source), (FOR_TYPES, ast.comprehension))): self.report(messages.RedefinedInListComp, node, value.name, existing.source) elif not existing.used and value.redefines(existing): if value.name != '_' or isinstance(existing, Importation): if not is_typing_overload(existing, self.scopeStack): self.report(messages.RedefinedWhileUnused, node, value.name, existing.source) elif isinstance(existing, Importation) and value.redefines(existing): existing.redefined.append(node) if value.name in self.scope: # then assume the rebound name is used as a global or within a loop value.used = self.scope[value.name].used self.scope[value.name] = value def getNodeHandler(self, node_class): try: return self._nodeHandlers[node_class] except KeyError: nodeType = getNodeType(node_class) self._nodeHandlers[node_class] = handler = getattr(self, nodeType) return handler def handleNodeLoad(self, node): name = getNodeName(node) if not name: return in_generators = None importStarred = None # try enclosing function scopes and global scope for scope in self.scopeStack[-1::-1]: if isinstance(scope, ClassScope): if not PY2 and name == '__class__': return elif in_generators is False: # only generators used in a class scope can access the # names of the class. this is skipped during the first # iteration continue if (name == 'print' and isinstance(scope.get(name, None), Builtin)): parent = self.getParent(node) if (isinstance(parent, ast.BinOp) and isinstance(parent.op, ast.RShift)): self.report(messages.InvalidPrintSyntax, node) try: scope[name].used = (self.scope, node) # if the name of SubImportation is same as # alias of other Importation and the alias # is used, SubImportation also should be marked as used. n = scope[name] if isinstance(n, Importation) and n._has_alias(): try: scope[n.fullName].used = (self.scope, node) except KeyError: pass except KeyError: pass else: return importStarred = importStarred or scope.importStarred if in_generators is not False: in_generators = isinstance(scope, GeneratorScope) if importStarred: from_list = [] for scope in self.scopeStack[-1::-1]: for binding in scope.values(): if isinstance(binding, StarImportation): # mark '*' imports as used for each scope binding.used = (self.scope, node) from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) self.report(messages.ImportStarUsage, node, name, from_list) return if name == '__path__' and os.path.basename(self.filename) == '__init__.py': # the special name __path__ is valid only in packages return if name == '__module__' and isinstance(self.scope, ClassScope): return # protected with a NameError handler? if 'NameError' not in self.exceptHandlers[-1]: self.report(messages.UndefinedName, node, name) def handleNodeStore(self, node): name = getNodeName(node) if not name: return # if the name hasn't already been defined in the current scope if isinstance(self.scope, FunctionScope) and name not in self.scope: # for each function or module scope above us for scope in self.scopeStack[:-1]: if not isinstance(scope, (FunctionScope, ModuleScope)): continue # if the name was defined in that scope, and the name has # been accessed already in the current scope, and hasn't # been declared global used = name in scope and scope[name].used if used and used[0] is self.scope and name not in self.scope.globals: # then it's probably a mistake self.report(messages.UndefinedLocal, scope[name].used[1], name, scope[name].source) break parent_stmt = self.getParent(node) if isinstance(parent_stmt, (FOR_TYPES, ast.comprehension)) or ( parent_stmt != node.parent and not self.isLiteralTupleUnpacking(parent_stmt)): binding = Binding(name, node) elif name == '__all__' and isinstance(self.scope, ModuleScope): binding = ExportBinding(name, node.parent, self.scope) elif isinstance(getattr(node, 'ctx', None), ast.Param): binding = Argument(name, self.getScopeNode(node)) else: binding = Assignment(name, node) self.addBinding(node, binding) def handleNodeDelete(self, node): def on_conditional_branch(): """ Return `True` if node is part of a conditional body. """ current = getattr(node, 'parent', None) while current: if isinstance(current, (ast.If, ast.While, ast.IfExp)): return True current = getattr(current, 'parent', None) return False name = getNodeName(node) if not name: return if on_conditional_branch(): # We cannot predict if this conditional branch is going to # be executed. return if isinstance(self.scope, FunctionScope) and name in self.scope.globals: self.scope.globals.remove(name) else: try: del self.scope[name] except KeyError: self.report(messages.UndefinedName, node, name) def _handle_type_comments(self, node): for (lineno, col_offset), comment in self._type_comments.get(node, ()): comment = comment.split(':', 1)[1].strip() func_match = TYPE_FUNC_RE.match(comment) if func_match: parts = ( func_match.group(1).replace('*', ''), func_match.group(2).strip(), ) else: parts = (comment,) for part in parts: if PY2: part = part.replace('...', 'Ellipsis') self.deferFunction(functools.partial( self.handleStringAnnotation, part, DummyNode(lineno, col_offset), lineno, col_offset, messages.CommentAnnotationSyntaxError, )) def handleChildren(self, tree, omit=None): self._handle_type_comments(tree) for node in iter_child_nodes(tree, omit=omit): self.handleNode(node, tree) def isLiteralTupleUnpacking(self, node): if isinstance(node, ast.Assign): for child in node.targets + [node.value]: if not hasattr(child, 'elts'): return False return True def isDocstring(self, node): """ Determine if the given node is a docstring, as long as it is at the correct place in the node tree. """ return isinstance(node, ast.Str) or (isinstance(node, ast.Expr) and isinstance(node.value, ast.Str)) def getDocstring(self, node): if isinstance(node, ast.Expr): node = node.value if not isinstance(node, ast.Str): return (None, None) if PYPY or PY38_PLUS: doctest_lineno = node.lineno - 1 else: # Computed incorrectly if the docstring has backslash doctest_lineno = node.lineno - node.s.count('\n') - 1 return (node.s, doctest_lineno) def handleNode(self, node, parent): if node is None: return if self.offset and getattr(node, 'lineno', None) is not None: node.lineno += self.offset[0] node.col_offset += self.offset[1] if self.traceTree: print(' ' * self.nodeDepth + node.__class__.__name__) if self.futuresAllowed and not (isinstance(node, ast.ImportFrom) or self.isDocstring(node)): self.futuresAllowed = False self.nodeDepth += 1 node.depth = self.nodeDepth node.parent = parent try: handler = self.getNodeHandler(node.__class__) handler(node) finally: self.nodeDepth -= 1 if self.traceTree: print(' ' * self.nodeDepth + 'end ' + node.__class__.__name__) _getDoctestExamples = doctest.DocTestParser().get_examples def handleDoctests(self, node): try: if hasattr(node, 'docstring'): docstring = node.docstring # This is just a reasonable guess. In Python 3.7, docstrings no # longer have line numbers associated with them. This will be # incorrect if there are empty lines between the beginning # of the function and the docstring. node_lineno = node.lineno if hasattr(node, 'args'): node_lineno = max([node_lineno] + [arg.lineno for arg in node.args.args]) else: (docstring, node_lineno) = self.getDocstring(node.body[0]) examples = docstring and self._getDoctestExamples(docstring) except (ValueError, IndexError): # e.g. line 6 of the docstring for <string> has inconsistent # leading whitespace: ... return if not examples: return # Place doctest in module scope saved_stack = self.scopeStack self.scopeStack = [self.scopeStack[0]] node_offset = self.offset or (0, 0) self.pushScope(DoctestScope) self.addBinding(None, Builtin('_')) for example in examples: try: tree = ast.parse(example.source, "<doctest>") except SyntaxError: e = sys.exc_info()[1] if PYPY: e.offset += 1 position = (node_lineno + example.lineno + e.lineno, example.indent + 4 + (e.offset or 0)) self.report(messages.DoctestSyntaxError, node, position) else: self.offset = (node_offset[0] + node_lineno + example.lineno, node_offset[1] + example.indent + 4) self.handleChildren(tree) self.offset = node_offset self.popScope() self.scopeStack = saved_stack def handleStringAnnotation(self, s, node, ref_lineno, ref_col_offset, err): try: tree = ast.parse(s) except SyntaxError: self.report(err, node, s) return body = tree.body if len(body) != 1 or not isinstance(body[0], ast.Expr): self.report(err, node, s) return parsed_annotation = tree.body[0].value for descendant in ast.walk(parsed_annotation): if ( 'lineno' in descendant._attributes and 'col_offset' in descendant._attributes ): descendant.lineno = ref_lineno descendant.col_offset = ref_col_offset self.handleNode(parsed_annotation, node) def handleAnnotation(self, annotation, node): if isinstance(annotation, ast.Str): # Defer handling forward annotation. self.deferFunction(functools.partial( self.handleStringAnnotation, annotation.s, node, annotation.lineno, annotation.col_offset, messages.ForwardAnnotationSyntaxError, )) elif self.annotationsFutureEnabled: self.deferFunction(lambda: self.handleNode(annotation, node)) else: self.handleNode(annotation, node) def ignore(self, node): pass # "stmt" type nodes DELETE = PRINT = FOR = ASYNCFOR = WHILE = IF = WITH = WITHITEM = \ ASYNCWITH = ASYNCWITHITEM = TRYFINALLY = EXEC = \ EXPR = ASSIGN = handleChildren PASS = ignore # "expr" type nodes BOOLOP = BINOP = UNARYOP = IFEXP = SET = \ CALL = REPR = ATTRIBUTE = SUBSCRIPT = \ STARRED = NAMECONSTANT = handleChildren NUM = STR = BYTES = ELLIPSIS = CONSTANT = ignore # "slice" type nodes SLICE = EXTSLICE = INDEX = handleChildren # expression contexts are node instances too, though being constants LOAD = STORE = DEL = AUGLOAD = AUGSTORE = PARAM = ignore # same for operators AND = OR = ADD = SUB = MULT = DIV = MOD = POW = LSHIFT = RSHIFT = \ BITOR = BITXOR = BITAND = FLOORDIV = INVERT = NOT = UADD = USUB = \ EQ = NOTEQ = LT = LTE = GT = GTE = IS = ISNOT = IN = NOTIN = \ MATMULT = ignore def RAISE(self, node): self.handleChildren(node) arg = get_raise_argument(node) if isinstance(arg, ast.Call): if is_notimplemented_name_node(arg.func): # Handle "raise NotImplemented(...)" self.report(messages.RaiseNotImplemented, node) elif is_notimplemented_name_node(arg): # Handle "raise NotImplemented" self.report(messages.RaiseNotImplemented, node) # additional node types COMPREHENSION = KEYWORD = FORMATTEDVALUE = JOINEDSTR = handleChildren def DICT(self, node): # Complain if there are duplicate keys with different values # If they have the same value it's not going to cause potentially # unexpected behaviour so we'll not complain. keys = [ convert_to_value(key) for key in node.keys ] key_counts = counter(keys) duplicate_keys = [ key for key, count in key_counts.items() if count > 1 ] for key in duplicate_keys: key_indices = [i for i, i_key in enumerate(keys) if i_key == key] values = counter( convert_to_value(node.values[index]) for index in key_indices ) if any(count == 1 for value, count in values.items()): for key_index in key_indices: key_node = node.keys[key_index] if isinstance(key, VariableKey): self.report(messages.MultiValueRepeatedKeyVariable, key_node, key.name) else: self.report( messages.MultiValueRepeatedKeyLiteral, key_node, key, ) self.handleChildren(node) def ASSERT(self, node): if isinstance(node.test, ast.Tuple) and node.test.elts != []: self.report(messages.AssertTuple, node) self.handleChildren(node) def GLOBAL(self, node): """ Keep track of globals declarations. """ global_scope_index = 1 if self._in_doctest() else 0 global_scope = self.scopeStack[global_scope_index] # Ignore 'global' statement in global scope. if self.scope is not global_scope: # One 'global' statement can bind multiple (comma-delimited) names. for node_name in node.names: node_value = Assignment(node_name, node) # Remove UndefinedName messages already reported for this name. # TODO: if the global is not used in this scope, it does not # become a globally defined name. See test_unused_global. self.messages = [ m for m in self.messages if not isinstance(m, messages.UndefinedName) or m.message_args[0] != node_name] # Bind name to global scope if it doesn't exist already. global_scope.setdefault(node_name, node_value) # Bind name to non-global scopes, but as already "used". node_value.used = (global_scope, node) for scope in self.scopeStack[global_scope_index + 1:]: scope[node_name] = node_value NONLOCAL = GLOBAL def GENERATOREXP(self, node): self.pushScope(GeneratorScope) self.handleChildren(node) self.popScope() LISTCOMP = handleChildren if PY2 else GENERATOREXP DICTCOMP = SETCOMP = GENERATOREXP def NAME(self, node): """ Handle occurrence of Name (which can be a load/store/delete access.) """ # Locate the name in locals / function / globals scopes. if isinstance(node.ctx, (ast.Load, ast.AugLoad)): self.handleNodeLoad(node) if (node.id == 'locals' and isinstance(self.scope, FunctionScope) and isinstance(node.parent, ast.Call)): # we are doing locals() call in current scope self.scope.usesLocals = True elif isinstance(node.ctx, (ast.Store, ast.AugStore, ast.Param)): self.handleNodeStore(node) elif isinstance(node.ctx, ast.Del): self.handleNodeDelete(node) else: # Unknown context raise RuntimeError("Got impossible expression context: %r" % (node.ctx,)) def CONTINUE(self, node): # Walk the tree up until we see a loop (OK), a function or class # definition (not OK), for 'continue', a finally block (not OK), or # the top module scope (not OK) n = node while hasattr(n, 'parent'): n, n_child = n.parent, n if isinstance(n, LOOP_TYPES): # Doesn't apply unless it's in the loop itself if n_child not in n.orelse: return if isinstance(n, (ast.FunctionDef, ast.ClassDef)): break # Handle Try/TryFinally difference in Python < and >= 3.3 if hasattr(n, 'finalbody') and isinstance(node, ast.Continue): if n_child in n.finalbody: self.report(messages.ContinueInFinally, node) return if isinstance(node, ast.Continue): self.report(messages.ContinueOutsideLoop, node) else: # ast.Break self.report(messages.BreakOutsideLoop, node) BREAK = CONTINUE def RETURN(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.ReturnOutsideFunction, node) return if ( node.value and hasattr(self.scope, 'returnValue') and not self.scope.returnValue ): self.scope.returnValue = node.value self.handleNode(node.value, node) def YIELD(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.YieldOutsideFunction, node) return self.scope.isGenerator = True self.handleNode(node.value, node) AWAIT = YIELDFROM = YIELD def FUNCTIONDEF(self, node): for deco in node.decorator_list: self.handleNode(deco, node) self.LAMBDA(node) self.addBinding(node, FunctionDefinition(node.name, node)) # doctest does not process doctest within a doctest, # or in nested functions. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) ASYNCFUNCTIONDEF = FUNCTIONDEF def LAMBDA(self, node): args = [] annotations = [] if PY2: def addArgs(arglist): for arg in arglist: if isinstance(arg, ast.Tuple): addArgs(arg.elts) else: args.append(arg.id) addArgs(node.args.args) defaults = node.args.defaults else: for arg in node.args.args + node.args.kwonlyargs: args.append(arg.arg) annotations.append(arg.annotation) defaults = node.args.defaults + node.args.kw_defaults # Only for Python3 FunctionDefs is_py3_func = hasattr(node, 'returns') for arg_name in ('vararg', 'kwarg'): wildcard = getattr(node.args, arg_name) if not wildcard: continue args.append(wildcard if PY2 else wildcard.arg) if is_py3_func: if PY2: # Python 2.7 argannotation = arg_name + 'annotation' annotations.append(getattr(node.args, argannotation)) else: # Python >= 3.4 annotations.append(wildcard.annotation) if is_py3_func: annotations.append(node.returns) if len(set(args)) < len(args): for (idx, arg) in enumerate(args): if arg in args[:idx]: self.report(messages.DuplicateArgument, node, arg) for annotation in annotations: self.handleAnnotation(annotation, node) for default in defaults: self.handleNode(default, node) def runFunction(): self.pushScope() self.handleChildren(node, omit='decorator_list') def checkUnusedAssignments(): """ Check to see if any assignments have not been used. """ for name, binding in self.scope.unusedAssignments(): self.report(messages.UnusedVariable, binding.source, name) self.deferAssignment(checkUnusedAssignments) if PY2: def checkReturnWithArgumentInsideGenerator(): """ Check to see if there is any return statement with arguments but the function is a generator. """ if self.scope.isGenerator and self.scope.returnValue: self.report(messages.ReturnWithArgsInsideGenerator, self.scope.returnValue) self.deferAssignment(checkReturnWithArgumentInsideGenerator) self.popScope() self.deferFunction(runFunction) def ARGUMENTS(self, node): self.handleChildren(node, omit=('defaults', 'kw_defaults')) if PY2: scope_node = self.getScopeNode(node) if node.vararg: self.addBinding(node, Argument(node.vararg, scope_node)) if node.kwarg: self.addBinding(node, Argument(node.kwarg, scope_node)) def ARG(self, node): self.addBinding(node, Argument(node.arg, self.getScopeNode(node))) def CLASSDEF(self, node): """ Check names used in a class definition, including its decorators, base classes, and the body of its definition. Additionally, add its name to the current scope. """ for deco in node.decorator_list: self.handleNode(deco, node) for baseNode in node.bases: self.handleNode(baseNode, node) if not PY2: for keywordNode in node.keywords: self.handleNode(keywordNode, node) self.pushScope(ClassScope) # doctest does not process doctest within a doctest # classes within classes are processed. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) for stmt in node.body: self.handleNode(stmt, node) self.popScope() self.addBinding(node, ClassDefinition(node.name, node)) def AUGASSIGN(self, node): self.handleNodeLoad(node.target) self.handleNode(node.value, node) self.handleNode(node.target, node) def TUPLE(self, node): if not PY2 and isinstance(node.ctx, ast.Store): # Python 3 advanced tuple unpacking: a, *b, c = d. # Only one starred expression is allowed, and no more than 1<<8 # assignments are allowed before a stared expression. There is # also a limit of 1<<24 expressions after the starred expression, # which is impossible to test due to memory restrictions, but we # add it here anyway has_starred = False star_loc = -1 for i, n in enumerate(node.elts): if isinstance(n, ast.Starred): if has_starred: self.report(messages.TwoStarredExpressions, node) # The SyntaxError doesn't distinguish two from more # than two. break has_starred = True star_loc = i if star_loc >= 1 << 8 or len(node.elts) - star_loc - 1 >= 1 << 24: self.report(messages.TooManyExpressionsInStarredAssignment, node) self.handleChildren(node) LIST = TUPLE def IMPORT(self, node): for alias in node.names: if '.' in alias.name and not alias.asname: importation = SubmoduleImportation(alias.name, node) else: name = alias.asname or alias.name importation = Importation(name, node, alias.name) self.addBinding(node, importation) def IMPORTFROM(self, node): if node.module == '__future__': if not self.futuresAllowed: self.report(messages.LateFutureImport, node, [n.name for n in node.names]) else: self.futuresAllowed = False module = ('.' * node.level) + (node.module or '') for alias in node.names: name = alias.asname or alias.name if node.module == '__future__': importation = FutureImportation(name, node, self.scope) if alias.name not in __future__.all_feature_names: self.report(messages.FutureFeatureNotDefined, node, alias.name) if alias.name == 'annotations': self.annotationsFutureEnabled = True elif alias.name == '*': # Only Python 2, local import * is a SyntaxWarning if not PY2 and not isinstance(self.scope, ModuleScope): self.report(messages.ImportStarNotPermitted, node, module) continue self.scope.importStarred = True self.report(messages.ImportStarUsed, node, module) importation = StarImportation(module, node) else: importation = ImportationFrom(name, node, module, alias.name) self.addBinding(node, importation) def TRY(self, node): handler_names = [] # List the exception handlers for i, handler in enumerate(node.handlers): if isinstance(handler.type, ast.Tuple): for exc_type in handler.type.elts: handler_names.append(getNodeName(exc_type)) elif handler.type: handler_names.append(getNodeName(handler.type)) if handler.type is None and i < len(node.handlers) - 1: self.report(messages.DefaultExceptNotLast, handler) # Memorize the except handlers and process the body self.exceptHandlers.append(handler_names) for child in node.body: self.handleNode(child, node) self.exceptHandlers.pop() # Process the other nodes: "except:", "else:", "finally:" self.handleChildren(node, omit='body') TRYEXCEPT = TRY def EXCEPTHANDLER(self, node): if PY2 or node.name is None: self.handleChildren(node) return # If the name already exists in the scope, modify state of existing # binding. if node.name in self.scope: self.handleNodeStore(node) # 3.x: the name of the exception, which is not a Name node, but a # simple string, creates a local that is only bound within the scope of # the except: block. As such, temporarily remove the existing binding # to more accurately determine if the name is used in the except: # block. try: prev_definition = self.scope.pop(node.name) except KeyError: prev_definition = None self.handleNodeStore(node) self.handleChildren(node) # See discussion on https://github.com/PyCQA/pyflakes/pull/59 # We're removing the local name since it's being unbound after leaving # the except: block and it's always unbound if the except: block is # never entered. This will cause an "undefined name" error raised if # the checked code tries to use the name afterwards. # # Unless it's been removed already. Then do nothing. try: binding = self.scope.pop(node.name) except KeyError: pass else: if not binding.used: self.report(messages.UnusedVariable, node, node.name) # Restore. if prev_definition: self.scope[node.name] = prev_definition def ANNASSIGN(self, node): if node.value: # Only bind the *targets* if the assignment has a value. # Otherwise it's not really ast.Store and shouldn't silence # UndefinedLocal warnings. self.handleNode(node.target, node) self.handleAnnotation(node.annotation, node) if node.value: # If the assignment has value, handle the *value* now. self.handleNode(node.value, node) def COMPARE(self, node): literals = (ast.Str, ast.Num) if not PY2: literals += (ast.Bytes,) left = node.left for op, right in zip(node.ops, node.comparators): if (isinstance(op, (ast.Is, ast.IsNot)) and (isinstance(left, literals) or isinstance(right, literals))): self.report(messages.IsLiteral, node) left = right self.handleChildren(node)
PyCQA/pyflakes
pyflakes/checker.py
Checker.checkDeadScopes
python
def checkDeadScopes(self): for scope in self.deadScopes: # imports in classes are public members if isinstance(scope, ClassScope): continue all_binding = scope.get('__all__') if all_binding and not isinstance(all_binding, ExportBinding): all_binding = None if all_binding: all_names = set(all_binding.names) undefined = all_names.difference(scope) else: all_names = undefined = [] if undefined: if not scope.importStarred and \ os.path.basename(self.filename) != '__init__.py': # Look for possible mistakes in the export list for name in undefined: self.report(messages.UndefinedExport, scope['__all__'].source, name) # mark all import '*' as used by the undefined in __all__ if scope.importStarred: from_list = [] for binding in scope.values(): if isinstance(binding, StarImportation): binding.used = all_binding from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) for name in undefined: self.report(messages.ImportStarUsage, scope['__all__'].source, name, from_list) # Look for imported names that aren't used. for value in scope.values(): if isinstance(value, Importation): used = value.used or value.name in all_names if not used: messg = messages.UnusedImport self.report(messg, value.source, str(value)) for node in value.redefined: if isinstance(self.getParent(node), FOR_TYPES): messg = messages.ImportShadowedByLoopVar elif used: continue else: messg = messages.RedefinedWhileUnused self.report(messg, node, value.name, value.source)
Look at scopes which have been fully examined and report names in them which were imported but unused.
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/checker.py#L759-L814
[ "def report(self, messageClass, *args, **kwargs):\n self.messages.append(messageClass(self.filename, *args, **kwargs))\n" ]
class Checker(object): """ I check the cleanliness and sanity of Python code. @ivar _deferredFunctions: Tracking list used by L{deferFunction}. Elements of the list are two-tuples. The first element is the callable passed to L{deferFunction}. The second element is a copy of the scope stack at the time L{deferFunction} was called. @ivar _deferredAssignments: Similar to C{_deferredFunctions}, but for callables which are deferred assignment checks. """ _ast_node_scope = { ast.Module: ModuleScope, ast.ClassDef: ClassScope, ast.FunctionDef: FunctionScope, ast.Lambda: FunctionScope, ast.ListComp: GeneratorScope, ast.SetComp: GeneratorScope, ast.GeneratorExp: GeneratorScope, ast.DictComp: GeneratorScope, } if PY35_PLUS: _ast_node_scope[ast.AsyncFunctionDef] = FunctionScope, nodeDepth = 0 offset = None traceTree = False builtIns = set(builtin_vars).union(_MAGIC_GLOBALS) _customBuiltIns = os.environ.get('PYFLAKES_BUILTINS') if _customBuiltIns: builtIns.update(_customBuiltIns.split(',')) del _customBuiltIns # TODO: file_tokens= is required to perform checks on type comments, # eventually make this a required positional argument. For now it # is defaulted to `()` for api compatibility. def __init__(self, tree, filename='(none)', builtins=None, withDoctest='PYFLAKES_DOCTEST' in os.environ, file_tokens=()): self._nodeHandlers = {} self._deferredFunctions = [] self._deferredAssignments = [] self.deadScopes = [] self.messages = [] self.filename = filename if builtins: self.builtIns = self.builtIns.union(builtins) self.withDoctest = withDoctest try: self.scopeStack = [Checker._ast_node_scope[type(tree)]()] except KeyError: raise RuntimeError('No scope implemented for the node %r' % tree) self.exceptHandlers = [()] self.root = tree self._type_comments = _collect_type_comments(tree, file_tokens) for builtin in self.builtIns: self.addBinding(None, Builtin(builtin)) self.handleChildren(tree) self.runDeferred(self._deferredFunctions) # Set _deferredFunctions to None so that deferFunction will fail # noisily if called after we've run through the deferred functions. self._deferredFunctions = None self.runDeferred(self._deferredAssignments) # Set _deferredAssignments to None so that deferAssignment will fail # noisily if called after we've run through the deferred assignments. self._deferredAssignments = None del self.scopeStack[1:] self.popScope() self.checkDeadScopes() def deferFunction(self, callable): """ Schedule a function handler to be called just before completion. This is used for handling function bodies, which must be deferred because code later in the file might modify the global scope. When `callable` is called, the scope at the time this is called will be restored, however it will contain any new bindings added to it. """ self._deferredFunctions.append((callable, self.scopeStack[:], self.offset)) def deferAssignment(self, callable): """ Schedule an assignment handler to be called just after deferred function handlers. """ self._deferredAssignments.append((callable, self.scopeStack[:], self.offset)) def runDeferred(self, deferred): """ Run the callables in C{deferred} using their associated scope stack. """ for handler, scope, offset in deferred: self.scopeStack = scope self.offset = offset handler() def _in_doctest(self): return (len(self.scopeStack) >= 2 and isinstance(self.scopeStack[1], DoctestScope)) @property def futuresAllowed(self): if not all(isinstance(scope, ModuleScope) for scope in self.scopeStack): return False return self.scope._futures_allowed @futuresAllowed.setter def futuresAllowed(self, value): assert value is False if isinstance(self.scope, ModuleScope): self.scope._futures_allowed = False @property def annotationsFutureEnabled(self): scope = self.scopeStack[0] if not isinstance(scope, ModuleScope): return False return scope._annotations_future_enabled @annotationsFutureEnabled.setter def annotationsFutureEnabled(self, value): assert value is True assert isinstance(self.scope, ModuleScope) self.scope._annotations_future_enabled = True @property def scope(self): return self.scopeStack[-1] def popScope(self): self.deadScopes.append(self.scopeStack.pop()) def pushScope(self, scopeClass=FunctionScope): self.scopeStack.append(scopeClass()) def report(self, messageClass, *args, **kwargs): self.messages.append(messageClass(self.filename, *args, **kwargs)) def getParent(self, node): # Lookup the first parent which is not Tuple, List or Starred while True: node = node.parent if not hasattr(node, 'elts') and not hasattr(node, 'ctx'): return node def getCommonAncestor(self, lnode, rnode, stop): if stop in (lnode, rnode) or not (hasattr(lnode, 'parent') and hasattr(rnode, 'parent')): return None if lnode is rnode: return lnode if (lnode.depth > rnode.depth): return self.getCommonAncestor(lnode.parent, rnode, stop) if (lnode.depth < rnode.depth): return self.getCommonAncestor(lnode, rnode.parent, stop) return self.getCommonAncestor(lnode.parent, rnode.parent, stop) def descendantOf(self, node, ancestors, stop): for a in ancestors: if self.getCommonAncestor(node, a, stop): return True return False def _getAncestor(self, node, ancestor_type): parent = node while True: if parent is self.root: return None parent = self.getParent(parent) if isinstance(parent, ancestor_type): return parent def getScopeNode(self, node): return self._getAncestor(node, tuple(Checker._ast_node_scope.keys())) def differentForks(self, lnode, rnode): """True, if lnode and rnode are located on different forks of IF/TRY""" ancestor = self.getCommonAncestor(lnode, rnode, self.root) parts = getAlternatives(ancestor) if parts: for items in parts: if self.descendantOf(lnode, items, ancestor) ^ \ self.descendantOf(rnode, items, ancestor): return True return False def addBinding(self, node, value): """ Called when a binding is altered. - `node` is the statement responsible for the change - `value` is the new value, a Binding instance """ # assert value.source in (node, node.parent): for scope in self.scopeStack[::-1]: if value.name in scope: break existing = scope.get(value.name) if (existing and not isinstance(existing, Builtin) and not self.differentForks(node, existing.source)): parent_stmt = self.getParent(value.source) if isinstance(existing, Importation) and isinstance(parent_stmt, FOR_TYPES): self.report(messages.ImportShadowedByLoopVar, node, value.name, existing.source) elif scope is self.scope: if (isinstance(parent_stmt, ast.comprehension) and not isinstance(self.getParent(existing.source), (FOR_TYPES, ast.comprehension))): self.report(messages.RedefinedInListComp, node, value.name, existing.source) elif not existing.used and value.redefines(existing): if value.name != '_' or isinstance(existing, Importation): if not is_typing_overload(existing, self.scopeStack): self.report(messages.RedefinedWhileUnused, node, value.name, existing.source) elif isinstance(existing, Importation) and value.redefines(existing): existing.redefined.append(node) if value.name in self.scope: # then assume the rebound name is used as a global or within a loop value.used = self.scope[value.name].used self.scope[value.name] = value def getNodeHandler(self, node_class): try: return self._nodeHandlers[node_class] except KeyError: nodeType = getNodeType(node_class) self._nodeHandlers[node_class] = handler = getattr(self, nodeType) return handler def handleNodeLoad(self, node): name = getNodeName(node) if not name: return in_generators = None importStarred = None # try enclosing function scopes and global scope for scope in self.scopeStack[-1::-1]: if isinstance(scope, ClassScope): if not PY2 and name == '__class__': return elif in_generators is False: # only generators used in a class scope can access the # names of the class. this is skipped during the first # iteration continue if (name == 'print' and isinstance(scope.get(name, None), Builtin)): parent = self.getParent(node) if (isinstance(parent, ast.BinOp) and isinstance(parent.op, ast.RShift)): self.report(messages.InvalidPrintSyntax, node) try: scope[name].used = (self.scope, node) # if the name of SubImportation is same as # alias of other Importation and the alias # is used, SubImportation also should be marked as used. n = scope[name] if isinstance(n, Importation) and n._has_alias(): try: scope[n.fullName].used = (self.scope, node) except KeyError: pass except KeyError: pass else: return importStarred = importStarred or scope.importStarred if in_generators is not False: in_generators = isinstance(scope, GeneratorScope) if importStarred: from_list = [] for scope in self.scopeStack[-1::-1]: for binding in scope.values(): if isinstance(binding, StarImportation): # mark '*' imports as used for each scope binding.used = (self.scope, node) from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) self.report(messages.ImportStarUsage, node, name, from_list) return if name == '__path__' and os.path.basename(self.filename) == '__init__.py': # the special name __path__ is valid only in packages return if name == '__module__' and isinstance(self.scope, ClassScope): return # protected with a NameError handler? if 'NameError' not in self.exceptHandlers[-1]: self.report(messages.UndefinedName, node, name) def handleNodeStore(self, node): name = getNodeName(node) if not name: return # if the name hasn't already been defined in the current scope if isinstance(self.scope, FunctionScope) and name not in self.scope: # for each function or module scope above us for scope in self.scopeStack[:-1]: if not isinstance(scope, (FunctionScope, ModuleScope)): continue # if the name was defined in that scope, and the name has # been accessed already in the current scope, and hasn't # been declared global used = name in scope and scope[name].used if used and used[0] is self.scope and name not in self.scope.globals: # then it's probably a mistake self.report(messages.UndefinedLocal, scope[name].used[1], name, scope[name].source) break parent_stmt = self.getParent(node) if isinstance(parent_stmt, (FOR_TYPES, ast.comprehension)) or ( parent_stmt != node.parent and not self.isLiteralTupleUnpacking(parent_stmt)): binding = Binding(name, node) elif name == '__all__' and isinstance(self.scope, ModuleScope): binding = ExportBinding(name, node.parent, self.scope) elif isinstance(getattr(node, 'ctx', None), ast.Param): binding = Argument(name, self.getScopeNode(node)) else: binding = Assignment(name, node) self.addBinding(node, binding) def handleNodeDelete(self, node): def on_conditional_branch(): """ Return `True` if node is part of a conditional body. """ current = getattr(node, 'parent', None) while current: if isinstance(current, (ast.If, ast.While, ast.IfExp)): return True current = getattr(current, 'parent', None) return False name = getNodeName(node) if not name: return if on_conditional_branch(): # We cannot predict if this conditional branch is going to # be executed. return if isinstance(self.scope, FunctionScope) and name in self.scope.globals: self.scope.globals.remove(name) else: try: del self.scope[name] except KeyError: self.report(messages.UndefinedName, node, name) def _handle_type_comments(self, node): for (lineno, col_offset), comment in self._type_comments.get(node, ()): comment = comment.split(':', 1)[1].strip() func_match = TYPE_FUNC_RE.match(comment) if func_match: parts = ( func_match.group(1).replace('*', ''), func_match.group(2).strip(), ) else: parts = (comment,) for part in parts: if PY2: part = part.replace('...', 'Ellipsis') self.deferFunction(functools.partial( self.handleStringAnnotation, part, DummyNode(lineno, col_offset), lineno, col_offset, messages.CommentAnnotationSyntaxError, )) def handleChildren(self, tree, omit=None): self._handle_type_comments(tree) for node in iter_child_nodes(tree, omit=omit): self.handleNode(node, tree) def isLiteralTupleUnpacking(self, node): if isinstance(node, ast.Assign): for child in node.targets + [node.value]: if not hasattr(child, 'elts'): return False return True def isDocstring(self, node): """ Determine if the given node is a docstring, as long as it is at the correct place in the node tree. """ return isinstance(node, ast.Str) or (isinstance(node, ast.Expr) and isinstance(node.value, ast.Str)) def getDocstring(self, node): if isinstance(node, ast.Expr): node = node.value if not isinstance(node, ast.Str): return (None, None) if PYPY or PY38_PLUS: doctest_lineno = node.lineno - 1 else: # Computed incorrectly if the docstring has backslash doctest_lineno = node.lineno - node.s.count('\n') - 1 return (node.s, doctest_lineno) def handleNode(self, node, parent): if node is None: return if self.offset and getattr(node, 'lineno', None) is not None: node.lineno += self.offset[0] node.col_offset += self.offset[1] if self.traceTree: print(' ' * self.nodeDepth + node.__class__.__name__) if self.futuresAllowed and not (isinstance(node, ast.ImportFrom) or self.isDocstring(node)): self.futuresAllowed = False self.nodeDepth += 1 node.depth = self.nodeDepth node.parent = parent try: handler = self.getNodeHandler(node.__class__) handler(node) finally: self.nodeDepth -= 1 if self.traceTree: print(' ' * self.nodeDepth + 'end ' + node.__class__.__name__) _getDoctestExamples = doctest.DocTestParser().get_examples def handleDoctests(self, node): try: if hasattr(node, 'docstring'): docstring = node.docstring # This is just a reasonable guess. In Python 3.7, docstrings no # longer have line numbers associated with them. This will be # incorrect if there are empty lines between the beginning # of the function and the docstring. node_lineno = node.lineno if hasattr(node, 'args'): node_lineno = max([node_lineno] + [arg.lineno for arg in node.args.args]) else: (docstring, node_lineno) = self.getDocstring(node.body[0]) examples = docstring and self._getDoctestExamples(docstring) except (ValueError, IndexError): # e.g. line 6 of the docstring for <string> has inconsistent # leading whitespace: ... return if not examples: return # Place doctest in module scope saved_stack = self.scopeStack self.scopeStack = [self.scopeStack[0]] node_offset = self.offset or (0, 0) self.pushScope(DoctestScope) self.addBinding(None, Builtin('_')) for example in examples: try: tree = ast.parse(example.source, "<doctest>") except SyntaxError: e = sys.exc_info()[1] if PYPY: e.offset += 1 position = (node_lineno + example.lineno + e.lineno, example.indent + 4 + (e.offset or 0)) self.report(messages.DoctestSyntaxError, node, position) else: self.offset = (node_offset[0] + node_lineno + example.lineno, node_offset[1] + example.indent + 4) self.handleChildren(tree) self.offset = node_offset self.popScope() self.scopeStack = saved_stack def handleStringAnnotation(self, s, node, ref_lineno, ref_col_offset, err): try: tree = ast.parse(s) except SyntaxError: self.report(err, node, s) return body = tree.body if len(body) != 1 or not isinstance(body[0], ast.Expr): self.report(err, node, s) return parsed_annotation = tree.body[0].value for descendant in ast.walk(parsed_annotation): if ( 'lineno' in descendant._attributes and 'col_offset' in descendant._attributes ): descendant.lineno = ref_lineno descendant.col_offset = ref_col_offset self.handleNode(parsed_annotation, node) def handleAnnotation(self, annotation, node): if isinstance(annotation, ast.Str): # Defer handling forward annotation. self.deferFunction(functools.partial( self.handleStringAnnotation, annotation.s, node, annotation.lineno, annotation.col_offset, messages.ForwardAnnotationSyntaxError, )) elif self.annotationsFutureEnabled: self.deferFunction(lambda: self.handleNode(annotation, node)) else: self.handleNode(annotation, node) def ignore(self, node): pass # "stmt" type nodes DELETE = PRINT = FOR = ASYNCFOR = WHILE = IF = WITH = WITHITEM = \ ASYNCWITH = ASYNCWITHITEM = TRYFINALLY = EXEC = \ EXPR = ASSIGN = handleChildren PASS = ignore # "expr" type nodes BOOLOP = BINOP = UNARYOP = IFEXP = SET = \ CALL = REPR = ATTRIBUTE = SUBSCRIPT = \ STARRED = NAMECONSTANT = handleChildren NUM = STR = BYTES = ELLIPSIS = CONSTANT = ignore # "slice" type nodes SLICE = EXTSLICE = INDEX = handleChildren # expression contexts are node instances too, though being constants LOAD = STORE = DEL = AUGLOAD = AUGSTORE = PARAM = ignore # same for operators AND = OR = ADD = SUB = MULT = DIV = MOD = POW = LSHIFT = RSHIFT = \ BITOR = BITXOR = BITAND = FLOORDIV = INVERT = NOT = UADD = USUB = \ EQ = NOTEQ = LT = LTE = GT = GTE = IS = ISNOT = IN = NOTIN = \ MATMULT = ignore def RAISE(self, node): self.handleChildren(node) arg = get_raise_argument(node) if isinstance(arg, ast.Call): if is_notimplemented_name_node(arg.func): # Handle "raise NotImplemented(...)" self.report(messages.RaiseNotImplemented, node) elif is_notimplemented_name_node(arg): # Handle "raise NotImplemented" self.report(messages.RaiseNotImplemented, node) # additional node types COMPREHENSION = KEYWORD = FORMATTEDVALUE = JOINEDSTR = handleChildren def DICT(self, node): # Complain if there are duplicate keys with different values # If they have the same value it's not going to cause potentially # unexpected behaviour so we'll not complain. keys = [ convert_to_value(key) for key in node.keys ] key_counts = counter(keys) duplicate_keys = [ key for key, count in key_counts.items() if count > 1 ] for key in duplicate_keys: key_indices = [i for i, i_key in enumerate(keys) if i_key == key] values = counter( convert_to_value(node.values[index]) for index in key_indices ) if any(count == 1 for value, count in values.items()): for key_index in key_indices: key_node = node.keys[key_index] if isinstance(key, VariableKey): self.report(messages.MultiValueRepeatedKeyVariable, key_node, key.name) else: self.report( messages.MultiValueRepeatedKeyLiteral, key_node, key, ) self.handleChildren(node) def ASSERT(self, node): if isinstance(node.test, ast.Tuple) and node.test.elts != []: self.report(messages.AssertTuple, node) self.handleChildren(node) def GLOBAL(self, node): """ Keep track of globals declarations. """ global_scope_index = 1 if self._in_doctest() else 0 global_scope = self.scopeStack[global_scope_index] # Ignore 'global' statement in global scope. if self.scope is not global_scope: # One 'global' statement can bind multiple (comma-delimited) names. for node_name in node.names: node_value = Assignment(node_name, node) # Remove UndefinedName messages already reported for this name. # TODO: if the global is not used in this scope, it does not # become a globally defined name. See test_unused_global. self.messages = [ m for m in self.messages if not isinstance(m, messages.UndefinedName) or m.message_args[0] != node_name] # Bind name to global scope if it doesn't exist already. global_scope.setdefault(node_name, node_value) # Bind name to non-global scopes, but as already "used". node_value.used = (global_scope, node) for scope in self.scopeStack[global_scope_index + 1:]: scope[node_name] = node_value NONLOCAL = GLOBAL def GENERATOREXP(self, node): self.pushScope(GeneratorScope) self.handleChildren(node) self.popScope() LISTCOMP = handleChildren if PY2 else GENERATOREXP DICTCOMP = SETCOMP = GENERATOREXP def NAME(self, node): """ Handle occurrence of Name (which can be a load/store/delete access.) """ # Locate the name in locals / function / globals scopes. if isinstance(node.ctx, (ast.Load, ast.AugLoad)): self.handleNodeLoad(node) if (node.id == 'locals' and isinstance(self.scope, FunctionScope) and isinstance(node.parent, ast.Call)): # we are doing locals() call in current scope self.scope.usesLocals = True elif isinstance(node.ctx, (ast.Store, ast.AugStore, ast.Param)): self.handleNodeStore(node) elif isinstance(node.ctx, ast.Del): self.handleNodeDelete(node) else: # Unknown context raise RuntimeError("Got impossible expression context: %r" % (node.ctx,)) def CONTINUE(self, node): # Walk the tree up until we see a loop (OK), a function or class # definition (not OK), for 'continue', a finally block (not OK), or # the top module scope (not OK) n = node while hasattr(n, 'parent'): n, n_child = n.parent, n if isinstance(n, LOOP_TYPES): # Doesn't apply unless it's in the loop itself if n_child not in n.orelse: return if isinstance(n, (ast.FunctionDef, ast.ClassDef)): break # Handle Try/TryFinally difference in Python < and >= 3.3 if hasattr(n, 'finalbody') and isinstance(node, ast.Continue): if n_child in n.finalbody: self.report(messages.ContinueInFinally, node) return if isinstance(node, ast.Continue): self.report(messages.ContinueOutsideLoop, node) else: # ast.Break self.report(messages.BreakOutsideLoop, node) BREAK = CONTINUE def RETURN(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.ReturnOutsideFunction, node) return if ( node.value and hasattr(self.scope, 'returnValue') and not self.scope.returnValue ): self.scope.returnValue = node.value self.handleNode(node.value, node) def YIELD(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.YieldOutsideFunction, node) return self.scope.isGenerator = True self.handleNode(node.value, node) AWAIT = YIELDFROM = YIELD def FUNCTIONDEF(self, node): for deco in node.decorator_list: self.handleNode(deco, node) self.LAMBDA(node) self.addBinding(node, FunctionDefinition(node.name, node)) # doctest does not process doctest within a doctest, # or in nested functions. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) ASYNCFUNCTIONDEF = FUNCTIONDEF def LAMBDA(self, node): args = [] annotations = [] if PY2: def addArgs(arglist): for arg in arglist: if isinstance(arg, ast.Tuple): addArgs(arg.elts) else: args.append(arg.id) addArgs(node.args.args) defaults = node.args.defaults else: for arg in node.args.args + node.args.kwonlyargs: args.append(arg.arg) annotations.append(arg.annotation) defaults = node.args.defaults + node.args.kw_defaults # Only for Python3 FunctionDefs is_py3_func = hasattr(node, 'returns') for arg_name in ('vararg', 'kwarg'): wildcard = getattr(node.args, arg_name) if not wildcard: continue args.append(wildcard if PY2 else wildcard.arg) if is_py3_func: if PY2: # Python 2.7 argannotation = arg_name + 'annotation' annotations.append(getattr(node.args, argannotation)) else: # Python >= 3.4 annotations.append(wildcard.annotation) if is_py3_func: annotations.append(node.returns) if len(set(args)) < len(args): for (idx, arg) in enumerate(args): if arg in args[:idx]: self.report(messages.DuplicateArgument, node, arg) for annotation in annotations: self.handleAnnotation(annotation, node) for default in defaults: self.handleNode(default, node) def runFunction(): self.pushScope() self.handleChildren(node, omit='decorator_list') def checkUnusedAssignments(): """ Check to see if any assignments have not been used. """ for name, binding in self.scope.unusedAssignments(): self.report(messages.UnusedVariable, binding.source, name) self.deferAssignment(checkUnusedAssignments) if PY2: def checkReturnWithArgumentInsideGenerator(): """ Check to see if there is any return statement with arguments but the function is a generator. """ if self.scope.isGenerator and self.scope.returnValue: self.report(messages.ReturnWithArgsInsideGenerator, self.scope.returnValue) self.deferAssignment(checkReturnWithArgumentInsideGenerator) self.popScope() self.deferFunction(runFunction) def ARGUMENTS(self, node): self.handleChildren(node, omit=('defaults', 'kw_defaults')) if PY2: scope_node = self.getScopeNode(node) if node.vararg: self.addBinding(node, Argument(node.vararg, scope_node)) if node.kwarg: self.addBinding(node, Argument(node.kwarg, scope_node)) def ARG(self, node): self.addBinding(node, Argument(node.arg, self.getScopeNode(node))) def CLASSDEF(self, node): """ Check names used in a class definition, including its decorators, base classes, and the body of its definition. Additionally, add its name to the current scope. """ for deco in node.decorator_list: self.handleNode(deco, node) for baseNode in node.bases: self.handleNode(baseNode, node) if not PY2: for keywordNode in node.keywords: self.handleNode(keywordNode, node) self.pushScope(ClassScope) # doctest does not process doctest within a doctest # classes within classes are processed. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) for stmt in node.body: self.handleNode(stmt, node) self.popScope() self.addBinding(node, ClassDefinition(node.name, node)) def AUGASSIGN(self, node): self.handleNodeLoad(node.target) self.handleNode(node.value, node) self.handleNode(node.target, node) def TUPLE(self, node): if not PY2 and isinstance(node.ctx, ast.Store): # Python 3 advanced tuple unpacking: a, *b, c = d. # Only one starred expression is allowed, and no more than 1<<8 # assignments are allowed before a stared expression. There is # also a limit of 1<<24 expressions after the starred expression, # which is impossible to test due to memory restrictions, but we # add it here anyway has_starred = False star_loc = -1 for i, n in enumerate(node.elts): if isinstance(n, ast.Starred): if has_starred: self.report(messages.TwoStarredExpressions, node) # The SyntaxError doesn't distinguish two from more # than two. break has_starred = True star_loc = i if star_loc >= 1 << 8 or len(node.elts) - star_loc - 1 >= 1 << 24: self.report(messages.TooManyExpressionsInStarredAssignment, node) self.handleChildren(node) LIST = TUPLE def IMPORT(self, node): for alias in node.names: if '.' in alias.name and not alias.asname: importation = SubmoduleImportation(alias.name, node) else: name = alias.asname or alias.name importation = Importation(name, node, alias.name) self.addBinding(node, importation) def IMPORTFROM(self, node): if node.module == '__future__': if not self.futuresAllowed: self.report(messages.LateFutureImport, node, [n.name for n in node.names]) else: self.futuresAllowed = False module = ('.' * node.level) + (node.module or '') for alias in node.names: name = alias.asname or alias.name if node.module == '__future__': importation = FutureImportation(name, node, self.scope) if alias.name not in __future__.all_feature_names: self.report(messages.FutureFeatureNotDefined, node, alias.name) if alias.name == 'annotations': self.annotationsFutureEnabled = True elif alias.name == '*': # Only Python 2, local import * is a SyntaxWarning if not PY2 and not isinstance(self.scope, ModuleScope): self.report(messages.ImportStarNotPermitted, node, module) continue self.scope.importStarred = True self.report(messages.ImportStarUsed, node, module) importation = StarImportation(module, node) else: importation = ImportationFrom(name, node, module, alias.name) self.addBinding(node, importation) def TRY(self, node): handler_names = [] # List the exception handlers for i, handler in enumerate(node.handlers): if isinstance(handler.type, ast.Tuple): for exc_type in handler.type.elts: handler_names.append(getNodeName(exc_type)) elif handler.type: handler_names.append(getNodeName(handler.type)) if handler.type is None and i < len(node.handlers) - 1: self.report(messages.DefaultExceptNotLast, handler) # Memorize the except handlers and process the body self.exceptHandlers.append(handler_names) for child in node.body: self.handleNode(child, node) self.exceptHandlers.pop() # Process the other nodes: "except:", "else:", "finally:" self.handleChildren(node, omit='body') TRYEXCEPT = TRY def EXCEPTHANDLER(self, node): if PY2 or node.name is None: self.handleChildren(node) return # If the name already exists in the scope, modify state of existing # binding. if node.name in self.scope: self.handleNodeStore(node) # 3.x: the name of the exception, which is not a Name node, but a # simple string, creates a local that is only bound within the scope of # the except: block. As such, temporarily remove the existing binding # to more accurately determine if the name is used in the except: # block. try: prev_definition = self.scope.pop(node.name) except KeyError: prev_definition = None self.handleNodeStore(node) self.handleChildren(node) # See discussion on https://github.com/PyCQA/pyflakes/pull/59 # We're removing the local name since it's being unbound after leaving # the except: block and it's always unbound if the except: block is # never entered. This will cause an "undefined name" error raised if # the checked code tries to use the name afterwards. # # Unless it's been removed already. Then do nothing. try: binding = self.scope.pop(node.name) except KeyError: pass else: if not binding.used: self.report(messages.UnusedVariable, node, node.name) # Restore. if prev_definition: self.scope[node.name] = prev_definition def ANNASSIGN(self, node): if node.value: # Only bind the *targets* if the assignment has a value. # Otherwise it's not really ast.Store and shouldn't silence # UndefinedLocal warnings. self.handleNode(node.target, node) self.handleAnnotation(node.annotation, node) if node.value: # If the assignment has value, handle the *value* now. self.handleNode(node.value, node) def COMPARE(self, node): literals = (ast.Str, ast.Num) if not PY2: literals += (ast.Bytes,) left = node.left for op, right in zip(node.ops, node.comparators): if (isinstance(op, (ast.Is, ast.IsNot)) and (isinstance(left, literals) or isinstance(right, literals))): self.report(messages.IsLiteral, node) left = right self.handleChildren(node)
PyCQA/pyflakes
pyflakes/checker.py
Checker.differentForks
python
def differentForks(self, lnode, rnode): ancestor = self.getCommonAncestor(lnode, rnode, self.root) parts = getAlternatives(ancestor) if parts: for items in parts: if self.descendantOf(lnode, items, ancestor) ^ \ self.descendantOf(rnode, items, ancestor): return True return False
True, if lnode and rnode are located on different forks of IF/TRY
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/checker.py#L860-L869
[ "def getAlternatives(n):\n if isinstance(n, ast.If):\n return [n.body]\n if isinstance(n, ast.Try):\n return [n.body + n.orelse] + [[hdl] for hdl in n.handlers]\n", "def getCommonAncestor(self, lnode, rnode, stop):\n if stop in (lnode, rnode) or not (hasattr(lnode, 'parent') and\n hasattr(rnode, 'parent')):\n return None\n if lnode is rnode:\n return lnode\n\n if (lnode.depth > rnode.depth):\n return self.getCommonAncestor(lnode.parent, rnode, stop)\n if (lnode.depth < rnode.depth):\n return self.getCommonAncestor(lnode, rnode.parent, stop)\n return self.getCommonAncestor(lnode.parent, rnode.parent, stop)\n", "def descendantOf(self, node, ancestors, stop):\n for a in ancestors:\n if self.getCommonAncestor(node, a, stop):\n return True\n return False\n" ]
class Checker(object): """ I check the cleanliness and sanity of Python code. @ivar _deferredFunctions: Tracking list used by L{deferFunction}. Elements of the list are two-tuples. The first element is the callable passed to L{deferFunction}. The second element is a copy of the scope stack at the time L{deferFunction} was called. @ivar _deferredAssignments: Similar to C{_deferredFunctions}, but for callables which are deferred assignment checks. """ _ast_node_scope = { ast.Module: ModuleScope, ast.ClassDef: ClassScope, ast.FunctionDef: FunctionScope, ast.Lambda: FunctionScope, ast.ListComp: GeneratorScope, ast.SetComp: GeneratorScope, ast.GeneratorExp: GeneratorScope, ast.DictComp: GeneratorScope, } if PY35_PLUS: _ast_node_scope[ast.AsyncFunctionDef] = FunctionScope, nodeDepth = 0 offset = None traceTree = False builtIns = set(builtin_vars).union(_MAGIC_GLOBALS) _customBuiltIns = os.environ.get('PYFLAKES_BUILTINS') if _customBuiltIns: builtIns.update(_customBuiltIns.split(',')) del _customBuiltIns # TODO: file_tokens= is required to perform checks on type comments, # eventually make this a required positional argument. For now it # is defaulted to `()` for api compatibility. def __init__(self, tree, filename='(none)', builtins=None, withDoctest='PYFLAKES_DOCTEST' in os.environ, file_tokens=()): self._nodeHandlers = {} self._deferredFunctions = [] self._deferredAssignments = [] self.deadScopes = [] self.messages = [] self.filename = filename if builtins: self.builtIns = self.builtIns.union(builtins) self.withDoctest = withDoctest try: self.scopeStack = [Checker._ast_node_scope[type(tree)]()] except KeyError: raise RuntimeError('No scope implemented for the node %r' % tree) self.exceptHandlers = [()] self.root = tree self._type_comments = _collect_type_comments(tree, file_tokens) for builtin in self.builtIns: self.addBinding(None, Builtin(builtin)) self.handleChildren(tree) self.runDeferred(self._deferredFunctions) # Set _deferredFunctions to None so that deferFunction will fail # noisily if called after we've run through the deferred functions. self._deferredFunctions = None self.runDeferred(self._deferredAssignments) # Set _deferredAssignments to None so that deferAssignment will fail # noisily if called after we've run through the deferred assignments. self._deferredAssignments = None del self.scopeStack[1:] self.popScope() self.checkDeadScopes() def deferFunction(self, callable): """ Schedule a function handler to be called just before completion. This is used for handling function bodies, which must be deferred because code later in the file might modify the global scope. When `callable` is called, the scope at the time this is called will be restored, however it will contain any new bindings added to it. """ self._deferredFunctions.append((callable, self.scopeStack[:], self.offset)) def deferAssignment(self, callable): """ Schedule an assignment handler to be called just after deferred function handlers. """ self._deferredAssignments.append((callable, self.scopeStack[:], self.offset)) def runDeferred(self, deferred): """ Run the callables in C{deferred} using their associated scope stack. """ for handler, scope, offset in deferred: self.scopeStack = scope self.offset = offset handler() def _in_doctest(self): return (len(self.scopeStack) >= 2 and isinstance(self.scopeStack[1], DoctestScope)) @property def futuresAllowed(self): if not all(isinstance(scope, ModuleScope) for scope in self.scopeStack): return False return self.scope._futures_allowed @futuresAllowed.setter def futuresAllowed(self, value): assert value is False if isinstance(self.scope, ModuleScope): self.scope._futures_allowed = False @property def annotationsFutureEnabled(self): scope = self.scopeStack[0] if not isinstance(scope, ModuleScope): return False return scope._annotations_future_enabled @annotationsFutureEnabled.setter def annotationsFutureEnabled(self, value): assert value is True assert isinstance(self.scope, ModuleScope) self.scope._annotations_future_enabled = True @property def scope(self): return self.scopeStack[-1] def popScope(self): self.deadScopes.append(self.scopeStack.pop()) def checkDeadScopes(self): """ Look at scopes which have been fully examined and report names in them which were imported but unused. """ for scope in self.deadScopes: # imports in classes are public members if isinstance(scope, ClassScope): continue all_binding = scope.get('__all__') if all_binding and not isinstance(all_binding, ExportBinding): all_binding = None if all_binding: all_names = set(all_binding.names) undefined = all_names.difference(scope) else: all_names = undefined = [] if undefined: if not scope.importStarred and \ os.path.basename(self.filename) != '__init__.py': # Look for possible mistakes in the export list for name in undefined: self.report(messages.UndefinedExport, scope['__all__'].source, name) # mark all import '*' as used by the undefined in __all__ if scope.importStarred: from_list = [] for binding in scope.values(): if isinstance(binding, StarImportation): binding.used = all_binding from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) for name in undefined: self.report(messages.ImportStarUsage, scope['__all__'].source, name, from_list) # Look for imported names that aren't used. for value in scope.values(): if isinstance(value, Importation): used = value.used or value.name in all_names if not used: messg = messages.UnusedImport self.report(messg, value.source, str(value)) for node in value.redefined: if isinstance(self.getParent(node), FOR_TYPES): messg = messages.ImportShadowedByLoopVar elif used: continue else: messg = messages.RedefinedWhileUnused self.report(messg, node, value.name, value.source) def pushScope(self, scopeClass=FunctionScope): self.scopeStack.append(scopeClass()) def report(self, messageClass, *args, **kwargs): self.messages.append(messageClass(self.filename, *args, **kwargs)) def getParent(self, node): # Lookup the first parent which is not Tuple, List or Starred while True: node = node.parent if not hasattr(node, 'elts') and not hasattr(node, 'ctx'): return node def getCommonAncestor(self, lnode, rnode, stop): if stop in (lnode, rnode) or not (hasattr(lnode, 'parent') and hasattr(rnode, 'parent')): return None if lnode is rnode: return lnode if (lnode.depth > rnode.depth): return self.getCommonAncestor(lnode.parent, rnode, stop) if (lnode.depth < rnode.depth): return self.getCommonAncestor(lnode, rnode.parent, stop) return self.getCommonAncestor(lnode.parent, rnode.parent, stop) def descendantOf(self, node, ancestors, stop): for a in ancestors: if self.getCommonAncestor(node, a, stop): return True return False def _getAncestor(self, node, ancestor_type): parent = node while True: if parent is self.root: return None parent = self.getParent(parent) if isinstance(parent, ancestor_type): return parent def getScopeNode(self, node): return self._getAncestor(node, tuple(Checker._ast_node_scope.keys())) def addBinding(self, node, value): """ Called when a binding is altered. - `node` is the statement responsible for the change - `value` is the new value, a Binding instance """ # assert value.source in (node, node.parent): for scope in self.scopeStack[::-1]: if value.name in scope: break existing = scope.get(value.name) if (existing and not isinstance(existing, Builtin) and not self.differentForks(node, existing.source)): parent_stmt = self.getParent(value.source) if isinstance(existing, Importation) and isinstance(parent_stmt, FOR_TYPES): self.report(messages.ImportShadowedByLoopVar, node, value.name, existing.source) elif scope is self.scope: if (isinstance(parent_stmt, ast.comprehension) and not isinstance(self.getParent(existing.source), (FOR_TYPES, ast.comprehension))): self.report(messages.RedefinedInListComp, node, value.name, existing.source) elif not existing.used and value.redefines(existing): if value.name != '_' or isinstance(existing, Importation): if not is_typing_overload(existing, self.scopeStack): self.report(messages.RedefinedWhileUnused, node, value.name, existing.source) elif isinstance(existing, Importation) and value.redefines(existing): existing.redefined.append(node) if value.name in self.scope: # then assume the rebound name is used as a global or within a loop value.used = self.scope[value.name].used self.scope[value.name] = value def getNodeHandler(self, node_class): try: return self._nodeHandlers[node_class] except KeyError: nodeType = getNodeType(node_class) self._nodeHandlers[node_class] = handler = getattr(self, nodeType) return handler def handleNodeLoad(self, node): name = getNodeName(node) if not name: return in_generators = None importStarred = None # try enclosing function scopes and global scope for scope in self.scopeStack[-1::-1]: if isinstance(scope, ClassScope): if not PY2 and name == '__class__': return elif in_generators is False: # only generators used in a class scope can access the # names of the class. this is skipped during the first # iteration continue if (name == 'print' and isinstance(scope.get(name, None), Builtin)): parent = self.getParent(node) if (isinstance(parent, ast.BinOp) and isinstance(parent.op, ast.RShift)): self.report(messages.InvalidPrintSyntax, node) try: scope[name].used = (self.scope, node) # if the name of SubImportation is same as # alias of other Importation and the alias # is used, SubImportation also should be marked as used. n = scope[name] if isinstance(n, Importation) and n._has_alias(): try: scope[n.fullName].used = (self.scope, node) except KeyError: pass except KeyError: pass else: return importStarred = importStarred or scope.importStarred if in_generators is not False: in_generators = isinstance(scope, GeneratorScope) if importStarred: from_list = [] for scope in self.scopeStack[-1::-1]: for binding in scope.values(): if isinstance(binding, StarImportation): # mark '*' imports as used for each scope binding.used = (self.scope, node) from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) self.report(messages.ImportStarUsage, node, name, from_list) return if name == '__path__' and os.path.basename(self.filename) == '__init__.py': # the special name __path__ is valid only in packages return if name == '__module__' and isinstance(self.scope, ClassScope): return # protected with a NameError handler? if 'NameError' not in self.exceptHandlers[-1]: self.report(messages.UndefinedName, node, name) def handleNodeStore(self, node): name = getNodeName(node) if not name: return # if the name hasn't already been defined in the current scope if isinstance(self.scope, FunctionScope) and name not in self.scope: # for each function or module scope above us for scope in self.scopeStack[:-1]: if not isinstance(scope, (FunctionScope, ModuleScope)): continue # if the name was defined in that scope, and the name has # been accessed already in the current scope, and hasn't # been declared global used = name in scope and scope[name].used if used and used[0] is self.scope and name not in self.scope.globals: # then it's probably a mistake self.report(messages.UndefinedLocal, scope[name].used[1], name, scope[name].source) break parent_stmt = self.getParent(node) if isinstance(parent_stmt, (FOR_TYPES, ast.comprehension)) or ( parent_stmt != node.parent and not self.isLiteralTupleUnpacking(parent_stmt)): binding = Binding(name, node) elif name == '__all__' and isinstance(self.scope, ModuleScope): binding = ExportBinding(name, node.parent, self.scope) elif isinstance(getattr(node, 'ctx', None), ast.Param): binding = Argument(name, self.getScopeNode(node)) else: binding = Assignment(name, node) self.addBinding(node, binding) def handleNodeDelete(self, node): def on_conditional_branch(): """ Return `True` if node is part of a conditional body. """ current = getattr(node, 'parent', None) while current: if isinstance(current, (ast.If, ast.While, ast.IfExp)): return True current = getattr(current, 'parent', None) return False name = getNodeName(node) if not name: return if on_conditional_branch(): # We cannot predict if this conditional branch is going to # be executed. return if isinstance(self.scope, FunctionScope) and name in self.scope.globals: self.scope.globals.remove(name) else: try: del self.scope[name] except KeyError: self.report(messages.UndefinedName, node, name) def _handle_type_comments(self, node): for (lineno, col_offset), comment in self._type_comments.get(node, ()): comment = comment.split(':', 1)[1].strip() func_match = TYPE_FUNC_RE.match(comment) if func_match: parts = ( func_match.group(1).replace('*', ''), func_match.group(2).strip(), ) else: parts = (comment,) for part in parts: if PY2: part = part.replace('...', 'Ellipsis') self.deferFunction(functools.partial( self.handleStringAnnotation, part, DummyNode(lineno, col_offset), lineno, col_offset, messages.CommentAnnotationSyntaxError, )) def handleChildren(self, tree, omit=None): self._handle_type_comments(tree) for node in iter_child_nodes(tree, omit=omit): self.handleNode(node, tree) def isLiteralTupleUnpacking(self, node): if isinstance(node, ast.Assign): for child in node.targets + [node.value]: if not hasattr(child, 'elts'): return False return True def isDocstring(self, node): """ Determine if the given node is a docstring, as long as it is at the correct place in the node tree. """ return isinstance(node, ast.Str) or (isinstance(node, ast.Expr) and isinstance(node.value, ast.Str)) def getDocstring(self, node): if isinstance(node, ast.Expr): node = node.value if not isinstance(node, ast.Str): return (None, None) if PYPY or PY38_PLUS: doctest_lineno = node.lineno - 1 else: # Computed incorrectly if the docstring has backslash doctest_lineno = node.lineno - node.s.count('\n') - 1 return (node.s, doctest_lineno) def handleNode(self, node, parent): if node is None: return if self.offset and getattr(node, 'lineno', None) is not None: node.lineno += self.offset[0] node.col_offset += self.offset[1] if self.traceTree: print(' ' * self.nodeDepth + node.__class__.__name__) if self.futuresAllowed and not (isinstance(node, ast.ImportFrom) or self.isDocstring(node)): self.futuresAllowed = False self.nodeDepth += 1 node.depth = self.nodeDepth node.parent = parent try: handler = self.getNodeHandler(node.__class__) handler(node) finally: self.nodeDepth -= 1 if self.traceTree: print(' ' * self.nodeDepth + 'end ' + node.__class__.__name__) _getDoctestExamples = doctest.DocTestParser().get_examples def handleDoctests(self, node): try: if hasattr(node, 'docstring'): docstring = node.docstring # This is just a reasonable guess. In Python 3.7, docstrings no # longer have line numbers associated with them. This will be # incorrect if there are empty lines between the beginning # of the function and the docstring. node_lineno = node.lineno if hasattr(node, 'args'): node_lineno = max([node_lineno] + [arg.lineno for arg in node.args.args]) else: (docstring, node_lineno) = self.getDocstring(node.body[0]) examples = docstring and self._getDoctestExamples(docstring) except (ValueError, IndexError): # e.g. line 6 of the docstring for <string> has inconsistent # leading whitespace: ... return if not examples: return # Place doctest in module scope saved_stack = self.scopeStack self.scopeStack = [self.scopeStack[0]] node_offset = self.offset or (0, 0) self.pushScope(DoctestScope) self.addBinding(None, Builtin('_')) for example in examples: try: tree = ast.parse(example.source, "<doctest>") except SyntaxError: e = sys.exc_info()[1] if PYPY: e.offset += 1 position = (node_lineno + example.lineno + e.lineno, example.indent + 4 + (e.offset or 0)) self.report(messages.DoctestSyntaxError, node, position) else: self.offset = (node_offset[0] + node_lineno + example.lineno, node_offset[1] + example.indent + 4) self.handleChildren(tree) self.offset = node_offset self.popScope() self.scopeStack = saved_stack def handleStringAnnotation(self, s, node, ref_lineno, ref_col_offset, err): try: tree = ast.parse(s) except SyntaxError: self.report(err, node, s) return body = tree.body if len(body) != 1 or not isinstance(body[0], ast.Expr): self.report(err, node, s) return parsed_annotation = tree.body[0].value for descendant in ast.walk(parsed_annotation): if ( 'lineno' in descendant._attributes and 'col_offset' in descendant._attributes ): descendant.lineno = ref_lineno descendant.col_offset = ref_col_offset self.handleNode(parsed_annotation, node) def handleAnnotation(self, annotation, node): if isinstance(annotation, ast.Str): # Defer handling forward annotation. self.deferFunction(functools.partial( self.handleStringAnnotation, annotation.s, node, annotation.lineno, annotation.col_offset, messages.ForwardAnnotationSyntaxError, )) elif self.annotationsFutureEnabled: self.deferFunction(lambda: self.handleNode(annotation, node)) else: self.handleNode(annotation, node) def ignore(self, node): pass # "stmt" type nodes DELETE = PRINT = FOR = ASYNCFOR = WHILE = IF = WITH = WITHITEM = \ ASYNCWITH = ASYNCWITHITEM = TRYFINALLY = EXEC = \ EXPR = ASSIGN = handleChildren PASS = ignore # "expr" type nodes BOOLOP = BINOP = UNARYOP = IFEXP = SET = \ CALL = REPR = ATTRIBUTE = SUBSCRIPT = \ STARRED = NAMECONSTANT = handleChildren NUM = STR = BYTES = ELLIPSIS = CONSTANT = ignore # "slice" type nodes SLICE = EXTSLICE = INDEX = handleChildren # expression contexts are node instances too, though being constants LOAD = STORE = DEL = AUGLOAD = AUGSTORE = PARAM = ignore # same for operators AND = OR = ADD = SUB = MULT = DIV = MOD = POW = LSHIFT = RSHIFT = \ BITOR = BITXOR = BITAND = FLOORDIV = INVERT = NOT = UADD = USUB = \ EQ = NOTEQ = LT = LTE = GT = GTE = IS = ISNOT = IN = NOTIN = \ MATMULT = ignore def RAISE(self, node): self.handleChildren(node) arg = get_raise_argument(node) if isinstance(arg, ast.Call): if is_notimplemented_name_node(arg.func): # Handle "raise NotImplemented(...)" self.report(messages.RaiseNotImplemented, node) elif is_notimplemented_name_node(arg): # Handle "raise NotImplemented" self.report(messages.RaiseNotImplemented, node) # additional node types COMPREHENSION = KEYWORD = FORMATTEDVALUE = JOINEDSTR = handleChildren def DICT(self, node): # Complain if there are duplicate keys with different values # If they have the same value it's not going to cause potentially # unexpected behaviour so we'll not complain. keys = [ convert_to_value(key) for key in node.keys ] key_counts = counter(keys) duplicate_keys = [ key for key, count in key_counts.items() if count > 1 ] for key in duplicate_keys: key_indices = [i for i, i_key in enumerate(keys) if i_key == key] values = counter( convert_to_value(node.values[index]) for index in key_indices ) if any(count == 1 for value, count in values.items()): for key_index in key_indices: key_node = node.keys[key_index] if isinstance(key, VariableKey): self.report(messages.MultiValueRepeatedKeyVariable, key_node, key.name) else: self.report( messages.MultiValueRepeatedKeyLiteral, key_node, key, ) self.handleChildren(node) def ASSERT(self, node): if isinstance(node.test, ast.Tuple) and node.test.elts != []: self.report(messages.AssertTuple, node) self.handleChildren(node) def GLOBAL(self, node): """ Keep track of globals declarations. """ global_scope_index = 1 if self._in_doctest() else 0 global_scope = self.scopeStack[global_scope_index] # Ignore 'global' statement in global scope. if self.scope is not global_scope: # One 'global' statement can bind multiple (comma-delimited) names. for node_name in node.names: node_value = Assignment(node_name, node) # Remove UndefinedName messages already reported for this name. # TODO: if the global is not used in this scope, it does not # become a globally defined name. See test_unused_global. self.messages = [ m for m in self.messages if not isinstance(m, messages.UndefinedName) or m.message_args[0] != node_name] # Bind name to global scope if it doesn't exist already. global_scope.setdefault(node_name, node_value) # Bind name to non-global scopes, but as already "used". node_value.used = (global_scope, node) for scope in self.scopeStack[global_scope_index + 1:]: scope[node_name] = node_value NONLOCAL = GLOBAL def GENERATOREXP(self, node): self.pushScope(GeneratorScope) self.handleChildren(node) self.popScope() LISTCOMP = handleChildren if PY2 else GENERATOREXP DICTCOMP = SETCOMP = GENERATOREXP def NAME(self, node): """ Handle occurrence of Name (which can be a load/store/delete access.) """ # Locate the name in locals / function / globals scopes. if isinstance(node.ctx, (ast.Load, ast.AugLoad)): self.handleNodeLoad(node) if (node.id == 'locals' and isinstance(self.scope, FunctionScope) and isinstance(node.parent, ast.Call)): # we are doing locals() call in current scope self.scope.usesLocals = True elif isinstance(node.ctx, (ast.Store, ast.AugStore, ast.Param)): self.handleNodeStore(node) elif isinstance(node.ctx, ast.Del): self.handleNodeDelete(node) else: # Unknown context raise RuntimeError("Got impossible expression context: %r" % (node.ctx,)) def CONTINUE(self, node): # Walk the tree up until we see a loop (OK), a function or class # definition (not OK), for 'continue', a finally block (not OK), or # the top module scope (not OK) n = node while hasattr(n, 'parent'): n, n_child = n.parent, n if isinstance(n, LOOP_TYPES): # Doesn't apply unless it's in the loop itself if n_child not in n.orelse: return if isinstance(n, (ast.FunctionDef, ast.ClassDef)): break # Handle Try/TryFinally difference in Python < and >= 3.3 if hasattr(n, 'finalbody') and isinstance(node, ast.Continue): if n_child in n.finalbody: self.report(messages.ContinueInFinally, node) return if isinstance(node, ast.Continue): self.report(messages.ContinueOutsideLoop, node) else: # ast.Break self.report(messages.BreakOutsideLoop, node) BREAK = CONTINUE def RETURN(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.ReturnOutsideFunction, node) return if ( node.value and hasattr(self.scope, 'returnValue') and not self.scope.returnValue ): self.scope.returnValue = node.value self.handleNode(node.value, node) def YIELD(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.YieldOutsideFunction, node) return self.scope.isGenerator = True self.handleNode(node.value, node) AWAIT = YIELDFROM = YIELD def FUNCTIONDEF(self, node): for deco in node.decorator_list: self.handleNode(deco, node) self.LAMBDA(node) self.addBinding(node, FunctionDefinition(node.name, node)) # doctest does not process doctest within a doctest, # or in nested functions. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) ASYNCFUNCTIONDEF = FUNCTIONDEF def LAMBDA(self, node): args = [] annotations = [] if PY2: def addArgs(arglist): for arg in arglist: if isinstance(arg, ast.Tuple): addArgs(arg.elts) else: args.append(arg.id) addArgs(node.args.args) defaults = node.args.defaults else: for arg in node.args.args + node.args.kwonlyargs: args.append(arg.arg) annotations.append(arg.annotation) defaults = node.args.defaults + node.args.kw_defaults # Only for Python3 FunctionDefs is_py3_func = hasattr(node, 'returns') for arg_name in ('vararg', 'kwarg'): wildcard = getattr(node.args, arg_name) if not wildcard: continue args.append(wildcard if PY2 else wildcard.arg) if is_py3_func: if PY2: # Python 2.7 argannotation = arg_name + 'annotation' annotations.append(getattr(node.args, argannotation)) else: # Python >= 3.4 annotations.append(wildcard.annotation) if is_py3_func: annotations.append(node.returns) if len(set(args)) < len(args): for (idx, arg) in enumerate(args): if arg in args[:idx]: self.report(messages.DuplicateArgument, node, arg) for annotation in annotations: self.handleAnnotation(annotation, node) for default in defaults: self.handleNode(default, node) def runFunction(): self.pushScope() self.handleChildren(node, omit='decorator_list') def checkUnusedAssignments(): """ Check to see if any assignments have not been used. """ for name, binding in self.scope.unusedAssignments(): self.report(messages.UnusedVariable, binding.source, name) self.deferAssignment(checkUnusedAssignments) if PY2: def checkReturnWithArgumentInsideGenerator(): """ Check to see if there is any return statement with arguments but the function is a generator. """ if self.scope.isGenerator and self.scope.returnValue: self.report(messages.ReturnWithArgsInsideGenerator, self.scope.returnValue) self.deferAssignment(checkReturnWithArgumentInsideGenerator) self.popScope() self.deferFunction(runFunction) def ARGUMENTS(self, node): self.handleChildren(node, omit=('defaults', 'kw_defaults')) if PY2: scope_node = self.getScopeNode(node) if node.vararg: self.addBinding(node, Argument(node.vararg, scope_node)) if node.kwarg: self.addBinding(node, Argument(node.kwarg, scope_node)) def ARG(self, node): self.addBinding(node, Argument(node.arg, self.getScopeNode(node))) def CLASSDEF(self, node): """ Check names used in a class definition, including its decorators, base classes, and the body of its definition. Additionally, add its name to the current scope. """ for deco in node.decorator_list: self.handleNode(deco, node) for baseNode in node.bases: self.handleNode(baseNode, node) if not PY2: for keywordNode in node.keywords: self.handleNode(keywordNode, node) self.pushScope(ClassScope) # doctest does not process doctest within a doctest # classes within classes are processed. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) for stmt in node.body: self.handleNode(stmt, node) self.popScope() self.addBinding(node, ClassDefinition(node.name, node)) def AUGASSIGN(self, node): self.handleNodeLoad(node.target) self.handleNode(node.value, node) self.handleNode(node.target, node) def TUPLE(self, node): if not PY2 and isinstance(node.ctx, ast.Store): # Python 3 advanced tuple unpacking: a, *b, c = d. # Only one starred expression is allowed, and no more than 1<<8 # assignments are allowed before a stared expression. There is # also a limit of 1<<24 expressions after the starred expression, # which is impossible to test due to memory restrictions, but we # add it here anyway has_starred = False star_loc = -1 for i, n in enumerate(node.elts): if isinstance(n, ast.Starred): if has_starred: self.report(messages.TwoStarredExpressions, node) # The SyntaxError doesn't distinguish two from more # than two. break has_starred = True star_loc = i if star_loc >= 1 << 8 or len(node.elts) - star_loc - 1 >= 1 << 24: self.report(messages.TooManyExpressionsInStarredAssignment, node) self.handleChildren(node) LIST = TUPLE def IMPORT(self, node): for alias in node.names: if '.' in alias.name and not alias.asname: importation = SubmoduleImportation(alias.name, node) else: name = alias.asname or alias.name importation = Importation(name, node, alias.name) self.addBinding(node, importation) def IMPORTFROM(self, node): if node.module == '__future__': if not self.futuresAllowed: self.report(messages.LateFutureImport, node, [n.name for n in node.names]) else: self.futuresAllowed = False module = ('.' * node.level) + (node.module or '') for alias in node.names: name = alias.asname or alias.name if node.module == '__future__': importation = FutureImportation(name, node, self.scope) if alias.name not in __future__.all_feature_names: self.report(messages.FutureFeatureNotDefined, node, alias.name) if alias.name == 'annotations': self.annotationsFutureEnabled = True elif alias.name == '*': # Only Python 2, local import * is a SyntaxWarning if not PY2 and not isinstance(self.scope, ModuleScope): self.report(messages.ImportStarNotPermitted, node, module) continue self.scope.importStarred = True self.report(messages.ImportStarUsed, node, module) importation = StarImportation(module, node) else: importation = ImportationFrom(name, node, module, alias.name) self.addBinding(node, importation) def TRY(self, node): handler_names = [] # List the exception handlers for i, handler in enumerate(node.handlers): if isinstance(handler.type, ast.Tuple): for exc_type in handler.type.elts: handler_names.append(getNodeName(exc_type)) elif handler.type: handler_names.append(getNodeName(handler.type)) if handler.type is None and i < len(node.handlers) - 1: self.report(messages.DefaultExceptNotLast, handler) # Memorize the except handlers and process the body self.exceptHandlers.append(handler_names) for child in node.body: self.handleNode(child, node) self.exceptHandlers.pop() # Process the other nodes: "except:", "else:", "finally:" self.handleChildren(node, omit='body') TRYEXCEPT = TRY def EXCEPTHANDLER(self, node): if PY2 or node.name is None: self.handleChildren(node) return # If the name already exists in the scope, modify state of existing # binding. if node.name in self.scope: self.handleNodeStore(node) # 3.x: the name of the exception, which is not a Name node, but a # simple string, creates a local that is only bound within the scope of # the except: block. As such, temporarily remove the existing binding # to more accurately determine if the name is used in the except: # block. try: prev_definition = self.scope.pop(node.name) except KeyError: prev_definition = None self.handleNodeStore(node) self.handleChildren(node) # See discussion on https://github.com/PyCQA/pyflakes/pull/59 # We're removing the local name since it's being unbound after leaving # the except: block and it's always unbound if the except: block is # never entered. This will cause an "undefined name" error raised if # the checked code tries to use the name afterwards. # # Unless it's been removed already. Then do nothing. try: binding = self.scope.pop(node.name) except KeyError: pass else: if not binding.used: self.report(messages.UnusedVariable, node, node.name) # Restore. if prev_definition: self.scope[node.name] = prev_definition def ANNASSIGN(self, node): if node.value: # Only bind the *targets* if the assignment has a value. # Otherwise it's not really ast.Store and shouldn't silence # UndefinedLocal warnings. self.handleNode(node.target, node) self.handleAnnotation(node.annotation, node) if node.value: # If the assignment has value, handle the *value* now. self.handleNode(node.value, node) def COMPARE(self, node): literals = (ast.Str, ast.Num) if not PY2: literals += (ast.Bytes,) left = node.left for op, right in zip(node.ops, node.comparators): if (isinstance(op, (ast.Is, ast.IsNot)) and (isinstance(left, literals) or isinstance(right, literals))): self.report(messages.IsLiteral, node) left = right self.handleChildren(node)
PyCQA/pyflakes
pyflakes/checker.py
Checker.addBinding
python
def addBinding(self, node, value): # assert value.source in (node, node.parent): for scope in self.scopeStack[::-1]: if value.name in scope: break existing = scope.get(value.name) if (existing and not isinstance(existing, Builtin) and not self.differentForks(node, existing.source)): parent_stmt = self.getParent(value.source) if isinstance(existing, Importation) and isinstance(parent_stmt, FOR_TYPES): self.report(messages.ImportShadowedByLoopVar, node, value.name, existing.source) elif scope is self.scope: if (isinstance(parent_stmt, ast.comprehension) and not isinstance(self.getParent(existing.source), (FOR_TYPES, ast.comprehension))): self.report(messages.RedefinedInListComp, node, value.name, existing.source) elif not existing.used and value.redefines(existing): if value.name != '_' or isinstance(existing, Importation): if not is_typing_overload(existing, self.scopeStack): self.report(messages.RedefinedWhileUnused, node, value.name, existing.source) elif isinstance(existing, Importation) and value.redefines(existing): existing.redefined.append(node) if value.name in self.scope: # then assume the rebound name is used as a global or within a loop value.used = self.scope[value.name].used self.scope[value.name] = value
Called when a binding is altered. - `node` is the statement responsible for the change - `value` is the new value, a Binding instance
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/checker.py#L871-L911
[ "def differentForks(self, lnode, rnode):\n \"\"\"True, if lnode and rnode are located on different forks of IF/TRY\"\"\"\n ancestor = self.getCommonAncestor(lnode, rnode, self.root)\n parts = getAlternatives(ancestor)\n if parts:\n for items in parts:\n if self.descendantOf(lnode, items, ancestor) ^ \\\n self.descendantOf(rnode, items, ancestor):\n return True\n return False\n" ]
class Checker(object): """ I check the cleanliness and sanity of Python code. @ivar _deferredFunctions: Tracking list used by L{deferFunction}. Elements of the list are two-tuples. The first element is the callable passed to L{deferFunction}. The second element is a copy of the scope stack at the time L{deferFunction} was called. @ivar _deferredAssignments: Similar to C{_deferredFunctions}, but for callables which are deferred assignment checks. """ _ast_node_scope = { ast.Module: ModuleScope, ast.ClassDef: ClassScope, ast.FunctionDef: FunctionScope, ast.Lambda: FunctionScope, ast.ListComp: GeneratorScope, ast.SetComp: GeneratorScope, ast.GeneratorExp: GeneratorScope, ast.DictComp: GeneratorScope, } if PY35_PLUS: _ast_node_scope[ast.AsyncFunctionDef] = FunctionScope, nodeDepth = 0 offset = None traceTree = False builtIns = set(builtin_vars).union(_MAGIC_GLOBALS) _customBuiltIns = os.environ.get('PYFLAKES_BUILTINS') if _customBuiltIns: builtIns.update(_customBuiltIns.split(',')) del _customBuiltIns # TODO: file_tokens= is required to perform checks on type comments, # eventually make this a required positional argument. For now it # is defaulted to `()` for api compatibility. def __init__(self, tree, filename='(none)', builtins=None, withDoctest='PYFLAKES_DOCTEST' in os.environ, file_tokens=()): self._nodeHandlers = {} self._deferredFunctions = [] self._deferredAssignments = [] self.deadScopes = [] self.messages = [] self.filename = filename if builtins: self.builtIns = self.builtIns.union(builtins) self.withDoctest = withDoctest try: self.scopeStack = [Checker._ast_node_scope[type(tree)]()] except KeyError: raise RuntimeError('No scope implemented for the node %r' % tree) self.exceptHandlers = [()] self.root = tree self._type_comments = _collect_type_comments(tree, file_tokens) for builtin in self.builtIns: self.addBinding(None, Builtin(builtin)) self.handleChildren(tree) self.runDeferred(self._deferredFunctions) # Set _deferredFunctions to None so that deferFunction will fail # noisily if called after we've run through the deferred functions. self._deferredFunctions = None self.runDeferred(self._deferredAssignments) # Set _deferredAssignments to None so that deferAssignment will fail # noisily if called after we've run through the deferred assignments. self._deferredAssignments = None del self.scopeStack[1:] self.popScope() self.checkDeadScopes() def deferFunction(self, callable): """ Schedule a function handler to be called just before completion. This is used for handling function bodies, which must be deferred because code later in the file might modify the global scope. When `callable` is called, the scope at the time this is called will be restored, however it will contain any new bindings added to it. """ self._deferredFunctions.append((callable, self.scopeStack[:], self.offset)) def deferAssignment(self, callable): """ Schedule an assignment handler to be called just after deferred function handlers. """ self._deferredAssignments.append((callable, self.scopeStack[:], self.offset)) def runDeferred(self, deferred): """ Run the callables in C{deferred} using their associated scope stack. """ for handler, scope, offset in deferred: self.scopeStack = scope self.offset = offset handler() def _in_doctest(self): return (len(self.scopeStack) >= 2 and isinstance(self.scopeStack[1], DoctestScope)) @property def futuresAllowed(self): if not all(isinstance(scope, ModuleScope) for scope in self.scopeStack): return False return self.scope._futures_allowed @futuresAllowed.setter def futuresAllowed(self, value): assert value is False if isinstance(self.scope, ModuleScope): self.scope._futures_allowed = False @property def annotationsFutureEnabled(self): scope = self.scopeStack[0] if not isinstance(scope, ModuleScope): return False return scope._annotations_future_enabled @annotationsFutureEnabled.setter def annotationsFutureEnabled(self, value): assert value is True assert isinstance(self.scope, ModuleScope) self.scope._annotations_future_enabled = True @property def scope(self): return self.scopeStack[-1] def popScope(self): self.deadScopes.append(self.scopeStack.pop()) def checkDeadScopes(self): """ Look at scopes which have been fully examined and report names in them which were imported but unused. """ for scope in self.deadScopes: # imports in classes are public members if isinstance(scope, ClassScope): continue all_binding = scope.get('__all__') if all_binding and not isinstance(all_binding, ExportBinding): all_binding = None if all_binding: all_names = set(all_binding.names) undefined = all_names.difference(scope) else: all_names = undefined = [] if undefined: if not scope.importStarred and \ os.path.basename(self.filename) != '__init__.py': # Look for possible mistakes in the export list for name in undefined: self.report(messages.UndefinedExport, scope['__all__'].source, name) # mark all import '*' as used by the undefined in __all__ if scope.importStarred: from_list = [] for binding in scope.values(): if isinstance(binding, StarImportation): binding.used = all_binding from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) for name in undefined: self.report(messages.ImportStarUsage, scope['__all__'].source, name, from_list) # Look for imported names that aren't used. for value in scope.values(): if isinstance(value, Importation): used = value.used or value.name in all_names if not used: messg = messages.UnusedImport self.report(messg, value.source, str(value)) for node in value.redefined: if isinstance(self.getParent(node), FOR_TYPES): messg = messages.ImportShadowedByLoopVar elif used: continue else: messg = messages.RedefinedWhileUnused self.report(messg, node, value.name, value.source) def pushScope(self, scopeClass=FunctionScope): self.scopeStack.append(scopeClass()) def report(self, messageClass, *args, **kwargs): self.messages.append(messageClass(self.filename, *args, **kwargs)) def getParent(self, node): # Lookup the first parent which is not Tuple, List or Starred while True: node = node.parent if not hasattr(node, 'elts') and not hasattr(node, 'ctx'): return node def getCommonAncestor(self, lnode, rnode, stop): if stop in (lnode, rnode) or not (hasattr(lnode, 'parent') and hasattr(rnode, 'parent')): return None if lnode is rnode: return lnode if (lnode.depth > rnode.depth): return self.getCommonAncestor(lnode.parent, rnode, stop) if (lnode.depth < rnode.depth): return self.getCommonAncestor(lnode, rnode.parent, stop) return self.getCommonAncestor(lnode.parent, rnode.parent, stop) def descendantOf(self, node, ancestors, stop): for a in ancestors: if self.getCommonAncestor(node, a, stop): return True return False def _getAncestor(self, node, ancestor_type): parent = node while True: if parent is self.root: return None parent = self.getParent(parent) if isinstance(parent, ancestor_type): return parent def getScopeNode(self, node): return self._getAncestor(node, tuple(Checker._ast_node_scope.keys())) def differentForks(self, lnode, rnode): """True, if lnode and rnode are located on different forks of IF/TRY""" ancestor = self.getCommonAncestor(lnode, rnode, self.root) parts = getAlternatives(ancestor) if parts: for items in parts: if self.descendantOf(lnode, items, ancestor) ^ \ self.descendantOf(rnode, items, ancestor): return True return False def getNodeHandler(self, node_class): try: return self._nodeHandlers[node_class] except KeyError: nodeType = getNodeType(node_class) self._nodeHandlers[node_class] = handler = getattr(self, nodeType) return handler def handleNodeLoad(self, node): name = getNodeName(node) if not name: return in_generators = None importStarred = None # try enclosing function scopes and global scope for scope in self.scopeStack[-1::-1]: if isinstance(scope, ClassScope): if not PY2 and name == '__class__': return elif in_generators is False: # only generators used in a class scope can access the # names of the class. this is skipped during the first # iteration continue if (name == 'print' and isinstance(scope.get(name, None), Builtin)): parent = self.getParent(node) if (isinstance(parent, ast.BinOp) and isinstance(parent.op, ast.RShift)): self.report(messages.InvalidPrintSyntax, node) try: scope[name].used = (self.scope, node) # if the name of SubImportation is same as # alias of other Importation and the alias # is used, SubImportation also should be marked as used. n = scope[name] if isinstance(n, Importation) and n._has_alias(): try: scope[n.fullName].used = (self.scope, node) except KeyError: pass except KeyError: pass else: return importStarred = importStarred or scope.importStarred if in_generators is not False: in_generators = isinstance(scope, GeneratorScope) if importStarred: from_list = [] for scope in self.scopeStack[-1::-1]: for binding in scope.values(): if isinstance(binding, StarImportation): # mark '*' imports as used for each scope binding.used = (self.scope, node) from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) self.report(messages.ImportStarUsage, node, name, from_list) return if name == '__path__' and os.path.basename(self.filename) == '__init__.py': # the special name __path__ is valid only in packages return if name == '__module__' and isinstance(self.scope, ClassScope): return # protected with a NameError handler? if 'NameError' not in self.exceptHandlers[-1]: self.report(messages.UndefinedName, node, name) def handleNodeStore(self, node): name = getNodeName(node) if not name: return # if the name hasn't already been defined in the current scope if isinstance(self.scope, FunctionScope) and name not in self.scope: # for each function or module scope above us for scope in self.scopeStack[:-1]: if not isinstance(scope, (FunctionScope, ModuleScope)): continue # if the name was defined in that scope, and the name has # been accessed already in the current scope, and hasn't # been declared global used = name in scope and scope[name].used if used and used[0] is self.scope and name not in self.scope.globals: # then it's probably a mistake self.report(messages.UndefinedLocal, scope[name].used[1], name, scope[name].source) break parent_stmt = self.getParent(node) if isinstance(parent_stmt, (FOR_TYPES, ast.comprehension)) or ( parent_stmt != node.parent and not self.isLiteralTupleUnpacking(parent_stmt)): binding = Binding(name, node) elif name == '__all__' and isinstance(self.scope, ModuleScope): binding = ExportBinding(name, node.parent, self.scope) elif isinstance(getattr(node, 'ctx', None), ast.Param): binding = Argument(name, self.getScopeNode(node)) else: binding = Assignment(name, node) self.addBinding(node, binding) def handleNodeDelete(self, node): def on_conditional_branch(): """ Return `True` if node is part of a conditional body. """ current = getattr(node, 'parent', None) while current: if isinstance(current, (ast.If, ast.While, ast.IfExp)): return True current = getattr(current, 'parent', None) return False name = getNodeName(node) if not name: return if on_conditional_branch(): # We cannot predict if this conditional branch is going to # be executed. return if isinstance(self.scope, FunctionScope) and name in self.scope.globals: self.scope.globals.remove(name) else: try: del self.scope[name] except KeyError: self.report(messages.UndefinedName, node, name) def _handle_type_comments(self, node): for (lineno, col_offset), comment in self._type_comments.get(node, ()): comment = comment.split(':', 1)[1].strip() func_match = TYPE_FUNC_RE.match(comment) if func_match: parts = ( func_match.group(1).replace('*', ''), func_match.group(2).strip(), ) else: parts = (comment,) for part in parts: if PY2: part = part.replace('...', 'Ellipsis') self.deferFunction(functools.partial( self.handleStringAnnotation, part, DummyNode(lineno, col_offset), lineno, col_offset, messages.CommentAnnotationSyntaxError, )) def handleChildren(self, tree, omit=None): self._handle_type_comments(tree) for node in iter_child_nodes(tree, omit=omit): self.handleNode(node, tree) def isLiteralTupleUnpacking(self, node): if isinstance(node, ast.Assign): for child in node.targets + [node.value]: if not hasattr(child, 'elts'): return False return True def isDocstring(self, node): """ Determine if the given node is a docstring, as long as it is at the correct place in the node tree. """ return isinstance(node, ast.Str) or (isinstance(node, ast.Expr) and isinstance(node.value, ast.Str)) def getDocstring(self, node): if isinstance(node, ast.Expr): node = node.value if not isinstance(node, ast.Str): return (None, None) if PYPY or PY38_PLUS: doctest_lineno = node.lineno - 1 else: # Computed incorrectly if the docstring has backslash doctest_lineno = node.lineno - node.s.count('\n') - 1 return (node.s, doctest_lineno) def handleNode(self, node, parent): if node is None: return if self.offset and getattr(node, 'lineno', None) is not None: node.lineno += self.offset[0] node.col_offset += self.offset[1] if self.traceTree: print(' ' * self.nodeDepth + node.__class__.__name__) if self.futuresAllowed and not (isinstance(node, ast.ImportFrom) or self.isDocstring(node)): self.futuresAllowed = False self.nodeDepth += 1 node.depth = self.nodeDepth node.parent = parent try: handler = self.getNodeHandler(node.__class__) handler(node) finally: self.nodeDepth -= 1 if self.traceTree: print(' ' * self.nodeDepth + 'end ' + node.__class__.__name__) _getDoctestExamples = doctest.DocTestParser().get_examples def handleDoctests(self, node): try: if hasattr(node, 'docstring'): docstring = node.docstring # This is just a reasonable guess. In Python 3.7, docstrings no # longer have line numbers associated with them. This will be # incorrect if there are empty lines between the beginning # of the function and the docstring. node_lineno = node.lineno if hasattr(node, 'args'): node_lineno = max([node_lineno] + [arg.lineno for arg in node.args.args]) else: (docstring, node_lineno) = self.getDocstring(node.body[0]) examples = docstring and self._getDoctestExamples(docstring) except (ValueError, IndexError): # e.g. line 6 of the docstring for <string> has inconsistent # leading whitespace: ... return if not examples: return # Place doctest in module scope saved_stack = self.scopeStack self.scopeStack = [self.scopeStack[0]] node_offset = self.offset or (0, 0) self.pushScope(DoctestScope) self.addBinding(None, Builtin('_')) for example in examples: try: tree = ast.parse(example.source, "<doctest>") except SyntaxError: e = sys.exc_info()[1] if PYPY: e.offset += 1 position = (node_lineno + example.lineno + e.lineno, example.indent + 4 + (e.offset or 0)) self.report(messages.DoctestSyntaxError, node, position) else: self.offset = (node_offset[0] + node_lineno + example.lineno, node_offset[1] + example.indent + 4) self.handleChildren(tree) self.offset = node_offset self.popScope() self.scopeStack = saved_stack def handleStringAnnotation(self, s, node, ref_lineno, ref_col_offset, err): try: tree = ast.parse(s) except SyntaxError: self.report(err, node, s) return body = tree.body if len(body) != 1 or not isinstance(body[0], ast.Expr): self.report(err, node, s) return parsed_annotation = tree.body[0].value for descendant in ast.walk(parsed_annotation): if ( 'lineno' in descendant._attributes and 'col_offset' in descendant._attributes ): descendant.lineno = ref_lineno descendant.col_offset = ref_col_offset self.handleNode(parsed_annotation, node) def handleAnnotation(self, annotation, node): if isinstance(annotation, ast.Str): # Defer handling forward annotation. self.deferFunction(functools.partial( self.handleStringAnnotation, annotation.s, node, annotation.lineno, annotation.col_offset, messages.ForwardAnnotationSyntaxError, )) elif self.annotationsFutureEnabled: self.deferFunction(lambda: self.handleNode(annotation, node)) else: self.handleNode(annotation, node) def ignore(self, node): pass # "stmt" type nodes DELETE = PRINT = FOR = ASYNCFOR = WHILE = IF = WITH = WITHITEM = \ ASYNCWITH = ASYNCWITHITEM = TRYFINALLY = EXEC = \ EXPR = ASSIGN = handleChildren PASS = ignore # "expr" type nodes BOOLOP = BINOP = UNARYOP = IFEXP = SET = \ CALL = REPR = ATTRIBUTE = SUBSCRIPT = \ STARRED = NAMECONSTANT = handleChildren NUM = STR = BYTES = ELLIPSIS = CONSTANT = ignore # "slice" type nodes SLICE = EXTSLICE = INDEX = handleChildren # expression contexts are node instances too, though being constants LOAD = STORE = DEL = AUGLOAD = AUGSTORE = PARAM = ignore # same for operators AND = OR = ADD = SUB = MULT = DIV = MOD = POW = LSHIFT = RSHIFT = \ BITOR = BITXOR = BITAND = FLOORDIV = INVERT = NOT = UADD = USUB = \ EQ = NOTEQ = LT = LTE = GT = GTE = IS = ISNOT = IN = NOTIN = \ MATMULT = ignore def RAISE(self, node): self.handleChildren(node) arg = get_raise_argument(node) if isinstance(arg, ast.Call): if is_notimplemented_name_node(arg.func): # Handle "raise NotImplemented(...)" self.report(messages.RaiseNotImplemented, node) elif is_notimplemented_name_node(arg): # Handle "raise NotImplemented" self.report(messages.RaiseNotImplemented, node) # additional node types COMPREHENSION = KEYWORD = FORMATTEDVALUE = JOINEDSTR = handleChildren def DICT(self, node): # Complain if there are duplicate keys with different values # If they have the same value it's not going to cause potentially # unexpected behaviour so we'll not complain. keys = [ convert_to_value(key) for key in node.keys ] key_counts = counter(keys) duplicate_keys = [ key for key, count in key_counts.items() if count > 1 ] for key in duplicate_keys: key_indices = [i for i, i_key in enumerate(keys) if i_key == key] values = counter( convert_to_value(node.values[index]) for index in key_indices ) if any(count == 1 for value, count in values.items()): for key_index in key_indices: key_node = node.keys[key_index] if isinstance(key, VariableKey): self.report(messages.MultiValueRepeatedKeyVariable, key_node, key.name) else: self.report( messages.MultiValueRepeatedKeyLiteral, key_node, key, ) self.handleChildren(node) def ASSERT(self, node): if isinstance(node.test, ast.Tuple) and node.test.elts != []: self.report(messages.AssertTuple, node) self.handleChildren(node) def GLOBAL(self, node): """ Keep track of globals declarations. """ global_scope_index = 1 if self._in_doctest() else 0 global_scope = self.scopeStack[global_scope_index] # Ignore 'global' statement in global scope. if self.scope is not global_scope: # One 'global' statement can bind multiple (comma-delimited) names. for node_name in node.names: node_value = Assignment(node_name, node) # Remove UndefinedName messages already reported for this name. # TODO: if the global is not used in this scope, it does not # become a globally defined name. See test_unused_global. self.messages = [ m for m in self.messages if not isinstance(m, messages.UndefinedName) or m.message_args[0] != node_name] # Bind name to global scope if it doesn't exist already. global_scope.setdefault(node_name, node_value) # Bind name to non-global scopes, but as already "used". node_value.used = (global_scope, node) for scope in self.scopeStack[global_scope_index + 1:]: scope[node_name] = node_value NONLOCAL = GLOBAL def GENERATOREXP(self, node): self.pushScope(GeneratorScope) self.handleChildren(node) self.popScope() LISTCOMP = handleChildren if PY2 else GENERATOREXP DICTCOMP = SETCOMP = GENERATOREXP def NAME(self, node): """ Handle occurrence of Name (which can be a load/store/delete access.) """ # Locate the name in locals / function / globals scopes. if isinstance(node.ctx, (ast.Load, ast.AugLoad)): self.handleNodeLoad(node) if (node.id == 'locals' and isinstance(self.scope, FunctionScope) and isinstance(node.parent, ast.Call)): # we are doing locals() call in current scope self.scope.usesLocals = True elif isinstance(node.ctx, (ast.Store, ast.AugStore, ast.Param)): self.handleNodeStore(node) elif isinstance(node.ctx, ast.Del): self.handleNodeDelete(node) else: # Unknown context raise RuntimeError("Got impossible expression context: %r" % (node.ctx,)) def CONTINUE(self, node): # Walk the tree up until we see a loop (OK), a function or class # definition (not OK), for 'continue', a finally block (not OK), or # the top module scope (not OK) n = node while hasattr(n, 'parent'): n, n_child = n.parent, n if isinstance(n, LOOP_TYPES): # Doesn't apply unless it's in the loop itself if n_child not in n.orelse: return if isinstance(n, (ast.FunctionDef, ast.ClassDef)): break # Handle Try/TryFinally difference in Python < and >= 3.3 if hasattr(n, 'finalbody') and isinstance(node, ast.Continue): if n_child in n.finalbody: self.report(messages.ContinueInFinally, node) return if isinstance(node, ast.Continue): self.report(messages.ContinueOutsideLoop, node) else: # ast.Break self.report(messages.BreakOutsideLoop, node) BREAK = CONTINUE def RETURN(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.ReturnOutsideFunction, node) return if ( node.value and hasattr(self.scope, 'returnValue') and not self.scope.returnValue ): self.scope.returnValue = node.value self.handleNode(node.value, node) def YIELD(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.YieldOutsideFunction, node) return self.scope.isGenerator = True self.handleNode(node.value, node) AWAIT = YIELDFROM = YIELD def FUNCTIONDEF(self, node): for deco in node.decorator_list: self.handleNode(deco, node) self.LAMBDA(node) self.addBinding(node, FunctionDefinition(node.name, node)) # doctest does not process doctest within a doctest, # or in nested functions. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) ASYNCFUNCTIONDEF = FUNCTIONDEF def LAMBDA(self, node): args = [] annotations = [] if PY2: def addArgs(arglist): for arg in arglist: if isinstance(arg, ast.Tuple): addArgs(arg.elts) else: args.append(arg.id) addArgs(node.args.args) defaults = node.args.defaults else: for arg in node.args.args + node.args.kwonlyargs: args.append(arg.arg) annotations.append(arg.annotation) defaults = node.args.defaults + node.args.kw_defaults # Only for Python3 FunctionDefs is_py3_func = hasattr(node, 'returns') for arg_name in ('vararg', 'kwarg'): wildcard = getattr(node.args, arg_name) if not wildcard: continue args.append(wildcard if PY2 else wildcard.arg) if is_py3_func: if PY2: # Python 2.7 argannotation = arg_name + 'annotation' annotations.append(getattr(node.args, argannotation)) else: # Python >= 3.4 annotations.append(wildcard.annotation) if is_py3_func: annotations.append(node.returns) if len(set(args)) < len(args): for (idx, arg) in enumerate(args): if arg in args[:idx]: self.report(messages.DuplicateArgument, node, arg) for annotation in annotations: self.handleAnnotation(annotation, node) for default in defaults: self.handleNode(default, node) def runFunction(): self.pushScope() self.handleChildren(node, omit='decorator_list') def checkUnusedAssignments(): """ Check to see if any assignments have not been used. """ for name, binding in self.scope.unusedAssignments(): self.report(messages.UnusedVariable, binding.source, name) self.deferAssignment(checkUnusedAssignments) if PY2: def checkReturnWithArgumentInsideGenerator(): """ Check to see if there is any return statement with arguments but the function is a generator. """ if self.scope.isGenerator and self.scope.returnValue: self.report(messages.ReturnWithArgsInsideGenerator, self.scope.returnValue) self.deferAssignment(checkReturnWithArgumentInsideGenerator) self.popScope() self.deferFunction(runFunction) def ARGUMENTS(self, node): self.handleChildren(node, omit=('defaults', 'kw_defaults')) if PY2: scope_node = self.getScopeNode(node) if node.vararg: self.addBinding(node, Argument(node.vararg, scope_node)) if node.kwarg: self.addBinding(node, Argument(node.kwarg, scope_node)) def ARG(self, node): self.addBinding(node, Argument(node.arg, self.getScopeNode(node))) def CLASSDEF(self, node): """ Check names used in a class definition, including its decorators, base classes, and the body of its definition. Additionally, add its name to the current scope. """ for deco in node.decorator_list: self.handleNode(deco, node) for baseNode in node.bases: self.handleNode(baseNode, node) if not PY2: for keywordNode in node.keywords: self.handleNode(keywordNode, node) self.pushScope(ClassScope) # doctest does not process doctest within a doctest # classes within classes are processed. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) for stmt in node.body: self.handleNode(stmt, node) self.popScope() self.addBinding(node, ClassDefinition(node.name, node)) def AUGASSIGN(self, node): self.handleNodeLoad(node.target) self.handleNode(node.value, node) self.handleNode(node.target, node) def TUPLE(self, node): if not PY2 and isinstance(node.ctx, ast.Store): # Python 3 advanced tuple unpacking: a, *b, c = d. # Only one starred expression is allowed, and no more than 1<<8 # assignments are allowed before a stared expression. There is # also a limit of 1<<24 expressions after the starred expression, # which is impossible to test due to memory restrictions, but we # add it here anyway has_starred = False star_loc = -1 for i, n in enumerate(node.elts): if isinstance(n, ast.Starred): if has_starred: self.report(messages.TwoStarredExpressions, node) # The SyntaxError doesn't distinguish two from more # than two. break has_starred = True star_loc = i if star_loc >= 1 << 8 or len(node.elts) - star_loc - 1 >= 1 << 24: self.report(messages.TooManyExpressionsInStarredAssignment, node) self.handleChildren(node) LIST = TUPLE def IMPORT(self, node): for alias in node.names: if '.' in alias.name and not alias.asname: importation = SubmoduleImportation(alias.name, node) else: name = alias.asname or alias.name importation = Importation(name, node, alias.name) self.addBinding(node, importation) def IMPORTFROM(self, node): if node.module == '__future__': if not self.futuresAllowed: self.report(messages.LateFutureImport, node, [n.name for n in node.names]) else: self.futuresAllowed = False module = ('.' * node.level) + (node.module or '') for alias in node.names: name = alias.asname or alias.name if node.module == '__future__': importation = FutureImportation(name, node, self.scope) if alias.name not in __future__.all_feature_names: self.report(messages.FutureFeatureNotDefined, node, alias.name) if alias.name == 'annotations': self.annotationsFutureEnabled = True elif alias.name == '*': # Only Python 2, local import * is a SyntaxWarning if not PY2 and not isinstance(self.scope, ModuleScope): self.report(messages.ImportStarNotPermitted, node, module) continue self.scope.importStarred = True self.report(messages.ImportStarUsed, node, module) importation = StarImportation(module, node) else: importation = ImportationFrom(name, node, module, alias.name) self.addBinding(node, importation) def TRY(self, node): handler_names = [] # List the exception handlers for i, handler in enumerate(node.handlers): if isinstance(handler.type, ast.Tuple): for exc_type in handler.type.elts: handler_names.append(getNodeName(exc_type)) elif handler.type: handler_names.append(getNodeName(handler.type)) if handler.type is None and i < len(node.handlers) - 1: self.report(messages.DefaultExceptNotLast, handler) # Memorize the except handlers and process the body self.exceptHandlers.append(handler_names) for child in node.body: self.handleNode(child, node) self.exceptHandlers.pop() # Process the other nodes: "except:", "else:", "finally:" self.handleChildren(node, omit='body') TRYEXCEPT = TRY def EXCEPTHANDLER(self, node): if PY2 or node.name is None: self.handleChildren(node) return # If the name already exists in the scope, modify state of existing # binding. if node.name in self.scope: self.handleNodeStore(node) # 3.x: the name of the exception, which is not a Name node, but a # simple string, creates a local that is only bound within the scope of # the except: block. As such, temporarily remove the existing binding # to more accurately determine if the name is used in the except: # block. try: prev_definition = self.scope.pop(node.name) except KeyError: prev_definition = None self.handleNodeStore(node) self.handleChildren(node) # See discussion on https://github.com/PyCQA/pyflakes/pull/59 # We're removing the local name since it's being unbound after leaving # the except: block and it's always unbound if the except: block is # never entered. This will cause an "undefined name" error raised if # the checked code tries to use the name afterwards. # # Unless it's been removed already. Then do nothing. try: binding = self.scope.pop(node.name) except KeyError: pass else: if not binding.used: self.report(messages.UnusedVariable, node, node.name) # Restore. if prev_definition: self.scope[node.name] = prev_definition def ANNASSIGN(self, node): if node.value: # Only bind the *targets* if the assignment has a value. # Otherwise it's not really ast.Store and shouldn't silence # UndefinedLocal warnings. self.handleNode(node.target, node) self.handleAnnotation(node.annotation, node) if node.value: # If the assignment has value, handle the *value* now. self.handleNode(node.value, node) def COMPARE(self, node): literals = (ast.Str, ast.Num) if not PY2: literals += (ast.Bytes,) left = node.left for op, right in zip(node.ops, node.comparators): if (isinstance(op, (ast.Is, ast.IsNot)) and (isinstance(left, literals) or isinstance(right, literals))): self.report(messages.IsLiteral, node) left = right self.handleChildren(node)
PyCQA/pyflakes
pyflakes/checker.py
Checker.isDocstring
python
def isDocstring(self, node): return isinstance(node, ast.Str) or (isinstance(node, ast.Expr) and isinstance(node.value, ast.Str))
Determine if the given node is a docstring, as long as it is at the correct place in the node tree.
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/checker.py#L1091-L1097
null
class Checker(object): """ I check the cleanliness and sanity of Python code. @ivar _deferredFunctions: Tracking list used by L{deferFunction}. Elements of the list are two-tuples. The first element is the callable passed to L{deferFunction}. The second element is a copy of the scope stack at the time L{deferFunction} was called. @ivar _deferredAssignments: Similar to C{_deferredFunctions}, but for callables which are deferred assignment checks. """ _ast_node_scope = { ast.Module: ModuleScope, ast.ClassDef: ClassScope, ast.FunctionDef: FunctionScope, ast.Lambda: FunctionScope, ast.ListComp: GeneratorScope, ast.SetComp: GeneratorScope, ast.GeneratorExp: GeneratorScope, ast.DictComp: GeneratorScope, } if PY35_PLUS: _ast_node_scope[ast.AsyncFunctionDef] = FunctionScope, nodeDepth = 0 offset = None traceTree = False builtIns = set(builtin_vars).union(_MAGIC_GLOBALS) _customBuiltIns = os.environ.get('PYFLAKES_BUILTINS') if _customBuiltIns: builtIns.update(_customBuiltIns.split(',')) del _customBuiltIns # TODO: file_tokens= is required to perform checks on type comments, # eventually make this a required positional argument. For now it # is defaulted to `()` for api compatibility. def __init__(self, tree, filename='(none)', builtins=None, withDoctest='PYFLAKES_DOCTEST' in os.environ, file_tokens=()): self._nodeHandlers = {} self._deferredFunctions = [] self._deferredAssignments = [] self.deadScopes = [] self.messages = [] self.filename = filename if builtins: self.builtIns = self.builtIns.union(builtins) self.withDoctest = withDoctest try: self.scopeStack = [Checker._ast_node_scope[type(tree)]()] except KeyError: raise RuntimeError('No scope implemented for the node %r' % tree) self.exceptHandlers = [()] self.root = tree self._type_comments = _collect_type_comments(tree, file_tokens) for builtin in self.builtIns: self.addBinding(None, Builtin(builtin)) self.handleChildren(tree) self.runDeferred(self._deferredFunctions) # Set _deferredFunctions to None so that deferFunction will fail # noisily if called after we've run through the deferred functions. self._deferredFunctions = None self.runDeferred(self._deferredAssignments) # Set _deferredAssignments to None so that deferAssignment will fail # noisily if called after we've run through the deferred assignments. self._deferredAssignments = None del self.scopeStack[1:] self.popScope() self.checkDeadScopes() def deferFunction(self, callable): """ Schedule a function handler to be called just before completion. This is used for handling function bodies, which must be deferred because code later in the file might modify the global scope. When `callable` is called, the scope at the time this is called will be restored, however it will contain any new bindings added to it. """ self._deferredFunctions.append((callable, self.scopeStack[:], self.offset)) def deferAssignment(self, callable): """ Schedule an assignment handler to be called just after deferred function handlers. """ self._deferredAssignments.append((callable, self.scopeStack[:], self.offset)) def runDeferred(self, deferred): """ Run the callables in C{deferred} using their associated scope stack. """ for handler, scope, offset in deferred: self.scopeStack = scope self.offset = offset handler() def _in_doctest(self): return (len(self.scopeStack) >= 2 and isinstance(self.scopeStack[1], DoctestScope)) @property def futuresAllowed(self): if not all(isinstance(scope, ModuleScope) for scope in self.scopeStack): return False return self.scope._futures_allowed @futuresAllowed.setter def futuresAllowed(self, value): assert value is False if isinstance(self.scope, ModuleScope): self.scope._futures_allowed = False @property def annotationsFutureEnabled(self): scope = self.scopeStack[0] if not isinstance(scope, ModuleScope): return False return scope._annotations_future_enabled @annotationsFutureEnabled.setter def annotationsFutureEnabled(self, value): assert value is True assert isinstance(self.scope, ModuleScope) self.scope._annotations_future_enabled = True @property def scope(self): return self.scopeStack[-1] def popScope(self): self.deadScopes.append(self.scopeStack.pop()) def checkDeadScopes(self): """ Look at scopes which have been fully examined and report names in them which were imported but unused. """ for scope in self.deadScopes: # imports in classes are public members if isinstance(scope, ClassScope): continue all_binding = scope.get('__all__') if all_binding and not isinstance(all_binding, ExportBinding): all_binding = None if all_binding: all_names = set(all_binding.names) undefined = all_names.difference(scope) else: all_names = undefined = [] if undefined: if not scope.importStarred and \ os.path.basename(self.filename) != '__init__.py': # Look for possible mistakes in the export list for name in undefined: self.report(messages.UndefinedExport, scope['__all__'].source, name) # mark all import '*' as used by the undefined in __all__ if scope.importStarred: from_list = [] for binding in scope.values(): if isinstance(binding, StarImportation): binding.used = all_binding from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) for name in undefined: self.report(messages.ImportStarUsage, scope['__all__'].source, name, from_list) # Look for imported names that aren't used. for value in scope.values(): if isinstance(value, Importation): used = value.used or value.name in all_names if not used: messg = messages.UnusedImport self.report(messg, value.source, str(value)) for node in value.redefined: if isinstance(self.getParent(node), FOR_TYPES): messg = messages.ImportShadowedByLoopVar elif used: continue else: messg = messages.RedefinedWhileUnused self.report(messg, node, value.name, value.source) def pushScope(self, scopeClass=FunctionScope): self.scopeStack.append(scopeClass()) def report(self, messageClass, *args, **kwargs): self.messages.append(messageClass(self.filename, *args, **kwargs)) def getParent(self, node): # Lookup the first parent which is not Tuple, List or Starred while True: node = node.parent if not hasattr(node, 'elts') and not hasattr(node, 'ctx'): return node def getCommonAncestor(self, lnode, rnode, stop): if stop in (lnode, rnode) or not (hasattr(lnode, 'parent') and hasattr(rnode, 'parent')): return None if lnode is rnode: return lnode if (lnode.depth > rnode.depth): return self.getCommonAncestor(lnode.parent, rnode, stop) if (lnode.depth < rnode.depth): return self.getCommonAncestor(lnode, rnode.parent, stop) return self.getCommonAncestor(lnode.parent, rnode.parent, stop) def descendantOf(self, node, ancestors, stop): for a in ancestors: if self.getCommonAncestor(node, a, stop): return True return False def _getAncestor(self, node, ancestor_type): parent = node while True: if parent is self.root: return None parent = self.getParent(parent) if isinstance(parent, ancestor_type): return parent def getScopeNode(self, node): return self._getAncestor(node, tuple(Checker._ast_node_scope.keys())) def differentForks(self, lnode, rnode): """True, if lnode and rnode are located on different forks of IF/TRY""" ancestor = self.getCommonAncestor(lnode, rnode, self.root) parts = getAlternatives(ancestor) if parts: for items in parts: if self.descendantOf(lnode, items, ancestor) ^ \ self.descendantOf(rnode, items, ancestor): return True return False def addBinding(self, node, value): """ Called when a binding is altered. - `node` is the statement responsible for the change - `value` is the new value, a Binding instance """ # assert value.source in (node, node.parent): for scope in self.scopeStack[::-1]: if value.name in scope: break existing = scope.get(value.name) if (existing and not isinstance(existing, Builtin) and not self.differentForks(node, existing.source)): parent_stmt = self.getParent(value.source) if isinstance(existing, Importation) and isinstance(parent_stmt, FOR_TYPES): self.report(messages.ImportShadowedByLoopVar, node, value.name, existing.source) elif scope is self.scope: if (isinstance(parent_stmt, ast.comprehension) and not isinstance(self.getParent(existing.source), (FOR_TYPES, ast.comprehension))): self.report(messages.RedefinedInListComp, node, value.name, existing.source) elif not existing.used and value.redefines(existing): if value.name != '_' or isinstance(existing, Importation): if not is_typing_overload(existing, self.scopeStack): self.report(messages.RedefinedWhileUnused, node, value.name, existing.source) elif isinstance(existing, Importation) and value.redefines(existing): existing.redefined.append(node) if value.name in self.scope: # then assume the rebound name is used as a global or within a loop value.used = self.scope[value.name].used self.scope[value.name] = value def getNodeHandler(self, node_class): try: return self._nodeHandlers[node_class] except KeyError: nodeType = getNodeType(node_class) self._nodeHandlers[node_class] = handler = getattr(self, nodeType) return handler def handleNodeLoad(self, node): name = getNodeName(node) if not name: return in_generators = None importStarred = None # try enclosing function scopes and global scope for scope in self.scopeStack[-1::-1]: if isinstance(scope, ClassScope): if not PY2 and name == '__class__': return elif in_generators is False: # only generators used in a class scope can access the # names of the class. this is skipped during the first # iteration continue if (name == 'print' and isinstance(scope.get(name, None), Builtin)): parent = self.getParent(node) if (isinstance(parent, ast.BinOp) and isinstance(parent.op, ast.RShift)): self.report(messages.InvalidPrintSyntax, node) try: scope[name].used = (self.scope, node) # if the name of SubImportation is same as # alias of other Importation and the alias # is used, SubImportation also should be marked as used. n = scope[name] if isinstance(n, Importation) and n._has_alias(): try: scope[n.fullName].used = (self.scope, node) except KeyError: pass except KeyError: pass else: return importStarred = importStarred or scope.importStarred if in_generators is not False: in_generators = isinstance(scope, GeneratorScope) if importStarred: from_list = [] for scope in self.scopeStack[-1::-1]: for binding in scope.values(): if isinstance(binding, StarImportation): # mark '*' imports as used for each scope binding.used = (self.scope, node) from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) self.report(messages.ImportStarUsage, node, name, from_list) return if name == '__path__' and os.path.basename(self.filename) == '__init__.py': # the special name __path__ is valid only in packages return if name == '__module__' and isinstance(self.scope, ClassScope): return # protected with a NameError handler? if 'NameError' not in self.exceptHandlers[-1]: self.report(messages.UndefinedName, node, name) def handleNodeStore(self, node): name = getNodeName(node) if not name: return # if the name hasn't already been defined in the current scope if isinstance(self.scope, FunctionScope) and name not in self.scope: # for each function or module scope above us for scope in self.scopeStack[:-1]: if not isinstance(scope, (FunctionScope, ModuleScope)): continue # if the name was defined in that scope, and the name has # been accessed already in the current scope, and hasn't # been declared global used = name in scope and scope[name].used if used and used[0] is self.scope and name not in self.scope.globals: # then it's probably a mistake self.report(messages.UndefinedLocal, scope[name].used[1], name, scope[name].source) break parent_stmt = self.getParent(node) if isinstance(parent_stmt, (FOR_TYPES, ast.comprehension)) or ( parent_stmt != node.parent and not self.isLiteralTupleUnpacking(parent_stmt)): binding = Binding(name, node) elif name == '__all__' and isinstance(self.scope, ModuleScope): binding = ExportBinding(name, node.parent, self.scope) elif isinstance(getattr(node, 'ctx', None), ast.Param): binding = Argument(name, self.getScopeNode(node)) else: binding = Assignment(name, node) self.addBinding(node, binding) def handleNodeDelete(self, node): def on_conditional_branch(): """ Return `True` if node is part of a conditional body. """ current = getattr(node, 'parent', None) while current: if isinstance(current, (ast.If, ast.While, ast.IfExp)): return True current = getattr(current, 'parent', None) return False name = getNodeName(node) if not name: return if on_conditional_branch(): # We cannot predict if this conditional branch is going to # be executed. return if isinstance(self.scope, FunctionScope) and name in self.scope.globals: self.scope.globals.remove(name) else: try: del self.scope[name] except KeyError: self.report(messages.UndefinedName, node, name) def _handle_type_comments(self, node): for (lineno, col_offset), comment in self._type_comments.get(node, ()): comment = comment.split(':', 1)[1].strip() func_match = TYPE_FUNC_RE.match(comment) if func_match: parts = ( func_match.group(1).replace('*', ''), func_match.group(2).strip(), ) else: parts = (comment,) for part in parts: if PY2: part = part.replace('...', 'Ellipsis') self.deferFunction(functools.partial( self.handleStringAnnotation, part, DummyNode(lineno, col_offset), lineno, col_offset, messages.CommentAnnotationSyntaxError, )) def handleChildren(self, tree, omit=None): self._handle_type_comments(tree) for node in iter_child_nodes(tree, omit=omit): self.handleNode(node, tree) def isLiteralTupleUnpacking(self, node): if isinstance(node, ast.Assign): for child in node.targets + [node.value]: if not hasattr(child, 'elts'): return False return True def getDocstring(self, node): if isinstance(node, ast.Expr): node = node.value if not isinstance(node, ast.Str): return (None, None) if PYPY or PY38_PLUS: doctest_lineno = node.lineno - 1 else: # Computed incorrectly if the docstring has backslash doctest_lineno = node.lineno - node.s.count('\n') - 1 return (node.s, doctest_lineno) def handleNode(self, node, parent): if node is None: return if self.offset and getattr(node, 'lineno', None) is not None: node.lineno += self.offset[0] node.col_offset += self.offset[1] if self.traceTree: print(' ' * self.nodeDepth + node.__class__.__name__) if self.futuresAllowed and not (isinstance(node, ast.ImportFrom) or self.isDocstring(node)): self.futuresAllowed = False self.nodeDepth += 1 node.depth = self.nodeDepth node.parent = parent try: handler = self.getNodeHandler(node.__class__) handler(node) finally: self.nodeDepth -= 1 if self.traceTree: print(' ' * self.nodeDepth + 'end ' + node.__class__.__name__) _getDoctestExamples = doctest.DocTestParser().get_examples def handleDoctests(self, node): try: if hasattr(node, 'docstring'): docstring = node.docstring # This is just a reasonable guess. In Python 3.7, docstrings no # longer have line numbers associated with them. This will be # incorrect if there are empty lines between the beginning # of the function and the docstring. node_lineno = node.lineno if hasattr(node, 'args'): node_lineno = max([node_lineno] + [arg.lineno for arg in node.args.args]) else: (docstring, node_lineno) = self.getDocstring(node.body[0]) examples = docstring and self._getDoctestExamples(docstring) except (ValueError, IndexError): # e.g. line 6 of the docstring for <string> has inconsistent # leading whitespace: ... return if not examples: return # Place doctest in module scope saved_stack = self.scopeStack self.scopeStack = [self.scopeStack[0]] node_offset = self.offset or (0, 0) self.pushScope(DoctestScope) self.addBinding(None, Builtin('_')) for example in examples: try: tree = ast.parse(example.source, "<doctest>") except SyntaxError: e = sys.exc_info()[1] if PYPY: e.offset += 1 position = (node_lineno + example.lineno + e.lineno, example.indent + 4 + (e.offset or 0)) self.report(messages.DoctestSyntaxError, node, position) else: self.offset = (node_offset[0] + node_lineno + example.lineno, node_offset[1] + example.indent + 4) self.handleChildren(tree) self.offset = node_offset self.popScope() self.scopeStack = saved_stack def handleStringAnnotation(self, s, node, ref_lineno, ref_col_offset, err): try: tree = ast.parse(s) except SyntaxError: self.report(err, node, s) return body = tree.body if len(body) != 1 or not isinstance(body[0], ast.Expr): self.report(err, node, s) return parsed_annotation = tree.body[0].value for descendant in ast.walk(parsed_annotation): if ( 'lineno' in descendant._attributes and 'col_offset' in descendant._attributes ): descendant.lineno = ref_lineno descendant.col_offset = ref_col_offset self.handleNode(parsed_annotation, node) def handleAnnotation(self, annotation, node): if isinstance(annotation, ast.Str): # Defer handling forward annotation. self.deferFunction(functools.partial( self.handleStringAnnotation, annotation.s, node, annotation.lineno, annotation.col_offset, messages.ForwardAnnotationSyntaxError, )) elif self.annotationsFutureEnabled: self.deferFunction(lambda: self.handleNode(annotation, node)) else: self.handleNode(annotation, node) def ignore(self, node): pass # "stmt" type nodes DELETE = PRINT = FOR = ASYNCFOR = WHILE = IF = WITH = WITHITEM = \ ASYNCWITH = ASYNCWITHITEM = TRYFINALLY = EXEC = \ EXPR = ASSIGN = handleChildren PASS = ignore # "expr" type nodes BOOLOP = BINOP = UNARYOP = IFEXP = SET = \ CALL = REPR = ATTRIBUTE = SUBSCRIPT = \ STARRED = NAMECONSTANT = handleChildren NUM = STR = BYTES = ELLIPSIS = CONSTANT = ignore # "slice" type nodes SLICE = EXTSLICE = INDEX = handleChildren # expression contexts are node instances too, though being constants LOAD = STORE = DEL = AUGLOAD = AUGSTORE = PARAM = ignore # same for operators AND = OR = ADD = SUB = MULT = DIV = MOD = POW = LSHIFT = RSHIFT = \ BITOR = BITXOR = BITAND = FLOORDIV = INVERT = NOT = UADD = USUB = \ EQ = NOTEQ = LT = LTE = GT = GTE = IS = ISNOT = IN = NOTIN = \ MATMULT = ignore def RAISE(self, node): self.handleChildren(node) arg = get_raise_argument(node) if isinstance(arg, ast.Call): if is_notimplemented_name_node(arg.func): # Handle "raise NotImplemented(...)" self.report(messages.RaiseNotImplemented, node) elif is_notimplemented_name_node(arg): # Handle "raise NotImplemented" self.report(messages.RaiseNotImplemented, node) # additional node types COMPREHENSION = KEYWORD = FORMATTEDVALUE = JOINEDSTR = handleChildren def DICT(self, node): # Complain if there are duplicate keys with different values # If they have the same value it's not going to cause potentially # unexpected behaviour so we'll not complain. keys = [ convert_to_value(key) for key in node.keys ] key_counts = counter(keys) duplicate_keys = [ key for key, count in key_counts.items() if count > 1 ] for key in duplicate_keys: key_indices = [i for i, i_key in enumerate(keys) if i_key == key] values = counter( convert_to_value(node.values[index]) for index in key_indices ) if any(count == 1 for value, count in values.items()): for key_index in key_indices: key_node = node.keys[key_index] if isinstance(key, VariableKey): self.report(messages.MultiValueRepeatedKeyVariable, key_node, key.name) else: self.report( messages.MultiValueRepeatedKeyLiteral, key_node, key, ) self.handleChildren(node) def ASSERT(self, node): if isinstance(node.test, ast.Tuple) and node.test.elts != []: self.report(messages.AssertTuple, node) self.handleChildren(node) def GLOBAL(self, node): """ Keep track of globals declarations. """ global_scope_index = 1 if self._in_doctest() else 0 global_scope = self.scopeStack[global_scope_index] # Ignore 'global' statement in global scope. if self.scope is not global_scope: # One 'global' statement can bind multiple (comma-delimited) names. for node_name in node.names: node_value = Assignment(node_name, node) # Remove UndefinedName messages already reported for this name. # TODO: if the global is not used in this scope, it does not # become a globally defined name. See test_unused_global. self.messages = [ m for m in self.messages if not isinstance(m, messages.UndefinedName) or m.message_args[0] != node_name] # Bind name to global scope if it doesn't exist already. global_scope.setdefault(node_name, node_value) # Bind name to non-global scopes, but as already "used". node_value.used = (global_scope, node) for scope in self.scopeStack[global_scope_index + 1:]: scope[node_name] = node_value NONLOCAL = GLOBAL def GENERATOREXP(self, node): self.pushScope(GeneratorScope) self.handleChildren(node) self.popScope() LISTCOMP = handleChildren if PY2 else GENERATOREXP DICTCOMP = SETCOMP = GENERATOREXP def NAME(self, node): """ Handle occurrence of Name (which can be a load/store/delete access.) """ # Locate the name in locals / function / globals scopes. if isinstance(node.ctx, (ast.Load, ast.AugLoad)): self.handleNodeLoad(node) if (node.id == 'locals' and isinstance(self.scope, FunctionScope) and isinstance(node.parent, ast.Call)): # we are doing locals() call in current scope self.scope.usesLocals = True elif isinstance(node.ctx, (ast.Store, ast.AugStore, ast.Param)): self.handleNodeStore(node) elif isinstance(node.ctx, ast.Del): self.handleNodeDelete(node) else: # Unknown context raise RuntimeError("Got impossible expression context: %r" % (node.ctx,)) def CONTINUE(self, node): # Walk the tree up until we see a loop (OK), a function or class # definition (not OK), for 'continue', a finally block (not OK), or # the top module scope (not OK) n = node while hasattr(n, 'parent'): n, n_child = n.parent, n if isinstance(n, LOOP_TYPES): # Doesn't apply unless it's in the loop itself if n_child not in n.orelse: return if isinstance(n, (ast.FunctionDef, ast.ClassDef)): break # Handle Try/TryFinally difference in Python < and >= 3.3 if hasattr(n, 'finalbody') and isinstance(node, ast.Continue): if n_child in n.finalbody: self.report(messages.ContinueInFinally, node) return if isinstance(node, ast.Continue): self.report(messages.ContinueOutsideLoop, node) else: # ast.Break self.report(messages.BreakOutsideLoop, node) BREAK = CONTINUE def RETURN(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.ReturnOutsideFunction, node) return if ( node.value and hasattr(self.scope, 'returnValue') and not self.scope.returnValue ): self.scope.returnValue = node.value self.handleNode(node.value, node) def YIELD(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.YieldOutsideFunction, node) return self.scope.isGenerator = True self.handleNode(node.value, node) AWAIT = YIELDFROM = YIELD def FUNCTIONDEF(self, node): for deco in node.decorator_list: self.handleNode(deco, node) self.LAMBDA(node) self.addBinding(node, FunctionDefinition(node.name, node)) # doctest does not process doctest within a doctest, # or in nested functions. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) ASYNCFUNCTIONDEF = FUNCTIONDEF def LAMBDA(self, node): args = [] annotations = [] if PY2: def addArgs(arglist): for arg in arglist: if isinstance(arg, ast.Tuple): addArgs(arg.elts) else: args.append(arg.id) addArgs(node.args.args) defaults = node.args.defaults else: for arg in node.args.args + node.args.kwonlyargs: args.append(arg.arg) annotations.append(arg.annotation) defaults = node.args.defaults + node.args.kw_defaults # Only for Python3 FunctionDefs is_py3_func = hasattr(node, 'returns') for arg_name in ('vararg', 'kwarg'): wildcard = getattr(node.args, arg_name) if not wildcard: continue args.append(wildcard if PY2 else wildcard.arg) if is_py3_func: if PY2: # Python 2.7 argannotation = arg_name + 'annotation' annotations.append(getattr(node.args, argannotation)) else: # Python >= 3.4 annotations.append(wildcard.annotation) if is_py3_func: annotations.append(node.returns) if len(set(args)) < len(args): for (idx, arg) in enumerate(args): if arg in args[:idx]: self.report(messages.DuplicateArgument, node, arg) for annotation in annotations: self.handleAnnotation(annotation, node) for default in defaults: self.handleNode(default, node) def runFunction(): self.pushScope() self.handleChildren(node, omit='decorator_list') def checkUnusedAssignments(): """ Check to see if any assignments have not been used. """ for name, binding in self.scope.unusedAssignments(): self.report(messages.UnusedVariable, binding.source, name) self.deferAssignment(checkUnusedAssignments) if PY2: def checkReturnWithArgumentInsideGenerator(): """ Check to see if there is any return statement with arguments but the function is a generator. """ if self.scope.isGenerator and self.scope.returnValue: self.report(messages.ReturnWithArgsInsideGenerator, self.scope.returnValue) self.deferAssignment(checkReturnWithArgumentInsideGenerator) self.popScope() self.deferFunction(runFunction) def ARGUMENTS(self, node): self.handleChildren(node, omit=('defaults', 'kw_defaults')) if PY2: scope_node = self.getScopeNode(node) if node.vararg: self.addBinding(node, Argument(node.vararg, scope_node)) if node.kwarg: self.addBinding(node, Argument(node.kwarg, scope_node)) def ARG(self, node): self.addBinding(node, Argument(node.arg, self.getScopeNode(node))) def CLASSDEF(self, node): """ Check names used in a class definition, including its decorators, base classes, and the body of its definition. Additionally, add its name to the current scope. """ for deco in node.decorator_list: self.handleNode(deco, node) for baseNode in node.bases: self.handleNode(baseNode, node) if not PY2: for keywordNode in node.keywords: self.handleNode(keywordNode, node) self.pushScope(ClassScope) # doctest does not process doctest within a doctest # classes within classes are processed. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) for stmt in node.body: self.handleNode(stmt, node) self.popScope() self.addBinding(node, ClassDefinition(node.name, node)) def AUGASSIGN(self, node): self.handleNodeLoad(node.target) self.handleNode(node.value, node) self.handleNode(node.target, node) def TUPLE(self, node): if not PY2 and isinstance(node.ctx, ast.Store): # Python 3 advanced tuple unpacking: a, *b, c = d. # Only one starred expression is allowed, and no more than 1<<8 # assignments are allowed before a stared expression. There is # also a limit of 1<<24 expressions after the starred expression, # which is impossible to test due to memory restrictions, but we # add it here anyway has_starred = False star_loc = -1 for i, n in enumerate(node.elts): if isinstance(n, ast.Starred): if has_starred: self.report(messages.TwoStarredExpressions, node) # The SyntaxError doesn't distinguish two from more # than two. break has_starred = True star_loc = i if star_loc >= 1 << 8 or len(node.elts) - star_loc - 1 >= 1 << 24: self.report(messages.TooManyExpressionsInStarredAssignment, node) self.handleChildren(node) LIST = TUPLE def IMPORT(self, node): for alias in node.names: if '.' in alias.name and not alias.asname: importation = SubmoduleImportation(alias.name, node) else: name = alias.asname or alias.name importation = Importation(name, node, alias.name) self.addBinding(node, importation) def IMPORTFROM(self, node): if node.module == '__future__': if not self.futuresAllowed: self.report(messages.LateFutureImport, node, [n.name for n in node.names]) else: self.futuresAllowed = False module = ('.' * node.level) + (node.module or '') for alias in node.names: name = alias.asname or alias.name if node.module == '__future__': importation = FutureImportation(name, node, self.scope) if alias.name not in __future__.all_feature_names: self.report(messages.FutureFeatureNotDefined, node, alias.name) if alias.name == 'annotations': self.annotationsFutureEnabled = True elif alias.name == '*': # Only Python 2, local import * is a SyntaxWarning if not PY2 and not isinstance(self.scope, ModuleScope): self.report(messages.ImportStarNotPermitted, node, module) continue self.scope.importStarred = True self.report(messages.ImportStarUsed, node, module) importation = StarImportation(module, node) else: importation = ImportationFrom(name, node, module, alias.name) self.addBinding(node, importation) def TRY(self, node): handler_names = [] # List the exception handlers for i, handler in enumerate(node.handlers): if isinstance(handler.type, ast.Tuple): for exc_type in handler.type.elts: handler_names.append(getNodeName(exc_type)) elif handler.type: handler_names.append(getNodeName(handler.type)) if handler.type is None and i < len(node.handlers) - 1: self.report(messages.DefaultExceptNotLast, handler) # Memorize the except handlers and process the body self.exceptHandlers.append(handler_names) for child in node.body: self.handleNode(child, node) self.exceptHandlers.pop() # Process the other nodes: "except:", "else:", "finally:" self.handleChildren(node, omit='body') TRYEXCEPT = TRY def EXCEPTHANDLER(self, node): if PY2 or node.name is None: self.handleChildren(node) return # If the name already exists in the scope, modify state of existing # binding. if node.name in self.scope: self.handleNodeStore(node) # 3.x: the name of the exception, which is not a Name node, but a # simple string, creates a local that is only bound within the scope of # the except: block. As such, temporarily remove the existing binding # to more accurately determine if the name is used in the except: # block. try: prev_definition = self.scope.pop(node.name) except KeyError: prev_definition = None self.handleNodeStore(node) self.handleChildren(node) # See discussion on https://github.com/PyCQA/pyflakes/pull/59 # We're removing the local name since it's being unbound after leaving # the except: block and it's always unbound if the except: block is # never entered. This will cause an "undefined name" error raised if # the checked code tries to use the name afterwards. # # Unless it's been removed already. Then do nothing. try: binding = self.scope.pop(node.name) except KeyError: pass else: if not binding.used: self.report(messages.UnusedVariable, node, node.name) # Restore. if prev_definition: self.scope[node.name] = prev_definition def ANNASSIGN(self, node): if node.value: # Only bind the *targets* if the assignment has a value. # Otherwise it's not really ast.Store and shouldn't silence # UndefinedLocal warnings. self.handleNode(node.target, node) self.handleAnnotation(node.annotation, node) if node.value: # If the assignment has value, handle the *value* now. self.handleNode(node.value, node) def COMPARE(self, node): literals = (ast.Str, ast.Num) if not PY2: literals += (ast.Bytes,) left = node.left for op, right in zip(node.ops, node.comparators): if (isinstance(op, (ast.Is, ast.IsNot)) and (isinstance(left, literals) or isinstance(right, literals))): self.report(messages.IsLiteral, node) left = right self.handleChildren(node)
PyCQA/pyflakes
pyflakes/checker.py
Checker.GLOBAL
python
def GLOBAL(self, node): global_scope_index = 1 if self._in_doctest() else 0 global_scope = self.scopeStack[global_scope_index] # Ignore 'global' statement in global scope. if self.scope is not global_scope: # One 'global' statement can bind multiple (comma-delimited) names. for node_name in node.names: node_value = Assignment(node_name, node) # Remove UndefinedName messages already reported for this name. # TODO: if the global is not used in this scope, it does not # become a globally defined name. See test_unused_global. self.messages = [ m for m in self.messages if not isinstance(m, messages.UndefinedName) or m.message_args[0] != node_name] # Bind name to global scope if it doesn't exist already. global_scope.setdefault(node_name, node_value) # Bind name to non-global scopes, but as already "used". node_value.used = (global_scope, node) for scope in self.scopeStack[global_scope_index + 1:]: scope[node_name] = node_value
Keep track of globals declarations.
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/checker.py#L1309-L1337
[ "def _in_doctest(self):\n return (len(self.scopeStack) >= 2 and\n isinstance(self.scopeStack[1], DoctestScope))\n" ]
class Checker(object): """ I check the cleanliness and sanity of Python code. @ivar _deferredFunctions: Tracking list used by L{deferFunction}. Elements of the list are two-tuples. The first element is the callable passed to L{deferFunction}. The second element is a copy of the scope stack at the time L{deferFunction} was called. @ivar _deferredAssignments: Similar to C{_deferredFunctions}, but for callables which are deferred assignment checks. """ _ast_node_scope = { ast.Module: ModuleScope, ast.ClassDef: ClassScope, ast.FunctionDef: FunctionScope, ast.Lambda: FunctionScope, ast.ListComp: GeneratorScope, ast.SetComp: GeneratorScope, ast.GeneratorExp: GeneratorScope, ast.DictComp: GeneratorScope, } if PY35_PLUS: _ast_node_scope[ast.AsyncFunctionDef] = FunctionScope, nodeDepth = 0 offset = None traceTree = False builtIns = set(builtin_vars).union(_MAGIC_GLOBALS) _customBuiltIns = os.environ.get('PYFLAKES_BUILTINS') if _customBuiltIns: builtIns.update(_customBuiltIns.split(',')) del _customBuiltIns # TODO: file_tokens= is required to perform checks on type comments, # eventually make this a required positional argument. For now it # is defaulted to `()` for api compatibility. def __init__(self, tree, filename='(none)', builtins=None, withDoctest='PYFLAKES_DOCTEST' in os.environ, file_tokens=()): self._nodeHandlers = {} self._deferredFunctions = [] self._deferredAssignments = [] self.deadScopes = [] self.messages = [] self.filename = filename if builtins: self.builtIns = self.builtIns.union(builtins) self.withDoctest = withDoctest try: self.scopeStack = [Checker._ast_node_scope[type(tree)]()] except KeyError: raise RuntimeError('No scope implemented for the node %r' % tree) self.exceptHandlers = [()] self.root = tree self._type_comments = _collect_type_comments(tree, file_tokens) for builtin in self.builtIns: self.addBinding(None, Builtin(builtin)) self.handleChildren(tree) self.runDeferred(self._deferredFunctions) # Set _deferredFunctions to None so that deferFunction will fail # noisily if called after we've run through the deferred functions. self._deferredFunctions = None self.runDeferred(self._deferredAssignments) # Set _deferredAssignments to None so that deferAssignment will fail # noisily if called after we've run through the deferred assignments. self._deferredAssignments = None del self.scopeStack[1:] self.popScope() self.checkDeadScopes() def deferFunction(self, callable): """ Schedule a function handler to be called just before completion. This is used for handling function bodies, which must be deferred because code later in the file might modify the global scope. When `callable` is called, the scope at the time this is called will be restored, however it will contain any new bindings added to it. """ self._deferredFunctions.append((callable, self.scopeStack[:], self.offset)) def deferAssignment(self, callable): """ Schedule an assignment handler to be called just after deferred function handlers. """ self._deferredAssignments.append((callable, self.scopeStack[:], self.offset)) def runDeferred(self, deferred): """ Run the callables in C{deferred} using their associated scope stack. """ for handler, scope, offset in deferred: self.scopeStack = scope self.offset = offset handler() def _in_doctest(self): return (len(self.scopeStack) >= 2 and isinstance(self.scopeStack[1], DoctestScope)) @property def futuresAllowed(self): if not all(isinstance(scope, ModuleScope) for scope in self.scopeStack): return False return self.scope._futures_allowed @futuresAllowed.setter def futuresAllowed(self, value): assert value is False if isinstance(self.scope, ModuleScope): self.scope._futures_allowed = False @property def annotationsFutureEnabled(self): scope = self.scopeStack[0] if not isinstance(scope, ModuleScope): return False return scope._annotations_future_enabled @annotationsFutureEnabled.setter def annotationsFutureEnabled(self, value): assert value is True assert isinstance(self.scope, ModuleScope) self.scope._annotations_future_enabled = True @property def scope(self): return self.scopeStack[-1] def popScope(self): self.deadScopes.append(self.scopeStack.pop()) def checkDeadScopes(self): """ Look at scopes which have been fully examined and report names in them which were imported but unused. """ for scope in self.deadScopes: # imports in classes are public members if isinstance(scope, ClassScope): continue all_binding = scope.get('__all__') if all_binding and not isinstance(all_binding, ExportBinding): all_binding = None if all_binding: all_names = set(all_binding.names) undefined = all_names.difference(scope) else: all_names = undefined = [] if undefined: if not scope.importStarred and \ os.path.basename(self.filename) != '__init__.py': # Look for possible mistakes in the export list for name in undefined: self.report(messages.UndefinedExport, scope['__all__'].source, name) # mark all import '*' as used by the undefined in __all__ if scope.importStarred: from_list = [] for binding in scope.values(): if isinstance(binding, StarImportation): binding.used = all_binding from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) for name in undefined: self.report(messages.ImportStarUsage, scope['__all__'].source, name, from_list) # Look for imported names that aren't used. for value in scope.values(): if isinstance(value, Importation): used = value.used or value.name in all_names if not used: messg = messages.UnusedImport self.report(messg, value.source, str(value)) for node in value.redefined: if isinstance(self.getParent(node), FOR_TYPES): messg = messages.ImportShadowedByLoopVar elif used: continue else: messg = messages.RedefinedWhileUnused self.report(messg, node, value.name, value.source) def pushScope(self, scopeClass=FunctionScope): self.scopeStack.append(scopeClass()) def report(self, messageClass, *args, **kwargs): self.messages.append(messageClass(self.filename, *args, **kwargs)) def getParent(self, node): # Lookup the first parent which is not Tuple, List or Starred while True: node = node.parent if not hasattr(node, 'elts') and not hasattr(node, 'ctx'): return node def getCommonAncestor(self, lnode, rnode, stop): if stop in (lnode, rnode) or not (hasattr(lnode, 'parent') and hasattr(rnode, 'parent')): return None if lnode is rnode: return lnode if (lnode.depth > rnode.depth): return self.getCommonAncestor(lnode.parent, rnode, stop) if (lnode.depth < rnode.depth): return self.getCommonAncestor(lnode, rnode.parent, stop) return self.getCommonAncestor(lnode.parent, rnode.parent, stop) def descendantOf(self, node, ancestors, stop): for a in ancestors: if self.getCommonAncestor(node, a, stop): return True return False def _getAncestor(self, node, ancestor_type): parent = node while True: if parent is self.root: return None parent = self.getParent(parent) if isinstance(parent, ancestor_type): return parent def getScopeNode(self, node): return self._getAncestor(node, tuple(Checker._ast_node_scope.keys())) def differentForks(self, lnode, rnode): """True, if lnode and rnode are located on different forks of IF/TRY""" ancestor = self.getCommonAncestor(lnode, rnode, self.root) parts = getAlternatives(ancestor) if parts: for items in parts: if self.descendantOf(lnode, items, ancestor) ^ \ self.descendantOf(rnode, items, ancestor): return True return False def addBinding(self, node, value): """ Called when a binding is altered. - `node` is the statement responsible for the change - `value` is the new value, a Binding instance """ # assert value.source in (node, node.parent): for scope in self.scopeStack[::-1]: if value.name in scope: break existing = scope.get(value.name) if (existing and not isinstance(existing, Builtin) and not self.differentForks(node, existing.source)): parent_stmt = self.getParent(value.source) if isinstance(existing, Importation) and isinstance(parent_stmt, FOR_TYPES): self.report(messages.ImportShadowedByLoopVar, node, value.name, existing.source) elif scope is self.scope: if (isinstance(parent_stmt, ast.comprehension) and not isinstance(self.getParent(existing.source), (FOR_TYPES, ast.comprehension))): self.report(messages.RedefinedInListComp, node, value.name, existing.source) elif not existing.used and value.redefines(existing): if value.name != '_' or isinstance(existing, Importation): if not is_typing_overload(existing, self.scopeStack): self.report(messages.RedefinedWhileUnused, node, value.name, existing.source) elif isinstance(existing, Importation) and value.redefines(existing): existing.redefined.append(node) if value.name in self.scope: # then assume the rebound name is used as a global or within a loop value.used = self.scope[value.name].used self.scope[value.name] = value def getNodeHandler(self, node_class): try: return self._nodeHandlers[node_class] except KeyError: nodeType = getNodeType(node_class) self._nodeHandlers[node_class] = handler = getattr(self, nodeType) return handler def handleNodeLoad(self, node): name = getNodeName(node) if not name: return in_generators = None importStarred = None # try enclosing function scopes and global scope for scope in self.scopeStack[-1::-1]: if isinstance(scope, ClassScope): if not PY2 and name == '__class__': return elif in_generators is False: # only generators used in a class scope can access the # names of the class. this is skipped during the first # iteration continue if (name == 'print' and isinstance(scope.get(name, None), Builtin)): parent = self.getParent(node) if (isinstance(parent, ast.BinOp) and isinstance(parent.op, ast.RShift)): self.report(messages.InvalidPrintSyntax, node) try: scope[name].used = (self.scope, node) # if the name of SubImportation is same as # alias of other Importation and the alias # is used, SubImportation also should be marked as used. n = scope[name] if isinstance(n, Importation) and n._has_alias(): try: scope[n.fullName].used = (self.scope, node) except KeyError: pass except KeyError: pass else: return importStarred = importStarred or scope.importStarred if in_generators is not False: in_generators = isinstance(scope, GeneratorScope) if importStarred: from_list = [] for scope in self.scopeStack[-1::-1]: for binding in scope.values(): if isinstance(binding, StarImportation): # mark '*' imports as used for each scope binding.used = (self.scope, node) from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) self.report(messages.ImportStarUsage, node, name, from_list) return if name == '__path__' and os.path.basename(self.filename) == '__init__.py': # the special name __path__ is valid only in packages return if name == '__module__' and isinstance(self.scope, ClassScope): return # protected with a NameError handler? if 'NameError' not in self.exceptHandlers[-1]: self.report(messages.UndefinedName, node, name) def handleNodeStore(self, node): name = getNodeName(node) if not name: return # if the name hasn't already been defined in the current scope if isinstance(self.scope, FunctionScope) and name not in self.scope: # for each function or module scope above us for scope in self.scopeStack[:-1]: if not isinstance(scope, (FunctionScope, ModuleScope)): continue # if the name was defined in that scope, and the name has # been accessed already in the current scope, and hasn't # been declared global used = name in scope and scope[name].used if used and used[0] is self.scope and name not in self.scope.globals: # then it's probably a mistake self.report(messages.UndefinedLocal, scope[name].used[1], name, scope[name].source) break parent_stmt = self.getParent(node) if isinstance(parent_stmt, (FOR_TYPES, ast.comprehension)) or ( parent_stmt != node.parent and not self.isLiteralTupleUnpacking(parent_stmt)): binding = Binding(name, node) elif name == '__all__' and isinstance(self.scope, ModuleScope): binding = ExportBinding(name, node.parent, self.scope) elif isinstance(getattr(node, 'ctx', None), ast.Param): binding = Argument(name, self.getScopeNode(node)) else: binding = Assignment(name, node) self.addBinding(node, binding) def handleNodeDelete(self, node): def on_conditional_branch(): """ Return `True` if node is part of a conditional body. """ current = getattr(node, 'parent', None) while current: if isinstance(current, (ast.If, ast.While, ast.IfExp)): return True current = getattr(current, 'parent', None) return False name = getNodeName(node) if not name: return if on_conditional_branch(): # We cannot predict if this conditional branch is going to # be executed. return if isinstance(self.scope, FunctionScope) and name in self.scope.globals: self.scope.globals.remove(name) else: try: del self.scope[name] except KeyError: self.report(messages.UndefinedName, node, name) def _handle_type_comments(self, node): for (lineno, col_offset), comment in self._type_comments.get(node, ()): comment = comment.split(':', 1)[1].strip() func_match = TYPE_FUNC_RE.match(comment) if func_match: parts = ( func_match.group(1).replace('*', ''), func_match.group(2).strip(), ) else: parts = (comment,) for part in parts: if PY2: part = part.replace('...', 'Ellipsis') self.deferFunction(functools.partial( self.handleStringAnnotation, part, DummyNode(lineno, col_offset), lineno, col_offset, messages.CommentAnnotationSyntaxError, )) def handleChildren(self, tree, omit=None): self._handle_type_comments(tree) for node in iter_child_nodes(tree, omit=omit): self.handleNode(node, tree) def isLiteralTupleUnpacking(self, node): if isinstance(node, ast.Assign): for child in node.targets + [node.value]: if not hasattr(child, 'elts'): return False return True def isDocstring(self, node): """ Determine if the given node is a docstring, as long as it is at the correct place in the node tree. """ return isinstance(node, ast.Str) or (isinstance(node, ast.Expr) and isinstance(node.value, ast.Str)) def getDocstring(self, node): if isinstance(node, ast.Expr): node = node.value if not isinstance(node, ast.Str): return (None, None) if PYPY or PY38_PLUS: doctest_lineno = node.lineno - 1 else: # Computed incorrectly if the docstring has backslash doctest_lineno = node.lineno - node.s.count('\n') - 1 return (node.s, doctest_lineno) def handleNode(self, node, parent): if node is None: return if self.offset and getattr(node, 'lineno', None) is not None: node.lineno += self.offset[0] node.col_offset += self.offset[1] if self.traceTree: print(' ' * self.nodeDepth + node.__class__.__name__) if self.futuresAllowed and not (isinstance(node, ast.ImportFrom) or self.isDocstring(node)): self.futuresAllowed = False self.nodeDepth += 1 node.depth = self.nodeDepth node.parent = parent try: handler = self.getNodeHandler(node.__class__) handler(node) finally: self.nodeDepth -= 1 if self.traceTree: print(' ' * self.nodeDepth + 'end ' + node.__class__.__name__) _getDoctestExamples = doctest.DocTestParser().get_examples def handleDoctests(self, node): try: if hasattr(node, 'docstring'): docstring = node.docstring # This is just a reasonable guess. In Python 3.7, docstrings no # longer have line numbers associated with them. This will be # incorrect if there are empty lines between the beginning # of the function and the docstring. node_lineno = node.lineno if hasattr(node, 'args'): node_lineno = max([node_lineno] + [arg.lineno for arg in node.args.args]) else: (docstring, node_lineno) = self.getDocstring(node.body[0]) examples = docstring and self._getDoctestExamples(docstring) except (ValueError, IndexError): # e.g. line 6 of the docstring for <string> has inconsistent # leading whitespace: ... return if not examples: return # Place doctest in module scope saved_stack = self.scopeStack self.scopeStack = [self.scopeStack[0]] node_offset = self.offset or (0, 0) self.pushScope(DoctestScope) self.addBinding(None, Builtin('_')) for example in examples: try: tree = ast.parse(example.source, "<doctest>") except SyntaxError: e = sys.exc_info()[1] if PYPY: e.offset += 1 position = (node_lineno + example.lineno + e.lineno, example.indent + 4 + (e.offset or 0)) self.report(messages.DoctestSyntaxError, node, position) else: self.offset = (node_offset[0] + node_lineno + example.lineno, node_offset[1] + example.indent + 4) self.handleChildren(tree) self.offset = node_offset self.popScope() self.scopeStack = saved_stack def handleStringAnnotation(self, s, node, ref_lineno, ref_col_offset, err): try: tree = ast.parse(s) except SyntaxError: self.report(err, node, s) return body = tree.body if len(body) != 1 or not isinstance(body[0], ast.Expr): self.report(err, node, s) return parsed_annotation = tree.body[0].value for descendant in ast.walk(parsed_annotation): if ( 'lineno' in descendant._attributes and 'col_offset' in descendant._attributes ): descendant.lineno = ref_lineno descendant.col_offset = ref_col_offset self.handleNode(parsed_annotation, node) def handleAnnotation(self, annotation, node): if isinstance(annotation, ast.Str): # Defer handling forward annotation. self.deferFunction(functools.partial( self.handleStringAnnotation, annotation.s, node, annotation.lineno, annotation.col_offset, messages.ForwardAnnotationSyntaxError, )) elif self.annotationsFutureEnabled: self.deferFunction(lambda: self.handleNode(annotation, node)) else: self.handleNode(annotation, node) def ignore(self, node): pass # "stmt" type nodes DELETE = PRINT = FOR = ASYNCFOR = WHILE = IF = WITH = WITHITEM = \ ASYNCWITH = ASYNCWITHITEM = TRYFINALLY = EXEC = \ EXPR = ASSIGN = handleChildren PASS = ignore # "expr" type nodes BOOLOP = BINOP = UNARYOP = IFEXP = SET = \ CALL = REPR = ATTRIBUTE = SUBSCRIPT = \ STARRED = NAMECONSTANT = handleChildren NUM = STR = BYTES = ELLIPSIS = CONSTANT = ignore # "slice" type nodes SLICE = EXTSLICE = INDEX = handleChildren # expression contexts are node instances too, though being constants LOAD = STORE = DEL = AUGLOAD = AUGSTORE = PARAM = ignore # same for operators AND = OR = ADD = SUB = MULT = DIV = MOD = POW = LSHIFT = RSHIFT = \ BITOR = BITXOR = BITAND = FLOORDIV = INVERT = NOT = UADD = USUB = \ EQ = NOTEQ = LT = LTE = GT = GTE = IS = ISNOT = IN = NOTIN = \ MATMULT = ignore def RAISE(self, node): self.handleChildren(node) arg = get_raise_argument(node) if isinstance(arg, ast.Call): if is_notimplemented_name_node(arg.func): # Handle "raise NotImplemented(...)" self.report(messages.RaiseNotImplemented, node) elif is_notimplemented_name_node(arg): # Handle "raise NotImplemented" self.report(messages.RaiseNotImplemented, node) # additional node types COMPREHENSION = KEYWORD = FORMATTEDVALUE = JOINEDSTR = handleChildren def DICT(self, node): # Complain if there are duplicate keys with different values # If they have the same value it's not going to cause potentially # unexpected behaviour so we'll not complain. keys = [ convert_to_value(key) for key in node.keys ] key_counts = counter(keys) duplicate_keys = [ key for key, count in key_counts.items() if count > 1 ] for key in duplicate_keys: key_indices = [i for i, i_key in enumerate(keys) if i_key == key] values = counter( convert_to_value(node.values[index]) for index in key_indices ) if any(count == 1 for value, count in values.items()): for key_index in key_indices: key_node = node.keys[key_index] if isinstance(key, VariableKey): self.report(messages.MultiValueRepeatedKeyVariable, key_node, key.name) else: self.report( messages.MultiValueRepeatedKeyLiteral, key_node, key, ) self.handleChildren(node) def ASSERT(self, node): if isinstance(node.test, ast.Tuple) and node.test.elts != []: self.report(messages.AssertTuple, node) self.handleChildren(node) NONLOCAL = GLOBAL def GENERATOREXP(self, node): self.pushScope(GeneratorScope) self.handleChildren(node) self.popScope() LISTCOMP = handleChildren if PY2 else GENERATOREXP DICTCOMP = SETCOMP = GENERATOREXP def NAME(self, node): """ Handle occurrence of Name (which can be a load/store/delete access.) """ # Locate the name in locals / function / globals scopes. if isinstance(node.ctx, (ast.Load, ast.AugLoad)): self.handleNodeLoad(node) if (node.id == 'locals' and isinstance(self.scope, FunctionScope) and isinstance(node.parent, ast.Call)): # we are doing locals() call in current scope self.scope.usesLocals = True elif isinstance(node.ctx, (ast.Store, ast.AugStore, ast.Param)): self.handleNodeStore(node) elif isinstance(node.ctx, ast.Del): self.handleNodeDelete(node) else: # Unknown context raise RuntimeError("Got impossible expression context: %r" % (node.ctx,)) def CONTINUE(self, node): # Walk the tree up until we see a loop (OK), a function or class # definition (not OK), for 'continue', a finally block (not OK), or # the top module scope (not OK) n = node while hasattr(n, 'parent'): n, n_child = n.parent, n if isinstance(n, LOOP_TYPES): # Doesn't apply unless it's in the loop itself if n_child not in n.orelse: return if isinstance(n, (ast.FunctionDef, ast.ClassDef)): break # Handle Try/TryFinally difference in Python < and >= 3.3 if hasattr(n, 'finalbody') and isinstance(node, ast.Continue): if n_child in n.finalbody: self.report(messages.ContinueInFinally, node) return if isinstance(node, ast.Continue): self.report(messages.ContinueOutsideLoop, node) else: # ast.Break self.report(messages.BreakOutsideLoop, node) BREAK = CONTINUE def RETURN(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.ReturnOutsideFunction, node) return if ( node.value and hasattr(self.scope, 'returnValue') and not self.scope.returnValue ): self.scope.returnValue = node.value self.handleNode(node.value, node) def YIELD(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.YieldOutsideFunction, node) return self.scope.isGenerator = True self.handleNode(node.value, node) AWAIT = YIELDFROM = YIELD def FUNCTIONDEF(self, node): for deco in node.decorator_list: self.handleNode(deco, node) self.LAMBDA(node) self.addBinding(node, FunctionDefinition(node.name, node)) # doctest does not process doctest within a doctest, # or in nested functions. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) ASYNCFUNCTIONDEF = FUNCTIONDEF def LAMBDA(self, node): args = [] annotations = [] if PY2: def addArgs(arglist): for arg in arglist: if isinstance(arg, ast.Tuple): addArgs(arg.elts) else: args.append(arg.id) addArgs(node.args.args) defaults = node.args.defaults else: for arg in node.args.args + node.args.kwonlyargs: args.append(arg.arg) annotations.append(arg.annotation) defaults = node.args.defaults + node.args.kw_defaults # Only for Python3 FunctionDefs is_py3_func = hasattr(node, 'returns') for arg_name in ('vararg', 'kwarg'): wildcard = getattr(node.args, arg_name) if not wildcard: continue args.append(wildcard if PY2 else wildcard.arg) if is_py3_func: if PY2: # Python 2.7 argannotation = arg_name + 'annotation' annotations.append(getattr(node.args, argannotation)) else: # Python >= 3.4 annotations.append(wildcard.annotation) if is_py3_func: annotations.append(node.returns) if len(set(args)) < len(args): for (idx, arg) in enumerate(args): if arg in args[:idx]: self.report(messages.DuplicateArgument, node, arg) for annotation in annotations: self.handleAnnotation(annotation, node) for default in defaults: self.handleNode(default, node) def runFunction(): self.pushScope() self.handleChildren(node, omit='decorator_list') def checkUnusedAssignments(): """ Check to see if any assignments have not been used. """ for name, binding in self.scope.unusedAssignments(): self.report(messages.UnusedVariable, binding.source, name) self.deferAssignment(checkUnusedAssignments) if PY2: def checkReturnWithArgumentInsideGenerator(): """ Check to see if there is any return statement with arguments but the function is a generator. """ if self.scope.isGenerator and self.scope.returnValue: self.report(messages.ReturnWithArgsInsideGenerator, self.scope.returnValue) self.deferAssignment(checkReturnWithArgumentInsideGenerator) self.popScope() self.deferFunction(runFunction) def ARGUMENTS(self, node): self.handleChildren(node, omit=('defaults', 'kw_defaults')) if PY2: scope_node = self.getScopeNode(node) if node.vararg: self.addBinding(node, Argument(node.vararg, scope_node)) if node.kwarg: self.addBinding(node, Argument(node.kwarg, scope_node)) def ARG(self, node): self.addBinding(node, Argument(node.arg, self.getScopeNode(node))) def CLASSDEF(self, node): """ Check names used in a class definition, including its decorators, base classes, and the body of its definition. Additionally, add its name to the current scope. """ for deco in node.decorator_list: self.handleNode(deco, node) for baseNode in node.bases: self.handleNode(baseNode, node) if not PY2: for keywordNode in node.keywords: self.handleNode(keywordNode, node) self.pushScope(ClassScope) # doctest does not process doctest within a doctest # classes within classes are processed. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) for stmt in node.body: self.handleNode(stmt, node) self.popScope() self.addBinding(node, ClassDefinition(node.name, node)) def AUGASSIGN(self, node): self.handleNodeLoad(node.target) self.handleNode(node.value, node) self.handleNode(node.target, node) def TUPLE(self, node): if not PY2 and isinstance(node.ctx, ast.Store): # Python 3 advanced tuple unpacking: a, *b, c = d. # Only one starred expression is allowed, and no more than 1<<8 # assignments are allowed before a stared expression. There is # also a limit of 1<<24 expressions after the starred expression, # which is impossible to test due to memory restrictions, but we # add it here anyway has_starred = False star_loc = -1 for i, n in enumerate(node.elts): if isinstance(n, ast.Starred): if has_starred: self.report(messages.TwoStarredExpressions, node) # The SyntaxError doesn't distinguish two from more # than two. break has_starred = True star_loc = i if star_loc >= 1 << 8 or len(node.elts) - star_loc - 1 >= 1 << 24: self.report(messages.TooManyExpressionsInStarredAssignment, node) self.handleChildren(node) LIST = TUPLE def IMPORT(self, node): for alias in node.names: if '.' in alias.name and not alias.asname: importation = SubmoduleImportation(alias.name, node) else: name = alias.asname or alias.name importation = Importation(name, node, alias.name) self.addBinding(node, importation) def IMPORTFROM(self, node): if node.module == '__future__': if not self.futuresAllowed: self.report(messages.LateFutureImport, node, [n.name for n in node.names]) else: self.futuresAllowed = False module = ('.' * node.level) + (node.module or '') for alias in node.names: name = alias.asname or alias.name if node.module == '__future__': importation = FutureImportation(name, node, self.scope) if alias.name not in __future__.all_feature_names: self.report(messages.FutureFeatureNotDefined, node, alias.name) if alias.name == 'annotations': self.annotationsFutureEnabled = True elif alias.name == '*': # Only Python 2, local import * is a SyntaxWarning if not PY2 and not isinstance(self.scope, ModuleScope): self.report(messages.ImportStarNotPermitted, node, module) continue self.scope.importStarred = True self.report(messages.ImportStarUsed, node, module) importation = StarImportation(module, node) else: importation = ImportationFrom(name, node, module, alias.name) self.addBinding(node, importation) def TRY(self, node): handler_names = [] # List the exception handlers for i, handler in enumerate(node.handlers): if isinstance(handler.type, ast.Tuple): for exc_type in handler.type.elts: handler_names.append(getNodeName(exc_type)) elif handler.type: handler_names.append(getNodeName(handler.type)) if handler.type is None and i < len(node.handlers) - 1: self.report(messages.DefaultExceptNotLast, handler) # Memorize the except handlers and process the body self.exceptHandlers.append(handler_names) for child in node.body: self.handleNode(child, node) self.exceptHandlers.pop() # Process the other nodes: "except:", "else:", "finally:" self.handleChildren(node, omit='body') TRYEXCEPT = TRY def EXCEPTHANDLER(self, node): if PY2 or node.name is None: self.handleChildren(node) return # If the name already exists in the scope, modify state of existing # binding. if node.name in self.scope: self.handleNodeStore(node) # 3.x: the name of the exception, which is not a Name node, but a # simple string, creates a local that is only bound within the scope of # the except: block. As such, temporarily remove the existing binding # to more accurately determine if the name is used in the except: # block. try: prev_definition = self.scope.pop(node.name) except KeyError: prev_definition = None self.handleNodeStore(node) self.handleChildren(node) # See discussion on https://github.com/PyCQA/pyflakes/pull/59 # We're removing the local name since it's being unbound after leaving # the except: block and it's always unbound if the except: block is # never entered. This will cause an "undefined name" error raised if # the checked code tries to use the name afterwards. # # Unless it's been removed already. Then do nothing. try: binding = self.scope.pop(node.name) except KeyError: pass else: if not binding.used: self.report(messages.UnusedVariable, node, node.name) # Restore. if prev_definition: self.scope[node.name] = prev_definition def ANNASSIGN(self, node): if node.value: # Only bind the *targets* if the assignment has a value. # Otherwise it's not really ast.Store and shouldn't silence # UndefinedLocal warnings. self.handleNode(node.target, node) self.handleAnnotation(node.annotation, node) if node.value: # If the assignment has value, handle the *value* now. self.handleNode(node.value, node) def COMPARE(self, node): literals = (ast.Str, ast.Num) if not PY2: literals += (ast.Bytes,) left = node.left for op, right in zip(node.ops, node.comparators): if (isinstance(op, (ast.Is, ast.IsNot)) and (isinstance(left, literals) or isinstance(right, literals))): self.report(messages.IsLiteral, node) left = right self.handleChildren(node)
PyCQA/pyflakes
pyflakes/checker.py
Checker.NAME
python
def NAME(self, node): # Locate the name in locals / function / globals scopes. if isinstance(node.ctx, (ast.Load, ast.AugLoad)): self.handleNodeLoad(node) if (node.id == 'locals' and isinstance(self.scope, FunctionScope) and isinstance(node.parent, ast.Call)): # we are doing locals() call in current scope self.scope.usesLocals = True elif isinstance(node.ctx, (ast.Store, ast.AugStore, ast.Param)): self.handleNodeStore(node) elif isinstance(node.ctx, ast.Del): self.handleNodeDelete(node) else: # Unknown context raise RuntimeError("Got impossible expression context: %r" % (node.ctx,))
Handle occurrence of Name (which can be a load/store/delete access.)
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/checker.py#L1350-L1367
[ "def handleNodeLoad(self, node):\n name = getNodeName(node)\n if not name:\n return\n\n in_generators = None\n importStarred = None\n\n # try enclosing function scopes and global scope\n for scope in self.scopeStack[-1::-1]:\n if isinstance(scope, ClassScope):\n if not PY2 and name == '__class__':\n return\n elif in_generators is False:\n # only generators used in a class scope can access the\n # names of the class. this is skipped during the first\n # iteration\n continue\n\n if (name == 'print' and\n isinstance(scope.get(name, None), Builtin)):\n parent = self.getParent(node)\n if (isinstance(parent, ast.BinOp) and\n isinstance(parent.op, ast.RShift)):\n self.report(messages.InvalidPrintSyntax, node)\n\n try:\n scope[name].used = (self.scope, node)\n\n # if the name of SubImportation is same as\n # alias of other Importation and the alias\n # is used, SubImportation also should be marked as used.\n n = scope[name]\n if isinstance(n, Importation) and n._has_alias():\n try:\n scope[n.fullName].used = (self.scope, node)\n except KeyError:\n pass\n except KeyError:\n pass\n else:\n return\n\n importStarred = importStarred or scope.importStarred\n\n if in_generators is not False:\n in_generators = isinstance(scope, GeneratorScope)\n\n if importStarred:\n from_list = []\n\n for scope in self.scopeStack[-1::-1]:\n for binding in scope.values():\n if isinstance(binding, StarImportation):\n # mark '*' imports as used for each scope\n binding.used = (self.scope, node)\n from_list.append(binding.fullName)\n\n # report * usage, with a list of possible sources\n from_list = ', '.join(sorted(from_list))\n self.report(messages.ImportStarUsage, node, name, from_list)\n return\n\n if name == '__path__' and os.path.basename(self.filename) == '__init__.py':\n # the special name __path__ is valid only in packages\n return\n\n if name == '__module__' and isinstance(self.scope, ClassScope):\n return\n\n # protected with a NameError handler?\n if 'NameError' not in self.exceptHandlers[-1]:\n self.report(messages.UndefinedName, node, name)\n", "def handleNodeStore(self, node):\n name = getNodeName(node)\n if not name:\n return\n # if the name hasn't already been defined in the current scope\n if isinstance(self.scope, FunctionScope) and name not in self.scope:\n # for each function or module scope above us\n for scope in self.scopeStack[:-1]:\n if not isinstance(scope, (FunctionScope, ModuleScope)):\n continue\n # if the name was defined in that scope, and the name has\n # been accessed already in the current scope, and hasn't\n # been declared global\n used = name in scope and scope[name].used\n if used and used[0] is self.scope and name not in self.scope.globals:\n # then it's probably a mistake\n self.report(messages.UndefinedLocal,\n scope[name].used[1], name, scope[name].source)\n break\n\n parent_stmt = self.getParent(node)\n if isinstance(parent_stmt, (FOR_TYPES, ast.comprehension)) or (\n parent_stmt != node.parent and\n not self.isLiteralTupleUnpacking(parent_stmt)):\n binding = Binding(name, node)\n elif name == '__all__' and isinstance(self.scope, ModuleScope):\n binding = ExportBinding(name, node.parent, self.scope)\n elif isinstance(getattr(node, 'ctx', None), ast.Param):\n binding = Argument(name, self.getScopeNode(node))\n else:\n binding = Assignment(name, node)\n self.addBinding(node, binding)\n" ]
class Checker(object): """ I check the cleanliness and sanity of Python code. @ivar _deferredFunctions: Tracking list used by L{deferFunction}. Elements of the list are two-tuples. The first element is the callable passed to L{deferFunction}. The second element is a copy of the scope stack at the time L{deferFunction} was called. @ivar _deferredAssignments: Similar to C{_deferredFunctions}, but for callables which are deferred assignment checks. """ _ast_node_scope = { ast.Module: ModuleScope, ast.ClassDef: ClassScope, ast.FunctionDef: FunctionScope, ast.Lambda: FunctionScope, ast.ListComp: GeneratorScope, ast.SetComp: GeneratorScope, ast.GeneratorExp: GeneratorScope, ast.DictComp: GeneratorScope, } if PY35_PLUS: _ast_node_scope[ast.AsyncFunctionDef] = FunctionScope, nodeDepth = 0 offset = None traceTree = False builtIns = set(builtin_vars).union(_MAGIC_GLOBALS) _customBuiltIns = os.environ.get('PYFLAKES_BUILTINS') if _customBuiltIns: builtIns.update(_customBuiltIns.split(',')) del _customBuiltIns # TODO: file_tokens= is required to perform checks on type comments, # eventually make this a required positional argument. For now it # is defaulted to `()` for api compatibility. def __init__(self, tree, filename='(none)', builtins=None, withDoctest='PYFLAKES_DOCTEST' in os.environ, file_tokens=()): self._nodeHandlers = {} self._deferredFunctions = [] self._deferredAssignments = [] self.deadScopes = [] self.messages = [] self.filename = filename if builtins: self.builtIns = self.builtIns.union(builtins) self.withDoctest = withDoctest try: self.scopeStack = [Checker._ast_node_scope[type(tree)]()] except KeyError: raise RuntimeError('No scope implemented for the node %r' % tree) self.exceptHandlers = [()] self.root = tree self._type_comments = _collect_type_comments(tree, file_tokens) for builtin in self.builtIns: self.addBinding(None, Builtin(builtin)) self.handleChildren(tree) self.runDeferred(self._deferredFunctions) # Set _deferredFunctions to None so that deferFunction will fail # noisily if called after we've run through the deferred functions. self._deferredFunctions = None self.runDeferred(self._deferredAssignments) # Set _deferredAssignments to None so that deferAssignment will fail # noisily if called after we've run through the deferred assignments. self._deferredAssignments = None del self.scopeStack[1:] self.popScope() self.checkDeadScopes() def deferFunction(self, callable): """ Schedule a function handler to be called just before completion. This is used for handling function bodies, which must be deferred because code later in the file might modify the global scope. When `callable` is called, the scope at the time this is called will be restored, however it will contain any new bindings added to it. """ self._deferredFunctions.append((callable, self.scopeStack[:], self.offset)) def deferAssignment(self, callable): """ Schedule an assignment handler to be called just after deferred function handlers. """ self._deferredAssignments.append((callable, self.scopeStack[:], self.offset)) def runDeferred(self, deferred): """ Run the callables in C{deferred} using their associated scope stack. """ for handler, scope, offset in deferred: self.scopeStack = scope self.offset = offset handler() def _in_doctest(self): return (len(self.scopeStack) >= 2 and isinstance(self.scopeStack[1], DoctestScope)) @property def futuresAllowed(self): if not all(isinstance(scope, ModuleScope) for scope in self.scopeStack): return False return self.scope._futures_allowed @futuresAllowed.setter def futuresAllowed(self, value): assert value is False if isinstance(self.scope, ModuleScope): self.scope._futures_allowed = False @property def annotationsFutureEnabled(self): scope = self.scopeStack[0] if not isinstance(scope, ModuleScope): return False return scope._annotations_future_enabled @annotationsFutureEnabled.setter def annotationsFutureEnabled(self, value): assert value is True assert isinstance(self.scope, ModuleScope) self.scope._annotations_future_enabled = True @property def scope(self): return self.scopeStack[-1] def popScope(self): self.deadScopes.append(self.scopeStack.pop()) def checkDeadScopes(self): """ Look at scopes which have been fully examined and report names in them which were imported but unused. """ for scope in self.deadScopes: # imports in classes are public members if isinstance(scope, ClassScope): continue all_binding = scope.get('__all__') if all_binding and not isinstance(all_binding, ExportBinding): all_binding = None if all_binding: all_names = set(all_binding.names) undefined = all_names.difference(scope) else: all_names = undefined = [] if undefined: if not scope.importStarred and \ os.path.basename(self.filename) != '__init__.py': # Look for possible mistakes in the export list for name in undefined: self.report(messages.UndefinedExport, scope['__all__'].source, name) # mark all import '*' as used by the undefined in __all__ if scope.importStarred: from_list = [] for binding in scope.values(): if isinstance(binding, StarImportation): binding.used = all_binding from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) for name in undefined: self.report(messages.ImportStarUsage, scope['__all__'].source, name, from_list) # Look for imported names that aren't used. for value in scope.values(): if isinstance(value, Importation): used = value.used or value.name in all_names if not used: messg = messages.UnusedImport self.report(messg, value.source, str(value)) for node in value.redefined: if isinstance(self.getParent(node), FOR_TYPES): messg = messages.ImportShadowedByLoopVar elif used: continue else: messg = messages.RedefinedWhileUnused self.report(messg, node, value.name, value.source) def pushScope(self, scopeClass=FunctionScope): self.scopeStack.append(scopeClass()) def report(self, messageClass, *args, **kwargs): self.messages.append(messageClass(self.filename, *args, **kwargs)) def getParent(self, node): # Lookup the first parent which is not Tuple, List or Starred while True: node = node.parent if not hasattr(node, 'elts') and not hasattr(node, 'ctx'): return node def getCommonAncestor(self, lnode, rnode, stop): if stop in (lnode, rnode) or not (hasattr(lnode, 'parent') and hasattr(rnode, 'parent')): return None if lnode is rnode: return lnode if (lnode.depth > rnode.depth): return self.getCommonAncestor(lnode.parent, rnode, stop) if (lnode.depth < rnode.depth): return self.getCommonAncestor(lnode, rnode.parent, stop) return self.getCommonAncestor(lnode.parent, rnode.parent, stop) def descendantOf(self, node, ancestors, stop): for a in ancestors: if self.getCommonAncestor(node, a, stop): return True return False def _getAncestor(self, node, ancestor_type): parent = node while True: if parent is self.root: return None parent = self.getParent(parent) if isinstance(parent, ancestor_type): return parent def getScopeNode(self, node): return self._getAncestor(node, tuple(Checker._ast_node_scope.keys())) def differentForks(self, lnode, rnode): """True, if lnode and rnode are located on different forks of IF/TRY""" ancestor = self.getCommonAncestor(lnode, rnode, self.root) parts = getAlternatives(ancestor) if parts: for items in parts: if self.descendantOf(lnode, items, ancestor) ^ \ self.descendantOf(rnode, items, ancestor): return True return False def addBinding(self, node, value): """ Called when a binding is altered. - `node` is the statement responsible for the change - `value` is the new value, a Binding instance """ # assert value.source in (node, node.parent): for scope in self.scopeStack[::-1]: if value.name in scope: break existing = scope.get(value.name) if (existing and not isinstance(existing, Builtin) and not self.differentForks(node, existing.source)): parent_stmt = self.getParent(value.source) if isinstance(existing, Importation) and isinstance(parent_stmt, FOR_TYPES): self.report(messages.ImportShadowedByLoopVar, node, value.name, existing.source) elif scope is self.scope: if (isinstance(parent_stmt, ast.comprehension) and not isinstance(self.getParent(existing.source), (FOR_TYPES, ast.comprehension))): self.report(messages.RedefinedInListComp, node, value.name, existing.source) elif not existing.used and value.redefines(existing): if value.name != '_' or isinstance(existing, Importation): if not is_typing_overload(existing, self.scopeStack): self.report(messages.RedefinedWhileUnused, node, value.name, existing.source) elif isinstance(existing, Importation) and value.redefines(existing): existing.redefined.append(node) if value.name in self.scope: # then assume the rebound name is used as a global or within a loop value.used = self.scope[value.name].used self.scope[value.name] = value def getNodeHandler(self, node_class): try: return self._nodeHandlers[node_class] except KeyError: nodeType = getNodeType(node_class) self._nodeHandlers[node_class] = handler = getattr(self, nodeType) return handler def handleNodeLoad(self, node): name = getNodeName(node) if not name: return in_generators = None importStarred = None # try enclosing function scopes and global scope for scope in self.scopeStack[-1::-1]: if isinstance(scope, ClassScope): if not PY2 and name == '__class__': return elif in_generators is False: # only generators used in a class scope can access the # names of the class. this is skipped during the first # iteration continue if (name == 'print' and isinstance(scope.get(name, None), Builtin)): parent = self.getParent(node) if (isinstance(parent, ast.BinOp) and isinstance(parent.op, ast.RShift)): self.report(messages.InvalidPrintSyntax, node) try: scope[name].used = (self.scope, node) # if the name of SubImportation is same as # alias of other Importation and the alias # is used, SubImportation also should be marked as used. n = scope[name] if isinstance(n, Importation) and n._has_alias(): try: scope[n.fullName].used = (self.scope, node) except KeyError: pass except KeyError: pass else: return importStarred = importStarred or scope.importStarred if in_generators is not False: in_generators = isinstance(scope, GeneratorScope) if importStarred: from_list = [] for scope in self.scopeStack[-1::-1]: for binding in scope.values(): if isinstance(binding, StarImportation): # mark '*' imports as used for each scope binding.used = (self.scope, node) from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) self.report(messages.ImportStarUsage, node, name, from_list) return if name == '__path__' and os.path.basename(self.filename) == '__init__.py': # the special name __path__ is valid only in packages return if name == '__module__' and isinstance(self.scope, ClassScope): return # protected with a NameError handler? if 'NameError' not in self.exceptHandlers[-1]: self.report(messages.UndefinedName, node, name) def handleNodeStore(self, node): name = getNodeName(node) if not name: return # if the name hasn't already been defined in the current scope if isinstance(self.scope, FunctionScope) and name not in self.scope: # for each function or module scope above us for scope in self.scopeStack[:-1]: if not isinstance(scope, (FunctionScope, ModuleScope)): continue # if the name was defined in that scope, and the name has # been accessed already in the current scope, and hasn't # been declared global used = name in scope and scope[name].used if used and used[0] is self.scope and name not in self.scope.globals: # then it's probably a mistake self.report(messages.UndefinedLocal, scope[name].used[1], name, scope[name].source) break parent_stmt = self.getParent(node) if isinstance(parent_stmt, (FOR_TYPES, ast.comprehension)) or ( parent_stmt != node.parent and not self.isLiteralTupleUnpacking(parent_stmt)): binding = Binding(name, node) elif name == '__all__' and isinstance(self.scope, ModuleScope): binding = ExportBinding(name, node.parent, self.scope) elif isinstance(getattr(node, 'ctx', None), ast.Param): binding = Argument(name, self.getScopeNode(node)) else: binding = Assignment(name, node) self.addBinding(node, binding) def handleNodeDelete(self, node): def on_conditional_branch(): """ Return `True` if node is part of a conditional body. """ current = getattr(node, 'parent', None) while current: if isinstance(current, (ast.If, ast.While, ast.IfExp)): return True current = getattr(current, 'parent', None) return False name = getNodeName(node) if not name: return if on_conditional_branch(): # We cannot predict if this conditional branch is going to # be executed. return if isinstance(self.scope, FunctionScope) and name in self.scope.globals: self.scope.globals.remove(name) else: try: del self.scope[name] except KeyError: self.report(messages.UndefinedName, node, name) def _handle_type_comments(self, node): for (lineno, col_offset), comment in self._type_comments.get(node, ()): comment = comment.split(':', 1)[1].strip() func_match = TYPE_FUNC_RE.match(comment) if func_match: parts = ( func_match.group(1).replace('*', ''), func_match.group(2).strip(), ) else: parts = (comment,) for part in parts: if PY2: part = part.replace('...', 'Ellipsis') self.deferFunction(functools.partial( self.handleStringAnnotation, part, DummyNode(lineno, col_offset), lineno, col_offset, messages.CommentAnnotationSyntaxError, )) def handleChildren(self, tree, omit=None): self._handle_type_comments(tree) for node in iter_child_nodes(tree, omit=omit): self.handleNode(node, tree) def isLiteralTupleUnpacking(self, node): if isinstance(node, ast.Assign): for child in node.targets + [node.value]: if not hasattr(child, 'elts'): return False return True def isDocstring(self, node): """ Determine if the given node is a docstring, as long as it is at the correct place in the node tree. """ return isinstance(node, ast.Str) or (isinstance(node, ast.Expr) and isinstance(node.value, ast.Str)) def getDocstring(self, node): if isinstance(node, ast.Expr): node = node.value if not isinstance(node, ast.Str): return (None, None) if PYPY or PY38_PLUS: doctest_lineno = node.lineno - 1 else: # Computed incorrectly if the docstring has backslash doctest_lineno = node.lineno - node.s.count('\n') - 1 return (node.s, doctest_lineno) def handleNode(self, node, parent): if node is None: return if self.offset and getattr(node, 'lineno', None) is not None: node.lineno += self.offset[0] node.col_offset += self.offset[1] if self.traceTree: print(' ' * self.nodeDepth + node.__class__.__name__) if self.futuresAllowed and not (isinstance(node, ast.ImportFrom) or self.isDocstring(node)): self.futuresAllowed = False self.nodeDepth += 1 node.depth = self.nodeDepth node.parent = parent try: handler = self.getNodeHandler(node.__class__) handler(node) finally: self.nodeDepth -= 1 if self.traceTree: print(' ' * self.nodeDepth + 'end ' + node.__class__.__name__) _getDoctestExamples = doctest.DocTestParser().get_examples def handleDoctests(self, node): try: if hasattr(node, 'docstring'): docstring = node.docstring # This is just a reasonable guess. In Python 3.7, docstrings no # longer have line numbers associated with them. This will be # incorrect if there are empty lines between the beginning # of the function and the docstring. node_lineno = node.lineno if hasattr(node, 'args'): node_lineno = max([node_lineno] + [arg.lineno for arg in node.args.args]) else: (docstring, node_lineno) = self.getDocstring(node.body[0]) examples = docstring and self._getDoctestExamples(docstring) except (ValueError, IndexError): # e.g. line 6 of the docstring for <string> has inconsistent # leading whitespace: ... return if not examples: return # Place doctest in module scope saved_stack = self.scopeStack self.scopeStack = [self.scopeStack[0]] node_offset = self.offset or (0, 0) self.pushScope(DoctestScope) self.addBinding(None, Builtin('_')) for example in examples: try: tree = ast.parse(example.source, "<doctest>") except SyntaxError: e = sys.exc_info()[1] if PYPY: e.offset += 1 position = (node_lineno + example.lineno + e.lineno, example.indent + 4 + (e.offset or 0)) self.report(messages.DoctestSyntaxError, node, position) else: self.offset = (node_offset[0] + node_lineno + example.lineno, node_offset[1] + example.indent + 4) self.handleChildren(tree) self.offset = node_offset self.popScope() self.scopeStack = saved_stack def handleStringAnnotation(self, s, node, ref_lineno, ref_col_offset, err): try: tree = ast.parse(s) except SyntaxError: self.report(err, node, s) return body = tree.body if len(body) != 1 or not isinstance(body[0], ast.Expr): self.report(err, node, s) return parsed_annotation = tree.body[0].value for descendant in ast.walk(parsed_annotation): if ( 'lineno' in descendant._attributes and 'col_offset' in descendant._attributes ): descendant.lineno = ref_lineno descendant.col_offset = ref_col_offset self.handleNode(parsed_annotation, node) def handleAnnotation(self, annotation, node): if isinstance(annotation, ast.Str): # Defer handling forward annotation. self.deferFunction(functools.partial( self.handleStringAnnotation, annotation.s, node, annotation.lineno, annotation.col_offset, messages.ForwardAnnotationSyntaxError, )) elif self.annotationsFutureEnabled: self.deferFunction(lambda: self.handleNode(annotation, node)) else: self.handleNode(annotation, node) def ignore(self, node): pass # "stmt" type nodes DELETE = PRINT = FOR = ASYNCFOR = WHILE = IF = WITH = WITHITEM = \ ASYNCWITH = ASYNCWITHITEM = TRYFINALLY = EXEC = \ EXPR = ASSIGN = handleChildren PASS = ignore # "expr" type nodes BOOLOP = BINOP = UNARYOP = IFEXP = SET = \ CALL = REPR = ATTRIBUTE = SUBSCRIPT = \ STARRED = NAMECONSTANT = handleChildren NUM = STR = BYTES = ELLIPSIS = CONSTANT = ignore # "slice" type nodes SLICE = EXTSLICE = INDEX = handleChildren # expression contexts are node instances too, though being constants LOAD = STORE = DEL = AUGLOAD = AUGSTORE = PARAM = ignore # same for operators AND = OR = ADD = SUB = MULT = DIV = MOD = POW = LSHIFT = RSHIFT = \ BITOR = BITXOR = BITAND = FLOORDIV = INVERT = NOT = UADD = USUB = \ EQ = NOTEQ = LT = LTE = GT = GTE = IS = ISNOT = IN = NOTIN = \ MATMULT = ignore def RAISE(self, node): self.handleChildren(node) arg = get_raise_argument(node) if isinstance(arg, ast.Call): if is_notimplemented_name_node(arg.func): # Handle "raise NotImplemented(...)" self.report(messages.RaiseNotImplemented, node) elif is_notimplemented_name_node(arg): # Handle "raise NotImplemented" self.report(messages.RaiseNotImplemented, node) # additional node types COMPREHENSION = KEYWORD = FORMATTEDVALUE = JOINEDSTR = handleChildren def DICT(self, node): # Complain if there are duplicate keys with different values # If they have the same value it's not going to cause potentially # unexpected behaviour so we'll not complain. keys = [ convert_to_value(key) for key in node.keys ] key_counts = counter(keys) duplicate_keys = [ key for key, count in key_counts.items() if count > 1 ] for key in duplicate_keys: key_indices = [i for i, i_key in enumerate(keys) if i_key == key] values = counter( convert_to_value(node.values[index]) for index in key_indices ) if any(count == 1 for value, count in values.items()): for key_index in key_indices: key_node = node.keys[key_index] if isinstance(key, VariableKey): self.report(messages.MultiValueRepeatedKeyVariable, key_node, key.name) else: self.report( messages.MultiValueRepeatedKeyLiteral, key_node, key, ) self.handleChildren(node) def ASSERT(self, node): if isinstance(node.test, ast.Tuple) and node.test.elts != []: self.report(messages.AssertTuple, node) self.handleChildren(node) def GLOBAL(self, node): """ Keep track of globals declarations. """ global_scope_index = 1 if self._in_doctest() else 0 global_scope = self.scopeStack[global_scope_index] # Ignore 'global' statement in global scope. if self.scope is not global_scope: # One 'global' statement can bind multiple (comma-delimited) names. for node_name in node.names: node_value = Assignment(node_name, node) # Remove UndefinedName messages already reported for this name. # TODO: if the global is not used in this scope, it does not # become a globally defined name. See test_unused_global. self.messages = [ m for m in self.messages if not isinstance(m, messages.UndefinedName) or m.message_args[0] != node_name] # Bind name to global scope if it doesn't exist already. global_scope.setdefault(node_name, node_value) # Bind name to non-global scopes, but as already "used". node_value.used = (global_scope, node) for scope in self.scopeStack[global_scope_index + 1:]: scope[node_name] = node_value NONLOCAL = GLOBAL def GENERATOREXP(self, node): self.pushScope(GeneratorScope) self.handleChildren(node) self.popScope() LISTCOMP = handleChildren if PY2 else GENERATOREXP DICTCOMP = SETCOMP = GENERATOREXP def CONTINUE(self, node): # Walk the tree up until we see a loop (OK), a function or class # definition (not OK), for 'continue', a finally block (not OK), or # the top module scope (not OK) n = node while hasattr(n, 'parent'): n, n_child = n.parent, n if isinstance(n, LOOP_TYPES): # Doesn't apply unless it's in the loop itself if n_child not in n.orelse: return if isinstance(n, (ast.FunctionDef, ast.ClassDef)): break # Handle Try/TryFinally difference in Python < and >= 3.3 if hasattr(n, 'finalbody') and isinstance(node, ast.Continue): if n_child in n.finalbody: self.report(messages.ContinueInFinally, node) return if isinstance(node, ast.Continue): self.report(messages.ContinueOutsideLoop, node) else: # ast.Break self.report(messages.BreakOutsideLoop, node) BREAK = CONTINUE def RETURN(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.ReturnOutsideFunction, node) return if ( node.value and hasattr(self.scope, 'returnValue') and not self.scope.returnValue ): self.scope.returnValue = node.value self.handleNode(node.value, node) def YIELD(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.YieldOutsideFunction, node) return self.scope.isGenerator = True self.handleNode(node.value, node) AWAIT = YIELDFROM = YIELD def FUNCTIONDEF(self, node): for deco in node.decorator_list: self.handleNode(deco, node) self.LAMBDA(node) self.addBinding(node, FunctionDefinition(node.name, node)) # doctest does not process doctest within a doctest, # or in nested functions. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) ASYNCFUNCTIONDEF = FUNCTIONDEF def LAMBDA(self, node): args = [] annotations = [] if PY2: def addArgs(arglist): for arg in arglist: if isinstance(arg, ast.Tuple): addArgs(arg.elts) else: args.append(arg.id) addArgs(node.args.args) defaults = node.args.defaults else: for arg in node.args.args + node.args.kwonlyargs: args.append(arg.arg) annotations.append(arg.annotation) defaults = node.args.defaults + node.args.kw_defaults # Only for Python3 FunctionDefs is_py3_func = hasattr(node, 'returns') for arg_name in ('vararg', 'kwarg'): wildcard = getattr(node.args, arg_name) if not wildcard: continue args.append(wildcard if PY2 else wildcard.arg) if is_py3_func: if PY2: # Python 2.7 argannotation = arg_name + 'annotation' annotations.append(getattr(node.args, argannotation)) else: # Python >= 3.4 annotations.append(wildcard.annotation) if is_py3_func: annotations.append(node.returns) if len(set(args)) < len(args): for (idx, arg) in enumerate(args): if arg in args[:idx]: self.report(messages.DuplicateArgument, node, arg) for annotation in annotations: self.handleAnnotation(annotation, node) for default in defaults: self.handleNode(default, node) def runFunction(): self.pushScope() self.handleChildren(node, omit='decorator_list') def checkUnusedAssignments(): """ Check to see if any assignments have not been used. """ for name, binding in self.scope.unusedAssignments(): self.report(messages.UnusedVariable, binding.source, name) self.deferAssignment(checkUnusedAssignments) if PY2: def checkReturnWithArgumentInsideGenerator(): """ Check to see if there is any return statement with arguments but the function is a generator. """ if self.scope.isGenerator and self.scope.returnValue: self.report(messages.ReturnWithArgsInsideGenerator, self.scope.returnValue) self.deferAssignment(checkReturnWithArgumentInsideGenerator) self.popScope() self.deferFunction(runFunction) def ARGUMENTS(self, node): self.handleChildren(node, omit=('defaults', 'kw_defaults')) if PY2: scope_node = self.getScopeNode(node) if node.vararg: self.addBinding(node, Argument(node.vararg, scope_node)) if node.kwarg: self.addBinding(node, Argument(node.kwarg, scope_node)) def ARG(self, node): self.addBinding(node, Argument(node.arg, self.getScopeNode(node))) def CLASSDEF(self, node): """ Check names used in a class definition, including its decorators, base classes, and the body of its definition. Additionally, add its name to the current scope. """ for deco in node.decorator_list: self.handleNode(deco, node) for baseNode in node.bases: self.handleNode(baseNode, node) if not PY2: for keywordNode in node.keywords: self.handleNode(keywordNode, node) self.pushScope(ClassScope) # doctest does not process doctest within a doctest # classes within classes are processed. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) for stmt in node.body: self.handleNode(stmt, node) self.popScope() self.addBinding(node, ClassDefinition(node.name, node)) def AUGASSIGN(self, node): self.handleNodeLoad(node.target) self.handleNode(node.value, node) self.handleNode(node.target, node) def TUPLE(self, node): if not PY2 and isinstance(node.ctx, ast.Store): # Python 3 advanced tuple unpacking: a, *b, c = d. # Only one starred expression is allowed, and no more than 1<<8 # assignments are allowed before a stared expression. There is # also a limit of 1<<24 expressions after the starred expression, # which is impossible to test due to memory restrictions, but we # add it here anyway has_starred = False star_loc = -1 for i, n in enumerate(node.elts): if isinstance(n, ast.Starred): if has_starred: self.report(messages.TwoStarredExpressions, node) # The SyntaxError doesn't distinguish two from more # than two. break has_starred = True star_loc = i if star_loc >= 1 << 8 or len(node.elts) - star_loc - 1 >= 1 << 24: self.report(messages.TooManyExpressionsInStarredAssignment, node) self.handleChildren(node) LIST = TUPLE def IMPORT(self, node): for alias in node.names: if '.' in alias.name and not alias.asname: importation = SubmoduleImportation(alias.name, node) else: name = alias.asname or alias.name importation = Importation(name, node, alias.name) self.addBinding(node, importation) def IMPORTFROM(self, node): if node.module == '__future__': if not self.futuresAllowed: self.report(messages.LateFutureImport, node, [n.name for n in node.names]) else: self.futuresAllowed = False module = ('.' * node.level) + (node.module or '') for alias in node.names: name = alias.asname or alias.name if node.module == '__future__': importation = FutureImportation(name, node, self.scope) if alias.name not in __future__.all_feature_names: self.report(messages.FutureFeatureNotDefined, node, alias.name) if alias.name == 'annotations': self.annotationsFutureEnabled = True elif alias.name == '*': # Only Python 2, local import * is a SyntaxWarning if not PY2 and not isinstance(self.scope, ModuleScope): self.report(messages.ImportStarNotPermitted, node, module) continue self.scope.importStarred = True self.report(messages.ImportStarUsed, node, module) importation = StarImportation(module, node) else: importation = ImportationFrom(name, node, module, alias.name) self.addBinding(node, importation) def TRY(self, node): handler_names = [] # List the exception handlers for i, handler in enumerate(node.handlers): if isinstance(handler.type, ast.Tuple): for exc_type in handler.type.elts: handler_names.append(getNodeName(exc_type)) elif handler.type: handler_names.append(getNodeName(handler.type)) if handler.type is None and i < len(node.handlers) - 1: self.report(messages.DefaultExceptNotLast, handler) # Memorize the except handlers and process the body self.exceptHandlers.append(handler_names) for child in node.body: self.handleNode(child, node) self.exceptHandlers.pop() # Process the other nodes: "except:", "else:", "finally:" self.handleChildren(node, omit='body') TRYEXCEPT = TRY def EXCEPTHANDLER(self, node): if PY2 or node.name is None: self.handleChildren(node) return # If the name already exists in the scope, modify state of existing # binding. if node.name in self.scope: self.handleNodeStore(node) # 3.x: the name of the exception, which is not a Name node, but a # simple string, creates a local that is only bound within the scope of # the except: block. As such, temporarily remove the existing binding # to more accurately determine if the name is used in the except: # block. try: prev_definition = self.scope.pop(node.name) except KeyError: prev_definition = None self.handleNodeStore(node) self.handleChildren(node) # See discussion on https://github.com/PyCQA/pyflakes/pull/59 # We're removing the local name since it's being unbound after leaving # the except: block and it's always unbound if the except: block is # never entered. This will cause an "undefined name" error raised if # the checked code tries to use the name afterwards. # # Unless it's been removed already. Then do nothing. try: binding = self.scope.pop(node.name) except KeyError: pass else: if not binding.used: self.report(messages.UnusedVariable, node, node.name) # Restore. if prev_definition: self.scope[node.name] = prev_definition def ANNASSIGN(self, node): if node.value: # Only bind the *targets* if the assignment has a value. # Otherwise it's not really ast.Store and shouldn't silence # UndefinedLocal warnings. self.handleNode(node.target, node) self.handleAnnotation(node.annotation, node) if node.value: # If the assignment has value, handle the *value* now. self.handleNode(node.value, node) def COMPARE(self, node): literals = (ast.Str, ast.Num) if not PY2: literals += (ast.Bytes,) left = node.left for op, right in zip(node.ops, node.comparators): if (isinstance(op, (ast.Is, ast.IsNot)) and (isinstance(left, literals) or isinstance(right, literals))): self.report(messages.IsLiteral, node) left = right self.handleChildren(node)
PyCQA/pyflakes
pyflakes/checker.py
Checker.CLASSDEF
python
def CLASSDEF(self, node): for deco in node.decorator_list: self.handleNode(deco, node) for baseNode in node.bases: self.handleNode(baseNode, node) if not PY2: for keywordNode in node.keywords: self.handleNode(keywordNode, node) self.pushScope(ClassScope) # doctest does not process doctest within a doctest # classes within classes are processed. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) for stmt in node.body: self.handleNode(stmt, node) self.popScope() self.addBinding(node, ClassDefinition(node.name, node))
Check names used in a class definition, including its decorators, base classes, and the body of its definition. Additionally, add its name to the current scope.
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/checker.py#L1519-L1542
[ "def _in_doctest(self):\n return (len(self.scopeStack) >= 2 and\n isinstance(self.scopeStack[1], DoctestScope))\n", "def popScope(self):\n self.deadScopes.append(self.scopeStack.pop())\n", "def pushScope(self, scopeClass=FunctionScope):\n self.scopeStack.append(scopeClass())\n", "def addBinding(self, node, value):\n \"\"\"\n Called when a binding is altered.\n\n - `node` is the statement responsible for the change\n - `value` is the new value, a Binding instance\n \"\"\"\n # assert value.source in (node, node.parent):\n for scope in self.scopeStack[::-1]:\n if value.name in scope:\n break\n existing = scope.get(value.name)\n\n if (existing and not isinstance(existing, Builtin) and\n not self.differentForks(node, existing.source)):\n\n parent_stmt = self.getParent(value.source)\n if isinstance(existing, Importation) and isinstance(parent_stmt, FOR_TYPES):\n self.report(messages.ImportShadowedByLoopVar,\n node, value.name, existing.source)\n\n elif scope is self.scope:\n if (isinstance(parent_stmt, ast.comprehension) and\n not isinstance(self.getParent(existing.source),\n (FOR_TYPES, ast.comprehension))):\n self.report(messages.RedefinedInListComp,\n node, value.name, existing.source)\n elif not existing.used and value.redefines(existing):\n if value.name != '_' or isinstance(existing, Importation):\n if not is_typing_overload(existing, self.scopeStack):\n self.report(messages.RedefinedWhileUnused,\n node, value.name, existing.source)\n\n elif isinstance(existing, Importation) and value.redefines(existing):\n existing.redefined.append(node)\n\n if value.name in self.scope:\n # then assume the rebound name is used as a global or within a loop\n value.used = self.scope[value.name].used\n\n self.scope[value.name] = value\n", "def handleNode(self, node, parent):\n if node is None:\n return\n if self.offset and getattr(node, 'lineno', None) is not None:\n node.lineno += self.offset[0]\n node.col_offset += self.offset[1]\n if self.traceTree:\n print(' ' * self.nodeDepth + node.__class__.__name__)\n if self.futuresAllowed and not (isinstance(node, ast.ImportFrom) or\n self.isDocstring(node)):\n self.futuresAllowed = False\n self.nodeDepth += 1\n node.depth = self.nodeDepth\n node.parent = parent\n try:\n handler = self.getNodeHandler(node.__class__)\n handler(node)\n finally:\n self.nodeDepth -= 1\n if self.traceTree:\n print(' ' * self.nodeDepth + 'end ' + node.__class__.__name__)\n" ]
class Checker(object): """ I check the cleanliness and sanity of Python code. @ivar _deferredFunctions: Tracking list used by L{deferFunction}. Elements of the list are two-tuples. The first element is the callable passed to L{deferFunction}. The second element is a copy of the scope stack at the time L{deferFunction} was called. @ivar _deferredAssignments: Similar to C{_deferredFunctions}, but for callables which are deferred assignment checks. """ _ast_node_scope = { ast.Module: ModuleScope, ast.ClassDef: ClassScope, ast.FunctionDef: FunctionScope, ast.Lambda: FunctionScope, ast.ListComp: GeneratorScope, ast.SetComp: GeneratorScope, ast.GeneratorExp: GeneratorScope, ast.DictComp: GeneratorScope, } if PY35_PLUS: _ast_node_scope[ast.AsyncFunctionDef] = FunctionScope, nodeDepth = 0 offset = None traceTree = False builtIns = set(builtin_vars).union(_MAGIC_GLOBALS) _customBuiltIns = os.environ.get('PYFLAKES_BUILTINS') if _customBuiltIns: builtIns.update(_customBuiltIns.split(',')) del _customBuiltIns # TODO: file_tokens= is required to perform checks on type comments, # eventually make this a required positional argument. For now it # is defaulted to `()` for api compatibility. def __init__(self, tree, filename='(none)', builtins=None, withDoctest='PYFLAKES_DOCTEST' in os.environ, file_tokens=()): self._nodeHandlers = {} self._deferredFunctions = [] self._deferredAssignments = [] self.deadScopes = [] self.messages = [] self.filename = filename if builtins: self.builtIns = self.builtIns.union(builtins) self.withDoctest = withDoctest try: self.scopeStack = [Checker._ast_node_scope[type(tree)]()] except KeyError: raise RuntimeError('No scope implemented for the node %r' % tree) self.exceptHandlers = [()] self.root = tree self._type_comments = _collect_type_comments(tree, file_tokens) for builtin in self.builtIns: self.addBinding(None, Builtin(builtin)) self.handleChildren(tree) self.runDeferred(self._deferredFunctions) # Set _deferredFunctions to None so that deferFunction will fail # noisily if called after we've run through the deferred functions. self._deferredFunctions = None self.runDeferred(self._deferredAssignments) # Set _deferredAssignments to None so that deferAssignment will fail # noisily if called after we've run through the deferred assignments. self._deferredAssignments = None del self.scopeStack[1:] self.popScope() self.checkDeadScopes() def deferFunction(self, callable): """ Schedule a function handler to be called just before completion. This is used for handling function bodies, which must be deferred because code later in the file might modify the global scope. When `callable` is called, the scope at the time this is called will be restored, however it will contain any new bindings added to it. """ self._deferredFunctions.append((callable, self.scopeStack[:], self.offset)) def deferAssignment(self, callable): """ Schedule an assignment handler to be called just after deferred function handlers. """ self._deferredAssignments.append((callable, self.scopeStack[:], self.offset)) def runDeferred(self, deferred): """ Run the callables in C{deferred} using their associated scope stack. """ for handler, scope, offset in deferred: self.scopeStack = scope self.offset = offset handler() def _in_doctest(self): return (len(self.scopeStack) >= 2 and isinstance(self.scopeStack[1], DoctestScope)) @property def futuresAllowed(self): if not all(isinstance(scope, ModuleScope) for scope in self.scopeStack): return False return self.scope._futures_allowed @futuresAllowed.setter def futuresAllowed(self, value): assert value is False if isinstance(self.scope, ModuleScope): self.scope._futures_allowed = False @property def annotationsFutureEnabled(self): scope = self.scopeStack[0] if not isinstance(scope, ModuleScope): return False return scope._annotations_future_enabled @annotationsFutureEnabled.setter def annotationsFutureEnabled(self, value): assert value is True assert isinstance(self.scope, ModuleScope) self.scope._annotations_future_enabled = True @property def scope(self): return self.scopeStack[-1] def popScope(self): self.deadScopes.append(self.scopeStack.pop()) def checkDeadScopes(self): """ Look at scopes which have been fully examined and report names in them which were imported but unused. """ for scope in self.deadScopes: # imports in classes are public members if isinstance(scope, ClassScope): continue all_binding = scope.get('__all__') if all_binding and not isinstance(all_binding, ExportBinding): all_binding = None if all_binding: all_names = set(all_binding.names) undefined = all_names.difference(scope) else: all_names = undefined = [] if undefined: if not scope.importStarred and \ os.path.basename(self.filename) != '__init__.py': # Look for possible mistakes in the export list for name in undefined: self.report(messages.UndefinedExport, scope['__all__'].source, name) # mark all import '*' as used by the undefined in __all__ if scope.importStarred: from_list = [] for binding in scope.values(): if isinstance(binding, StarImportation): binding.used = all_binding from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) for name in undefined: self.report(messages.ImportStarUsage, scope['__all__'].source, name, from_list) # Look for imported names that aren't used. for value in scope.values(): if isinstance(value, Importation): used = value.used or value.name in all_names if not used: messg = messages.UnusedImport self.report(messg, value.source, str(value)) for node in value.redefined: if isinstance(self.getParent(node), FOR_TYPES): messg = messages.ImportShadowedByLoopVar elif used: continue else: messg = messages.RedefinedWhileUnused self.report(messg, node, value.name, value.source) def pushScope(self, scopeClass=FunctionScope): self.scopeStack.append(scopeClass()) def report(self, messageClass, *args, **kwargs): self.messages.append(messageClass(self.filename, *args, **kwargs)) def getParent(self, node): # Lookup the first parent which is not Tuple, List or Starred while True: node = node.parent if not hasattr(node, 'elts') and not hasattr(node, 'ctx'): return node def getCommonAncestor(self, lnode, rnode, stop): if stop in (lnode, rnode) or not (hasattr(lnode, 'parent') and hasattr(rnode, 'parent')): return None if lnode is rnode: return lnode if (lnode.depth > rnode.depth): return self.getCommonAncestor(lnode.parent, rnode, stop) if (lnode.depth < rnode.depth): return self.getCommonAncestor(lnode, rnode.parent, stop) return self.getCommonAncestor(lnode.parent, rnode.parent, stop) def descendantOf(self, node, ancestors, stop): for a in ancestors: if self.getCommonAncestor(node, a, stop): return True return False def _getAncestor(self, node, ancestor_type): parent = node while True: if parent is self.root: return None parent = self.getParent(parent) if isinstance(parent, ancestor_type): return parent def getScopeNode(self, node): return self._getAncestor(node, tuple(Checker._ast_node_scope.keys())) def differentForks(self, lnode, rnode): """True, if lnode and rnode are located on different forks of IF/TRY""" ancestor = self.getCommonAncestor(lnode, rnode, self.root) parts = getAlternatives(ancestor) if parts: for items in parts: if self.descendantOf(lnode, items, ancestor) ^ \ self.descendantOf(rnode, items, ancestor): return True return False def addBinding(self, node, value): """ Called when a binding is altered. - `node` is the statement responsible for the change - `value` is the new value, a Binding instance """ # assert value.source in (node, node.parent): for scope in self.scopeStack[::-1]: if value.name in scope: break existing = scope.get(value.name) if (existing and not isinstance(existing, Builtin) and not self.differentForks(node, existing.source)): parent_stmt = self.getParent(value.source) if isinstance(existing, Importation) and isinstance(parent_stmt, FOR_TYPES): self.report(messages.ImportShadowedByLoopVar, node, value.name, existing.source) elif scope is self.scope: if (isinstance(parent_stmt, ast.comprehension) and not isinstance(self.getParent(existing.source), (FOR_TYPES, ast.comprehension))): self.report(messages.RedefinedInListComp, node, value.name, existing.source) elif not existing.used and value.redefines(existing): if value.name != '_' or isinstance(existing, Importation): if not is_typing_overload(existing, self.scopeStack): self.report(messages.RedefinedWhileUnused, node, value.name, existing.source) elif isinstance(existing, Importation) and value.redefines(existing): existing.redefined.append(node) if value.name in self.scope: # then assume the rebound name is used as a global or within a loop value.used = self.scope[value.name].used self.scope[value.name] = value def getNodeHandler(self, node_class): try: return self._nodeHandlers[node_class] except KeyError: nodeType = getNodeType(node_class) self._nodeHandlers[node_class] = handler = getattr(self, nodeType) return handler def handleNodeLoad(self, node): name = getNodeName(node) if not name: return in_generators = None importStarred = None # try enclosing function scopes and global scope for scope in self.scopeStack[-1::-1]: if isinstance(scope, ClassScope): if not PY2 and name == '__class__': return elif in_generators is False: # only generators used in a class scope can access the # names of the class. this is skipped during the first # iteration continue if (name == 'print' and isinstance(scope.get(name, None), Builtin)): parent = self.getParent(node) if (isinstance(parent, ast.BinOp) and isinstance(parent.op, ast.RShift)): self.report(messages.InvalidPrintSyntax, node) try: scope[name].used = (self.scope, node) # if the name of SubImportation is same as # alias of other Importation and the alias # is used, SubImportation also should be marked as used. n = scope[name] if isinstance(n, Importation) and n._has_alias(): try: scope[n.fullName].used = (self.scope, node) except KeyError: pass except KeyError: pass else: return importStarred = importStarred or scope.importStarred if in_generators is not False: in_generators = isinstance(scope, GeneratorScope) if importStarred: from_list = [] for scope in self.scopeStack[-1::-1]: for binding in scope.values(): if isinstance(binding, StarImportation): # mark '*' imports as used for each scope binding.used = (self.scope, node) from_list.append(binding.fullName) # report * usage, with a list of possible sources from_list = ', '.join(sorted(from_list)) self.report(messages.ImportStarUsage, node, name, from_list) return if name == '__path__' and os.path.basename(self.filename) == '__init__.py': # the special name __path__ is valid only in packages return if name == '__module__' and isinstance(self.scope, ClassScope): return # protected with a NameError handler? if 'NameError' not in self.exceptHandlers[-1]: self.report(messages.UndefinedName, node, name) def handleNodeStore(self, node): name = getNodeName(node) if not name: return # if the name hasn't already been defined in the current scope if isinstance(self.scope, FunctionScope) and name not in self.scope: # for each function or module scope above us for scope in self.scopeStack[:-1]: if not isinstance(scope, (FunctionScope, ModuleScope)): continue # if the name was defined in that scope, and the name has # been accessed already in the current scope, and hasn't # been declared global used = name in scope and scope[name].used if used and used[0] is self.scope and name not in self.scope.globals: # then it's probably a mistake self.report(messages.UndefinedLocal, scope[name].used[1], name, scope[name].source) break parent_stmt = self.getParent(node) if isinstance(parent_stmt, (FOR_TYPES, ast.comprehension)) or ( parent_stmt != node.parent and not self.isLiteralTupleUnpacking(parent_stmt)): binding = Binding(name, node) elif name == '__all__' and isinstance(self.scope, ModuleScope): binding = ExportBinding(name, node.parent, self.scope) elif isinstance(getattr(node, 'ctx', None), ast.Param): binding = Argument(name, self.getScopeNode(node)) else: binding = Assignment(name, node) self.addBinding(node, binding) def handleNodeDelete(self, node): def on_conditional_branch(): """ Return `True` if node is part of a conditional body. """ current = getattr(node, 'parent', None) while current: if isinstance(current, (ast.If, ast.While, ast.IfExp)): return True current = getattr(current, 'parent', None) return False name = getNodeName(node) if not name: return if on_conditional_branch(): # We cannot predict if this conditional branch is going to # be executed. return if isinstance(self.scope, FunctionScope) and name in self.scope.globals: self.scope.globals.remove(name) else: try: del self.scope[name] except KeyError: self.report(messages.UndefinedName, node, name) def _handle_type_comments(self, node): for (lineno, col_offset), comment in self._type_comments.get(node, ()): comment = comment.split(':', 1)[1].strip() func_match = TYPE_FUNC_RE.match(comment) if func_match: parts = ( func_match.group(1).replace('*', ''), func_match.group(2).strip(), ) else: parts = (comment,) for part in parts: if PY2: part = part.replace('...', 'Ellipsis') self.deferFunction(functools.partial( self.handleStringAnnotation, part, DummyNode(lineno, col_offset), lineno, col_offset, messages.CommentAnnotationSyntaxError, )) def handleChildren(self, tree, omit=None): self._handle_type_comments(tree) for node in iter_child_nodes(tree, omit=omit): self.handleNode(node, tree) def isLiteralTupleUnpacking(self, node): if isinstance(node, ast.Assign): for child in node.targets + [node.value]: if not hasattr(child, 'elts'): return False return True def isDocstring(self, node): """ Determine if the given node is a docstring, as long as it is at the correct place in the node tree. """ return isinstance(node, ast.Str) or (isinstance(node, ast.Expr) and isinstance(node.value, ast.Str)) def getDocstring(self, node): if isinstance(node, ast.Expr): node = node.value if not isinstance(node, ast.Str): return (None, None) if PYPY or PY38_PLUS: doctest_lineno = node.lineno - 1 else: # Computed incorrectly if the docstring has backslash doctest_lineno = node.lineno - node.s.count('\n') - 1 return (node.s, doctest_lineno) def handleNode(self, node, parent): if node is None: return if self.offset and getattr(node, 'lineno', None) is not None: node.lineno += self.offset[0] node.col_offset += self.offset[1] if self.traceTree: print(' ' * self.nodeDepth + node.__class__.__name__) if self.futuresAllowed and not (isinstance(node, ast.ImportFrom) or self.isDocstring(node)): self.futuresAllowed = False self.nodeDepth += 1 node.depth = self.nodeDepth node.parent = parent try: handler = self.getNodeHandler(node.__class__) handler(node) finally: self.nodeDepth -= 1 if self.traceTree: print(' ' * self.nodeDepth + 'end ' + node.__class__.__name__) _getDoctestExamples = doctest.DocTestParser().get_examples def handleDoctests(self, node): try: if hasattr(node, 'docstring'): docstring = node.docstring # This is just a reasonable guess. In Python 3.7, docstrings no # longer have line numbers associated with them. This will be # incorrect if there are empty lines between the beginning # of the function and the docstring. node_lineno = node.lineno if hasattr(node, 'args'): node_lineno = max([node_lineno] + [arg.lineno for arg in node.args.args]) else: (docstring, node_lineno) = self.getDocstring(node.body[0]) examples = docstring and self._getDoctestExamples(docstring) except (ValueError, IndexError): # e.g. line 6 of the docstring for <string> has inconsistent # leading whitespace: ... return if not examples: return # Place doctest in module scope saved_stack = self.scopeStack self.scopeStack = [self.scopeStack[0]] node_offset = self.offset or (0, 0) self.pushScope(DoctestScope) self.addBinding(None, Builtin('_')) for example in examples: try: tree = ast.parse(example.source, "<doctest>") except SyntaxError: e = sys.exc_info()[1] if PYPY: e.offset += 1 position = (node_lineno + example.lineno + e.lineno, example.indent + 4 + (e.offset or 0)) self.report(messages.DoctestSyntaxError, node, position) else: self.offset = (node_offset[0] + node_lineno + example.lineno, node_offset[1] + example.indent + 4) self.handleChildren(tree) self.offset = node_offset self.popScope() self.scopeStack = saved_stack def handleStringAnnotation(self, s, node, ref_lineno, ref_col_offset, err): try: tree = ast.parse(s) except SyntaxError: self.report(err, node, s) return body = tree.body if len(body) != 1 or not isinstance(body[0], ast.Expr): self.report(err, node, s) return parsed_annotation = tree.body[0].value for descendant in ast.walk(parsed_annotation): if ( 'lineno' in descendant._attributes and 'col_offset' in descendant._attributes ): descendant.lineno = ref_lineno descendant.col_offset = ref_col_offset self.handleNode(parsed_annotation, node) def handleAnnotation(self, annotation, node): if isinstance(annotation, ast.Str): # Defer handling forward annotation. self.deferFunction(functools.partial( self.handleStringAnnotation, annotation.s, node, annotation.lineno, annotation.col_offset, messages.ForwardAnnotationSyntaxError, )) elif self.annotationsFutureEnabled: self.deferFunction(lambda: self.handleNode(annotation, node)) else: self.handleNode(annotation, node) def ignore(self, node): pass # "stmt" type nodes DELETE = PRINT = FOR = ASYNCFOR = WHILE = IF = WITH = WITHITEM = \ ASYNCWITH = ASYNCWITHITEM = TRYFINALLY = EXEC = \ EXPR = ASSIGN = handleChildren PASS = ignore # "expr" type nodes BOOLOP = BINOP = UNARYOP = IFEXP = SET = \ CALL = REPR = ATTRIBUTE = SUBSCRIPT = \ STARRED = NAMECONSTANT = handleChildren NUM = STR = BYTES = ELLIPSIS = CONSTANT = ignore # "slice" type nodes SLICE = EXTSLICE = INDEX = handleChildren # expression contexts are node instances too, though being constants LOAD = STORE = DEL = AUGLOAD = AUGSTORE = PARAM = ignore # same for operators AND = OR = ADD = SUB = MULT = DIV = MOD = POW = LSHIFT = RSHIFT = \ BITOR = BITXOR = BITAND = FLOORDIV = INVERT = NOT = UADD = USUB = \ EQ = NOTEQ = LT = LTE = GT = GTE = IS = ISNOT = IN = NOTIN = \ MATMULT = ignore def RAISE(self, node): self.handleChildren(node) arg = get_raise_argument(node) if isinstance(arg, ast.Call): if is_notimplemented_name_node(arg.func): # Handle "raise NotImplemented(...)" self.report(messages.RaiseNotImplemented, node) elif is_notimplemented_name_node(arg): # Handle "raise NotImplemented" self.report(messages.RaiseNotImplemented, node) # additional node types COMPREHENSION = KEYWORD = FORMATTEDVALUE = JOINEDSTR = handleChildren def DICT(self, node): # Complain if there are duplicate keys with different values # If they have the same value it's not going to cause potentially # unexpected behaviour so we'll not complain. keys = [ convert_to_value(key) for key in node.keys ] key_counts = counter(keys) duplicate_keys = [ key for key, count in key_counts.items() if count > 1 ] for key in duplicate_keys: key_indices = [i for i, i_key in enumerate(keys) if i_key == key] values = counter( convert_to_value(node.values[index]) for index in key_indices ) if any(count == 1 for value, count in values.items()): for key_index in key_indices: key_node = node.keys[key_index] if isinstance(key, VariableKey): self.report(messages.MultiValueRepeatedKeyVariable, key_node, key.name) else: self.report( messages.MultiValueRepeatedKeyLiteral, key_node, key, ) self.handleChildren(node) def ASSERT(self, node): if isinstance(node.test, ast.Tuple) and node.test.elts != []: self.report(messages.AssertTuple, node) self.handleChildren(node) def GLOBAL(self, node): """ Keep track of globals declarations. """ global_scope_index = 1 if self._in_doctest() else 0 global_scope = self.scopeStack[global_scope_index] # Ignore 'global' statement in global scope. if self.scope is not global_scope: # One 'global' statement can bind multiple (comma-delimited) names. for node_name in node.names: node_value = Assignment(node_name, node) # Remove UndefinedName messages already reported for this name. # TODO: if the global is not used in this scope, it does not # become a globally defined name. See test_unused_global. self.messages = [ m for m in self.messages if not isinstance(m, messages.UndefinedName) or m.message_args[0] != node_name] # Bind name to global scope if it doesn't exist already. global_scope.setdefault(node_name, node_value) # Bind name to non-global scopes, but as already "used". node_value.used = (global_scope, node) for scope in self.scopeStack[global_scope_index + 1:]: scope[node_name] = node_value NONLOCAL = GLOBAL def GENERATOREXP(self, node): self.pushScope(GeneratorScope) self.handleChildren(node) self.popScope() LISTCOMP = handleChildren if PY2 else GENERATOREXP DICTCOMP = SETCOMP = GENERATOREXP def NAME(self, node): """ Handle occurrence of Name (which can be a load/store/delete access.) """ # Locate the name in locals / function / globals scopes. if isinstance(node.ctx, (ast.Load, ast.AugLoad)): self.handleNodeLoad(node) if (node.id == 'locals' and isinstance(self.scope, FunctionScope) and isinstance(node.parent, ast.Call)): # we are doing locals() call in current scope self.scope.usesLocals = True elif isinstance(node.ctx, (ast.Store, ast.AugStore, ast.Param)): self.handleNodeStore(node) elif isinstance(node.ctx, ast.Del): self.handleNodeDelete(node) else: # Unknown context raise RuntimeError("Got impossible expression context: %r" % (node.ctx,)) def CONTINUE(self, node): # Walk the tree up until we see a loop (OK), a function or class # definition (not OK), for 'continue', a finally block (not OK), or # the top module scope (not OK) n = node while hasattr(n, 'parent'): n, n_child = n.parent, n if isinstance(n, LOOP_TYPES): # Doesn't apply unless it's in the loop itself if n_child not in n.orelse: return if isinstance(n, (ast.FunctionDef, ast.ClassDef)): break # Handle Try/TryFinally difference in Python < and >= 3.3 if hasattr(n, 'finalbody') and isinstance(node, ast.Continue): if n_child in n.finalbody: self.report(messages.ContinueInFinally, node) return if isinstance(node, ast.Continue): self.report(messages.ContinueOutsideLoop, node) else: # ast.Break self.report(messages.BreakOutsideLoop, node) BREAK = CONTINUE def RETURN(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.ReturnOutsideFunction, node) return if ( node.value and hasattr(self.scope, 'returnValue') and not self.scope.returnValue ): self.scope.returnValue = node.value self.handleNode(node.value, node) def YIELD(self, node): if isinstance(self.scope, (ClassScope, ModuleScope)): self.report(messages.YieldOutsideFunction, node) return self.scope.isGenerator = True self.handleNode(node.value, node) AWAIT = YIELDFROM = YIELD def FUNCTIONDEF(self, node): for deco in node.decorator_list: self.handleNode(deco, node) self.LAMBDA(node) self.addBinding(node, FunctionDefinition(node.name, node)) # doctest does not process doctest within a doctest, # or in nested functions. if (self.withDoctest and not self._in_doctest() and not isinstance(self.scope, FunctionScope)): self.deferFunction(lambda: self.handleDoctests(node)) ASYNCFUNCTIONDEF = FUNCTIONDEF def LAMBDA(self, node): args = [] annotations = [] if PY2: def addArgs(arglist): for arg in arglist: if isinstance(arg, ast.Tuple): addArgs(arg.elts) else: args.append(arg.id) addArgs(node.args.args) defaults = node.args.defaults else: for arg in node.args.args + node.args.kwonlyargs: args.append(arg.arg) annotations.append(arg.annotation) defaults = node.args.defaults + node.args.kw_defaults # Only for Python3 FunctionDefs is_py3_func = hasattr(node, 'returns') for arg_name in ('vararg', 'kwarg'): wildcard = getattr(node.args, arg_name) if not wildcard: continue args.append(wildcard if PY2 else wildcard.arg) if is_py3_func: if PY2: # Python 2.7 argannotation = arg_name + 'annotation' annotations.append(getattr(node.args, argannotation)) else: # Python >= 3.4 annotations.append(wildcard.annotation) if is_py3_func: annotations.append(node.returns) if len(set(args)) < len(args): for (idx, arg) in enumerate(args): if arg in args[:idx]: self.report(messages.DuplicateArgument, node, arg) for annotation in annotations: self.handleAnnotation(annotation, node) for default in defaults: self.handleNode(default, node) def runFunction(): self.pushScope() self.handleChildren(node, omit='decorator_list') def checkUnusedAssignments(): """ Check to see if any assignments have not been used. """ for name, binding in self.scope.unusedAssignments(): self.report(messages.UnusedVariable, binding.source, name) self.deferAssignment(checkUnusedAssignments) if PY2: def checkReturnWithArgumentInsideGenerator(): """ Check to see if there is any return statement with arguments but the function is a generator. """ if self.scope.isGenerator and self.scope.returnValue: self.report(messages.ReturnWithArgsInsideGenerator, self.scope.returnValue) self.deferAssignment(checkReturnWithArgumentInsideGenerator) self.popScope() self.deferFunction(runFunction) def ARGUMENTS(self, node): self.handleChildren(node, omit=('defaults', 'kw_defaults')) if PY2: scope_node = self.getScopeNode(node) if node.vararg: self.addBinding(node, Argument(node.vararg, scope_node)) if node.kwarg: self.addBinding(node, Argument(node.kwarg, scope_node)) def ARG(self, node): self.addBinding(node, Argument(node.arg, self.getScopeNode(node))) def AUGASSIGN(self, node): self.handleNodeLoad(node.target) self.handleNode(node.value, node) self.handleNode(node.target, node) def TUPLE(self, node): if not PY2 and isinstance(node.ctx, ast.Store): # Python 3 advanced tuple unpacking: a, *b, c = d. # Only one starred expression is allowed, and no more than 1<<8 # assignments are allowed before a stared expression. There is # also a limit of 1<<24 expressions after the starred expression, # which is impossible to test due to memory restrictions, but we # add it here anyway has_starred = False star_loc = -1 for i, n in enumerate(node.elts): if isinstance(n, ast.Starred): if has_starred: self.report(messages.TwoStarredExpressions, node) # The SyntaxError doesn't distinguish two from more # than two. break has_starred = True star_loc = i if star_loc >= 1 << 8 or len(node.elts) - star_loc - 1 >= 1 << 24: self.report(messages.TooManyExpressionsInStarredAssignment, node) self.handleChildren(node) LIST = TUPLE def IMPORT(self, node): for alias in node.names: if '.' in alias.name and not alias.asname: importation = SubmoduleImportation(alias.name, node) else: name = alias.asname or alias.name importation = Importation(name, node, alias.name) self.addBinding(node, importation) def IMPORTFROM(self, node): if node.module == '__future__': if not self.futuresAllowed: self.report(messages.LateFutureImport, node, [n.name for n in node.names]) else: self.futuresAllowed = False module = ('.' * node.level) + (node.module or '') for alias in node.names: name = alias.asname or alias.name if node.module == '__future__': importation = FutureImportation(name, node, self.scope) if alias.name not in __future__.all_feature_names: self.report(messages.FutureFeatureNotDefined, node, alias.name) if alias.name == 'annotations': self.annotationsFutureEnabled = True elif alias.name == '*': # Only Python 2, local import * is a SyntaxWarning if not PY2 and not isinstance(self.scope, ModuleScope): self.report(messages.ImportStarNotPermitted, node, module) continue self.scope.importStarred = True self.report(messages.ImportStarUsed, node, module) importation = StarImportation(module, node) else: importation = ImportationFrom(name, node, module, alias.name) self.addBinding(node, importation) def TRY(self, node): handler_names = [] # List the exception handlers for i, handler in enumerate(node.handlers): if isinstance(handler.type, ast.Tuple): for exc_type in handler.type.elts: handler_names.append(getNodeName(exc_type)) elif handler.type: handler_names.append(getNodeName(handler.type)) if handler.type is None and i < len(node.handlers) - 1: self.report(messages.DefaultExceptNotLast, handler) # Memorize the except handlers and process the body self.exceptHandlers.append(handler_names) for child in node.body: self.handleNode(child, node) self.exceptHandlers.pop() # Process the other nodes: "except:", "else:", "finally:" self.handleChildren(node, omit='body') TRYEXCEPT = TRY def EXCEPTHANDLER(self, node): if PY2 or node.name is None: self.handleChildren(node) return # If the name already exists in the scope, modify state of existing # binding. if node.name in self.scope: self.handleNodeStore(node) # 3.x: the name of the exception, which is not a Name node, but a # simple string, creates a local that is only bound within the scope of # the except: block. As such, temporarily remove the existing binding # to more accurately determine if the name is used in the except: # block. try: prev_definition = self.scope.pop(node.name) except KeyError: prev_definition = None self.handleNodeStore(node) self.handleChildren(node) # See discussion on https://github.com/PyCQA/pyflakes/pull/59 # We're removing the local name since it's being unbound after leaving # the except: block and it's always unbound if the except: block is # never entered. This will cause an "undefined name" error raised if # the checked code tries to use the name afterwards. # # Unless it's been removed already. Then do nothing. try: binding = self.scope.pop(node.name) except KeyError: pass else: if not binding.used: self.report(messages.UnusedVariable, node, node.name) # Restore. if prev_definition: self.scope[node.name] = prev_definition def ANNASSIGN(self, node): if node.value: # Only bind the *targets* if the assignment has a value. # Otherwise it's not really ast.Store and shouldn't silence # UndefinedLocal warnings. self.handleNode(node.target, node) self.handleAnnotation(node.annotation, node) if node.value: # If the assignment has value, handle the *value* now. self.handleNode(node.value, node) def COMPARE(self, node): literals = (ast.Str, ast.Num) if not PY2: literals += (ast.Bytes,) left = node.left for op, right in zip(node.ops, node.comparators): if (isinstance(op, (ast.Is, ast.IsNot)) and (isinstance(left, literals) or isinstance(right, literals))): self.report(messages.IsLiteral, node) left = right self.handleChildren(node)
PyCQA/pyflakes
pyflakes/api.py
check
python
def check(codeString, filename, reporter=None): if reporter is None: reporter = modReporter._makeDefaultReporter() # First, compile into an AST and handle syntax errors. try: tree = ast.parse(codeString, filename=filename) except SyntaxError: value = sys.exc_info()[1] msg = value.args[0] (lineno, offset, text) = value.lineno, value.offset, value.text if checker.PYPY: if text is None: lines = codeString.splitlines() if len(lines) >= lineno: text = lines[lineno - 1] if sys.version_info >= (3, ) and isinstance(text, bytes): try: text = text.decode('ascii') except UnicodeDecodeError: text = None offset -= 1 # If there's an encoding problem with the file, the text is None. if text is None: # Avoid using msg, since for the only known case, it contains a # bogus message that claims the encoding the file declared was # unknown. reporter.unexpectedError(filename, 'problem decoding source') else: reporter.syntaxError(filename, msg, lineno, offset, text) return 1 except Exception: reporter.unexpectedError(filename, 'problem decoding source') return 1 # Okay, it's syntactically valid. Now check it. file_tokens = checker.make_tokens(codeString) w = checker.Checker(tree, file_tokens=file_tokens, filename=filename) w.messages.sort(key=lambda m: m.lineno) for warning in w.messages: reporter.flake(warning) return len(w.messages)
Check the Python source given by C{codeString} for flakes. @param codeString: The Python source to check. @type codeString: C{str} @param filename: The name of the file the source came from, used to report errors. @type filename: C{str} @param reporter: A L{Reporter} instance, where errors and warnings will be reported. @return: The number of warnings emitted. @rtype: C{int}
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/api.py#L20-L78
[ "def _makeDefaultReporter():\n \"\"\"\n Make a reporter that can be used when no reporter is specified.\n \"\"\"\n return Reporter(sys.stdout, sys.stderr)\n", "def make_tokens(code):\n # PY3: tokenize.tokenize requires readline of bytes\n if not isinstance(code, bytes):\n code = code.encode('UTF-8')\n lines = iter(code.splitlines(True))\n # next(lines, b'') is to prevent an error in pypy3\n return tuple(tokenize_tokenize(lambda: next(lines, b'')))\n", "def unexpectedError(self, filename, msg):\n \"\"\"\n An unexpected error occurred trying to process C{filename}.\n\n @param filename: The path to a file that we could not process.\n @ptype filename: C{unicode}\n @param msg: A message explaining the problem.\n @ptype msg: C{unicode}\n \"\"\"\n self._stderr.write(\"%s: %s\\n\" % (filename, msg))\n", "def syntaxError(self, filename, msg, lineno, offset, text):\n \"\"\"\n There was a syntax error in C{filename}.\n\n @param filename: The path to the file with the syntax error.\n @ptype filename: C{unicode}\n @param msg: An explanation of the syntax error.\n @ptype msg: C{unicode}\n @param lineno: The line number where the syntax error occurred.\n @ptype lineno: C{int}\n @param offset: The column on which the syntax error occurred, or None.\n @ptype offset: C{int}\n @param text: The source code containing the syntax error.\n @ptype text: C{unicode}\n \"\"\"\n line = text.splitlines()[-1]\n if offset is not None:\n if sys.version_info < (3, 8):\n offset = offset - (len(text) - len(line)) + 1\n self._stderr.write('%s:%d:%d: %s\\n' %\n (filename, lineno, offset, msg))\n else:\n self._stderr.write('%s:%d: %s\\n' % (filename, lineno, msg))\n self._stderr.write(line)\n self._stderr.write('\\n')\n if offset is not None:\n self._stderr.write(re.sub(r'\\S', ' ', line[:offset - 1]) +\n \"^\\n\")\n", "def flake(self, message):\n \"\"\"\n pyflakes found something wrong with the code.\n\n @param: A L{pyflakes.messages.Message}.\n \"\"\"\n self._stdout.write(str(message))\n self._stdout.write('\\n')\n" ]
""" API for the command-line I{pyflakes} tool. """ from __future__ import with_statement import ast import os import platform import re import sys from pyflakes import checker, __version__ from pyflakes import reporter as modReporter __all__ = ['check', 'checkPath', 'checkRecursive', 'iterSourceCode', 'main'] PYTHON_SHEBANG_REGEX = re.compile(br'^#!.*\bpython[23w]?\b\s*$') def checkPath(filename, reporter=None): """ Check the given path, printing out any warnings detected. @param reporter: A L{Reporter} instance, where errors and warnings will be reported. @return: the number of warnings printed """ if reporter is None: reporter = modReporter._makeDefaultReporter() try: with open(filename, 'rb') as f: codestr = f.read() except IOError: msg = sys.exc_info()[1] reporter.unexpectedError(filename, msg.args[1]) return 1 return check(codestr, filename, reporter) def isPythonFile(filename): """Return True if filename points to a Python file.""" if filename.endswith('.py'): return True # Avoid obvious Emacs backup files if filename.endswith("~"): return False max_bytes = 128 try: with open(filename, 'rb') as f: text = f.read(max_bytes) if not text: return False except IOError: return False first_line = text.splitlines()[0] return PYTHON_SHEBANG_REGEX.match(first_line) def iterSourceCode(paths): """ Iterate over all Python source files in C{paths}. @param paths: A list of paths. Directories will be recursed into and any .py files found will be yielded. Any non-directories will be yielded as-is. """ for path in paths: if os.path.isdir(path): for dirpath, dirnames, filenames in os.walk(path): for filename in filenames: full_path = os.path.join(dirpath, filename) if isPythonFile(full_path): yield full_path else: yield path def checkRecursive(paths, reporter): """ Recursively check all source files in C{paths}. @param paths: A list of paths to Python source files and directories containing Python source files. @param reporter: A L{Reporter} where all of the warnings and errors will be reported to. @return: The number of warnings found. """ warnings = 0 for sourcePath in iterSourceCode(paths): warnings += checkPath(sourcePath, reporter) return warnings def _exitOnSignal(sigName, message): """Handles a signal with sys.exit. Some of these signals (SIGPIPE, for example) don't exist or are invalid on Windows. So, ignore errors that might arise. """ import signal try: sigNumber = getattr(signal, sigName) except AttributeError: # the signal constants defined in the signal module are defined by # whether the C library supports them or not. So, SIGPIPE might not # even be defined. return def handler(sig, f): sys.exit(message) try: signal.signal(sigNumber, handler) except ValueError: # It's also possible the signal is defined, but then it's invalid. In # this case, signal.signal raises ValueError. pass def _get_version(): """ Retrieve and format package version along with python version & OS used """ return ('%s Python %s on %s' % (__version__, platform.python_version(), platform.system())) def main(prog=None, args=None): """Entry point for the script "pyflakes".""" import optparse # Handle "Keyboard Interrupt" and "Broken pipe" gracefully _exitOnSignal('SIGINT', '... stopped') _exitOnSignal('SIGPIPE', 1) parser = optparse.OptionParser(prog=prog, version=_get_version()) (__, args) = parser.parse_args(args=args) reporter = modReporter._makeDefaultReporter() if args: warnings = checkRecursive(args, reporter) else: warnings = check(sys.stdin.read(), '<stdin>', reporter) raise SystemExit(warnings > 0)
PyCQA/pyflakes
pyflakes/api.py
checkPath
python
def checkPath(filename, reporter=None): if reporter is None: reporter = modReporter._makeDefaultReporter() try: with open(filename, 'rb') as f: codestr = f.read() except IOError: msg = sys.exc_info()[1] reporter.unexpectedError(filename, msg.args[1]) return 1 return check(codestr, filename, reporter)
Check the given path, printing out any warnings detected. @param reporter: A L{Reporter} instance, where errors and warnings will be reported. @return: the number of warnings printed
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/api.py#L81-L99
[ "def check(codeString, filename, reporter=None):\n \"\"\"\n Check the Python source given by C{codeString} for flakes.\n\n @param codeString: The Python source to check.\n @type codeString: C{str}\n\n @param filename: The name of the file the source came from, used to report\n errors.\n @type filename: C{str}\n\n @param reporter: A L{Reporter} instance, where errors and warnings will be\n reported.\n\n @return: The number of warnings emitted.\n @rtype: C{int}\n \"\"\"\n if reporter is None:\n reporter = modReporter._makeDefaultReporter()\n # First, compile into an AST and handle syntax errors.\n try:\n tree = ast.parse(codeString, filename=filename)\n except SyntaxError:\n value = sys.exc_info()[1]\n msg = value.args[0]\n\n (lineno, offset, text) = value.lineno, value.offset, value.text\n\n if checker.PYPY:\n if text is None:\n lines = codeString.splitlines()\n if len(lines) >= lineno:\n text = lines[lineno - 1]\n if sys.version_info >= (3, ) and isinstance(text, bytes):\n try:\n text = text.decode('ascii')\n except UnicodeDecodeError:\n text = None\n offset -= 1\n\n # If there's an encoding problem with the file, the text is None.\n if text is None:\n # Avoid using msg, since for the only known case, it contains a\n # bogus message that claims the encoding the file declared was\n # unknown.\n reporter.unexpectedError(filename, 'problem decoding source')\n else:\n reporter.syntaxError(filename, msg, lineno, offset, text)\n return 1\n except Exception:\n reporter.unexpectedError(filename, 'problem decoding source')\n return 1\n # Okay, it's syntactically valid. Now check it.\n file_tokens = checker.make_tokens(codeString)\n w = checker.Checker(tree, file_tokens=file_tokens, filename=filename)\n w.messages.sort(key=lambda m: m.lineno)\n for warning in w.messages:\n reporter.flake(warning)\n return len(w.messages)\n", "def _makeDefaultReporter():\n \"\"\"\n Make a reporter that can be used when no reporter is specified.\n \"\"\"\n return Reporter(sys.stdout, sys.stderr)\n", "def unexpectedError(self, filename, msg):\n \"\"\"\n An unexpected error occurred trying to process C{filename}.\n\n @param filename: The path to a file that we could not process.\n @ptype filename: C{unicode}\n @param msg: A message explaining the problem.\n @ptype msg: C{unicode}\n \"\"\"\n self._stderr.write(\"%s: %s\\n\" % (filename, msg))\n", "def unexpectedError(self, filename, message):\n self.log.append(('unexpectedError', filename, message))\n" ]
""" API for the command-line I{pyflakes} tool. """ from __future__ import with_statement import ast import os import platform import re import sys from pyflakes import checker, __version__ from pyflakes import reporter as modReporter __all__ = ['check', 'checkPath', 'checkRecursive', 'iterSourceCode', 'main'] PYTHON_SHEBANG_REGEX = re.compile(br'^#!.*\bpython[23w]?\b\s*$') def check(codeString, filename, reporter=None): """ Check the Python source given by C{codeString} for flakes. @param codeString: The Python source to check. @type codeString: C{str} @param filename: The name of the file the source came from, used to report errors. @type filename: C{str} @param reporter: A L{Reporter} instance, where errors and warnings will be reported. @return: The number of warnings emitted. @rtype: C{int} """ if reporter is None: reporter = modReporter._makeDefaultReporter() # First, compile into an AST and handle syntax errors. try: tree = ast.parse(codeString, filename=filename) except SyntaxError: value = sys.exc_info()[1] msg = value.args[0] (lineno, offset, text) = value.lineno, value.offset, value.text if checker.PYPY: if text is None: lines = codeString.splitlines() if len(lines) >= lineno: text = lines[lineno - 1] if sys.version_info >= (3, ) and isinstance(text, bytes): try: text = text.decode('ascii') except UnicodeDecodeError: text = None offset -= 1 # If there's an encoding problem with the file, the text is None. if text is None: # Avoid using msg, since for the only known case, it contains a # bogus message that claims the encoding the file declared was # unknown. reporter.unexpectedError(filename, 'problem decoding source') else: reporter.syntaxError(filename, msg, lineno, offset, text) return 1 except Exception: reporter.unexpectedError(filename, 'problem decoding source') return 1 # Okay, it's syntactically valid. Now check it. file_tokens = checker.make_tokens(codeString) w = checker.Checker(tree, file_tokens=file_tokens, filename=filename) w.messages.sort(key=lambda m: m.lineno) for warning in w.messages: reporter.flake(warning) return len(w.messages) def isPythonFile(filename): """Return True if filename points to a Python file.""" if filename.endswith('.py'): return True # Avoid obvious Emacs backup files if filename.endswith("~"): return False max_bytes = 128 try: with open(filename, 'rb') as f: text = f.read(max_bytes) if not text: return False except IOError: return False first_line = text.splitlines()[0] return PYTHON_SHEBANG_REGEX.match(first_line) def iterSourceCode(paths): """ Iterate over all Python source files in C{paths}. @param paths: A list of paths. Directories will be recursed into and any .py files found will be yielded. Any non-directories will be yielded as-is. """ for path in paths: if os.path.isdir(path): for dirpath, dirnames, filenames in os.walk(path): for filename in filenames: full_path = os.path.join(dirpath, filename) if isPythonFile(full_path): yield full_path else: yield path def checkRecursive(paths, reporter): """ Recursively check all source files in C{paths}. @param paths: A list of paths to Python source files and directories containing Python source files. @param reporter: A L{Reporter} where all of the warnings and errors will be reported to. @return: The number of warnings found. """ warnings = 0 for sourcePath in iterSourceCode(paths): warnings += checkPath(sourcePath, reporter) return warnings def _exitOnSignal(sigName, message): """Handles a signal with sys.exit. Some of these signals (SIGPIPE, for example) don't exist or are invalid on Windows. So, ignore errors that might arise. """ import signal try: sigNumber = getattr(signal, sigName) except AttributeError: # the signal constants defined in the signal module are defined by # whether the C library supports them or not. So, SIGPIPE might not # even be defined. return def handler(sig, f): sys.exit(message) try: signal.signal(sigNumber, handler) except ValueError: # It's also possible the signal is defined, but then it's invalid. In # this case, signal.signal raises ValueError. pass def _get_version(): """ Retrieve and format package version along with python version & OS used """ return ('%s Python %s on %s' % (__version__, platform.python_version(), platform.system())) def main(prog=None, args=None): """Entry point for the script "pyflakes".""" import optparse # Handle "Keyboard Interrupt" and "Broken pipe" gracefully _exitOnSignal('SIGINT', '... stopped') _exitOnSignal('SIGPIPE', 1) parser = optparse.OptionParser(prog=prog, version=_get_version()) (__, args) = parser.parse_args(args=args) reporter = modReporter._makeDefaultReporter() if args: warnings = checkRecursive(args, reporter) else: warnings = check(sys.stdin.read(), '<stdin>', reporter) raise SystemExit(warnings > 0)
PyCQA/pyflakes
pyflakes/api.py
isPythonFile
python
def isPythonFile(filename): if filename.endswith('.py'): return True # Avoid obvious Emacs backup files if filename.endswith("~"): return False max_bytes = 128 try: with open(filename, 'rb') as f: text = f.read(max_bytes) if not text: return False except IOError: return False first_line = text.splitlines()[0] return PYTHON_SHEBANG_REGEX.match(first_line)
Return True if filename points to a Python file.
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/api.py#L102-L122
null
""" API for the command-line I{pyflakes} tool. """ from __future__ import with_statement import ast import os import platform import re import sys from pyflakes import checker, __version__ from pyflakes import reporter as modReporter __all__ = ['check', 'checkPath', 'checkRecursive', 'iterSourceCode', 'main'] PYTHON_SHEBANG_REGEX = re.compile(br'^#!.*\bpython[23w]?\b\s*$') def check(codeString, filename, reporter=None): """ Check the Python source given by C{codeString} for flakes. @param codeString: The Python source to check. @type codeString: C{str} @param filename: The name of the file the source came from, used to report errors. @type filename: C{str} @param reporter: A L{Reporter} instance, where errors and warnings will be reported. @return: The number of warnings emitted. @rtype: C{int} """ if reporter is None: reporter = modReporter._makeDefaultReporter() # First, compile into an AST and handle syntax errors. try: tree = ast.parse(codeString, filename=filename) except SyntaxError: value = sys.exc_info()[1] msg = value.args[0] (lineno, offset, text) = value.lineno, value.offset, value.text if checker.PYPY: if text is None: lines = codeString.splitlines() if len(lines) >= lineno: text = lines[lineno - 1] if sys.version_info >= (3, ) and isinstance(text, bytes): try: text = text.decode('ascii') except UnicodeDecodeError: text = None offset -= 1 # If there's an encoding problem with the file, the text is None. if text is None: # Avoid using msg, since for the only known case, it contains a # bogus message that claims the encoding the file declared was # unknown. reporter.unexpectedError(filename, 'problem decoding source') else: reporter.syntaxError(filename, msg, lineno, offset, text) return 1 except Exception: reporter.unexpectedError(filename, 'problem decoding source') return 1 # Okay, it's syntactically valid. Now check it. file_tokens = checker.make_tokens(codeString) w = checker.Checker(tree, file_tokens=file_tokens, filename=filename) w.messages.sort(key=lambda m: m.lineno) for warning in w.messages: reporter.flake(warning) return len(w.messages) def checkPath(filename, reporter=None): """ Check the given path, printing out any warnings detected. @param reporter: A L{Reporter} instance, where errors and warnings will be reported. @return: the number of warnings printed """ if reporter is None: reporter = modReporter._makeDefaultReporter() try: with open(filename, 'rb') as f: codestr = f.read() except IOError: msg = sys.exc_info()[1] reporter.unexpectedError(filename, msg.args[1]) return 1 return check(codestr, filename, reporter) def iterSourceCode(paths): """ Iterate over all Python source files in C{paths}. @param paths: A list of paths. Directories will be recursed into and any .py files found will be yielded. Any non-directories will be yielded as-is. """ for path in paths: if os.path.isdir(path): for dirpath, dirnames, filenames in os.walk(path): for filename in filenames: full_path = os.path.join(dirpath, filename) if isPythonFile(full_path): yield full_path else: yield path def checkRecursive(paths, reporter): """ Recursively check all source files in C{paths}. @param paths: A list of paths to Python source files and directories containing Python source files. @param reporter: A L{Reporter} where all of the warnings and errors will be reported to. @return: The number of warnings found. """ warnings = 0 for sourcePath in iterSourceCode(paths): warnings += checkPath(sourcePath, reporter) return warnings def _exitOnSignal(sigName, message): """Handles a signal with sys.exit. Some of these signals (SIGPIPE, for example) don't exist or are invalid on Windows. So, ignore errors that might arise. """ import signal try: sigNumber = getattr(signal, sigName) except AttributeError: # the signal constants defined in the signal module are defined by # whether the C library supports them or not. So, SIGPIPE might not # even be defined. return def handler(sig, f): sys.exit(message) try: signal.signal(sigNumber, handler) except ValueError: # It's also possible the signal is defined, but then it's invalid. In # this case, signal.signal raises ValueError. pass def _get_version(): """ Retrieve and format package version along with python version & OS used """ return ('%s Python %s on %s' % (__version__, platform.python_version(), platform.system())) def main(prog=None, args=None): """Entry point for the script "pyflakes".""" import optparse # Handle "Keyboard Interrupt" and "Broken pipe" gracefully _exitOnSignal('SIGINT', '... stopped') _exitOnSignal('SIGPIPE', 1) parser = optparse.OptionParser(prog=prog, version=_get_version()) (__, args) = parser.parse_args(args=args) reporter = modReporter._makeDefaultReporter() if args: warnings = checkRecursive(args, reporter) else: warnings = check(sys.stdin.read(), '<stdin>', reporter) raise SystemExit(warnings > 0)
PyCQA/pyflakes
pyflakes/api.py
iterSourceCode
python
def iterSourceCode(paths): for path in paths: if os.path.isdir(path): for dirpath, dirnames, filenames in os.walk(path): for filename in filenames: full_path = os.path.join(dirpath, filename) if isPythonFile(full_path): yield full_path else: yield path
Iterate over all Python source files in C{paths}. @param paths: A list of paths. Directories will be recursed into and any .py files found will be yielded. Any non-directories will be yielded as-is.
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/api.py#L125-L141
[ "def isPythonFile(filename):\n \"\"\"Return True if filename points to a Python file.\"\"\"\n if filename.endswith('.py'):\n return True\n\n # Avoid obvious Emacs backup files\n if filename.endswith(\"~\"):\n return False\n\n max_bytes = 128\n\n try:\n with open(filename, 'rb') as f:\n text = f.read(max_bytes)\n if not text:\n return False\n except IOError:\n return False\n\n first_line = text.splitlines()[0]\n return PYTHON_SHEBANG_REGEX.match(first_line)\n" ]
""" API for the command-line I{pyflakes} tool. """ from __future__ import with_statement import ast import os import platform import re import sys from pyflakes import checker, __version__ from pyflakes import reporter as modReporter __all__ = ['check', 'checkPath', 'checkRecursive', 'iterSourceCode', 'main'] PYTHON_SHEBANG_REGEX = re.compile(br'^#!.*\bpython[23w]?\b\s*$') def check(codeString, filename, reporter=None): """ Check the Python source given by C{codeString} for flakes. @param codeString: The Python source to check. @type codeString: C{str} @param filename: The name of the file the source came from, used to report errors. @type filename: C{str} @param reporter: A L{Reporter} instance, where errors and warnings will be reported. @return: The number of warnings emitted. @rtype: C{int} """ if reporter is None: reporter = modReporter._makeDefaultReporter() # First, compile into an AST and handle syntax errors. try: tree = ast.parse(codeString, filename=filename) except SyntaxError: value = sys.exc_info()[1] msg = value.args[0] (lineno, offset, text) = value.lineno, value.offset, value.text if checker.PYPY: if text is None: lines = codeString.splitlines() if len(lines) >= lineno: text = lines[lineno - 1] if sys.version_info >= (3, ) and isinstance(text, bytes): try: text = text.decode('ascii') except UnicodeDecodeError: text = None offset -= 1 # If there's an encoding problem with the file, the text is None. if text is None: # Avoid using msg, since for the only known case, it contains a # bogus message that claims the encoding the file declared was # unknown. reporter.unexpectedError(filename, 'problem decoding source') else: reporter.syntaxError(filename, msg, lineno, offset, text) return 1 except Exception: reporter.unexpectedError(filename, 'problem decoding source') return 1 # Okay, it's syntactically valid. Now check it. file_tokens = checker.make_tokens(codeString) w = checker.Checker(tree, file_tokens=file_tokens, filename=filename) w.messages.sort(key=lambda m: m.lineno) for warning in w.messages: reporter.flake(warning) return len(w.messages) def checkPath(filename, reporter=None): """ Check the given path, printing out any warnings detected. @param reporter: A L{Reporter} instance, where errors and warnings will be reported. @return: the number of warnings printed """ if reporter is None: reporter = modReporter._makeDefaultReporter() try: with open(filename, 'rb') as f: codestr = f.read() except IOError: msg = sys.exc_info()[1] reporter.unexpectedError(filename, msg.args[1]) return 1 return check(codestr, filename, reporter) def isPythonFile(filename): """Return True if filename points to a Python file.""" if filename.endswith('.py'): return True # Avoid obvious Emacs backup files if filename.endswith("~"): return False max_bytes = 128 try: with open(filename, 'rb') as f: text = f.read(max_bytes) if not text: return False except IOError: return False first_line = text.splitlines()[0] return PYTHON_SHEBANG_REGEX.match(first_line) def checkRecursive(paths, reporter): """ Recursively check all source files in C{paths}. @param paths: A list of paths to Python source files and directories containing Python source files. @param reporter: A L{Reporter} where all of the warnings and errors will be reported to. @return: The number of warnings found. """ warnings = 0 for sourcePath in iterSourceCode(paths): warnings += checkPath(sourcePath, reporter) return warnings def _exitOnSignal(sigName, message): """Handles a signal with sys.exit. Some of these signals (SIGPIPE, for example) don't exist or are invalid on Windows. So, ignore errors that might arise. """ import signal try: sigNumber = getattr(signal, sigName) except AttributeError: # the signal constants defined in the signal module are defined by # whether the C library supports them or not. So, SIGPIPE might not # even be defined. return def handler(sig, f): sys.exit(message) try: signal.signal(sigNumber, handler) except ValueError: # It's also possible the signal is defined, but then it's invalid. In # this case, signal.signal raises ValueError. pass def _get_version(): """ Retrieve and format package version along with python version & OS used """ return ('%s Python %s on %s' % (__version__, platform.python_version(), platform.system())) def main(prog=None, args=None): """Entry point for the script "pyflakes".""" import optparse # Handle "Keyboard Interrupt" and "Broken pipe" gracefully _exitOnSignal('SIGINT', '... stopped') _exitOnSignal('SIGPIPE', 1) parser = optparse.OptionParser(prog=prog, version=_get_version()) (__, args) = parser.parse_args(args=args) reporter = modReporter._makeDefaultReporter() if args: warnings = checkRecursive(args, reporter) else: warnings = check(sys.stdin.read(), '<stdin>', reporter) raise SystemExit(warnings > 0)
PyCQA/pyflakes
pyflakes/api.py
checkRecursive
python
def checkRecursive(paths, reporter): warnings = 0 for sourcePath in iterSourceCode(paths): warnings += checkPath(sourcePath, reporter) return warnings
Recursively check all source files in C{paths}. @param paths: A list of paths to Python source files and directories containing Python source files. @param reporter: A L{Reporter} where all of the warnings and errors will be reported to. @return: The number of warnings found.
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/api.py#L144-L157
[ "def checkPath(filename, reporter=None):\n \"\"\"\n Check the given path, printing out any warnings detected.\n\n @param reporter: A L{Reporter} instance, where errors and warnings will be\n reported.\n\n @return: the number of warnings printed\n \"\"\"\n if reporter is None:\n reporter = modReporter._makeDefaultReporter()\n try:\n with open(filename, 'rb') as f:\n codestr = f.read()\n except IOError:\n msg = sys.exc_info()[1]\n reporter.unexpectedError(filename, msg.args[1])\n return 1\n return check(codestr, filename, reporter)\n", "def iterSourceCode(paths):\n \"\"\"\n Iterate over all Python source files in C{paths}.\n\n @param paths: A list of paths. Directories will be recursed into and\n any .py files found will be yielded. Any non-directories will be\n yielded as-is.\n \"\"\"\n for path in paths:\n if os.path.isdir(path):\n for dirpath, dirnames, filenames in os.walk(path):\n for filename in filenames:\n full_path = os.path.join(dirpath, filename)\n if isPythonFile(full_path):\n yield full_path\n else:\n yield path\n" ]
""" API for the command-line I{pyflakes} tool. """ from __future__ import with_statement import ast import os import platform import re import sys from pyflakes import checker, __version__ from pyflakes import reporter as modReporter __all__ = ['check', 'checkPath', 'checkRecursive', 'iterSourceCode', 'main'] PYTHON_SHEBANG_REGEX = re.compile(br'^#!.*\bpython[23w]?\b\s*$') def check(codeString, filename, reporter=None): """ Check the Python source given by C{codeString} for flakes. @param codeString: The Python source to check. @type codeString: C{str} @param filename: The name of the file the source came from, used to report errors. @type filename: C{str} @param reporter: A L{Reporter} instance, where errors and warnings will be reported. @return: The number of warnings emitted. @rtype: C{int} """ if reporter is None: reporter = modReporter._makeDefaultReporter() # First, compile into an AST and handle syntax errors. try: tree = ast.parse(codeString, filename=filename) except SyntaxError: value = sys.exc_info()[1] msg = value.args[0] (lineno, offset, text) = value.lineno, value.offset, value.text if checker.PYPY: if text is None: lines = codeString.splitlines() if len(lines) >= lineno: text = lines[lineno - 1] if sys.version_info >= (3, ) and isinstance(text, bytes): try: text = text.decode('ascii') except UnicodeDecodeError: text = None offset -= 1 # If there's an encoding problem with the file, the text is None. if text is None: # Avoid using msg, since for the only known case, it contains a # bogus message that claims the encoding the file declared was # unknown. reporter.unexpectedError(filename, 'problem decoding source') else: reporter.syntaxError(filename, msg, lineno, offset, text) return 1 except Exception: reporter.unexpectedError(filename, 'problem decoding source') return 1 # Okay, it's syntactically valid. Now check it. file_tokens = checker.make_tokens(codeString) w = checker.Checker(tree, file_tokens=file_tokens, filename=filename) w.messages.sort(key=lambda m: m.lineno) for warning in w.messages: reporter.flake(warning) return len(w.messages) def checkPath(filename, reporter=None): """ Check the given path, printing out any warnings detected. @param reporter: A L{Reporter} instance, where errors and warnings will be reported. @return: the number of warnings printed """ if reporter is None: reporter = modReporter._makeDefaultReporter() try: with open(filename, 'rb') as f: codestr = f.read() except IOError: msg = sys.exc_info()[1] reporter.unexpectedError(filename, msg.args[1]) return 1 return check(codestr, filename, reporter) def isPythonFile(filename): """Return True if filename points to a Python file.""" if filename.endswith('.py'): return True # Avoid obvious Emacs backup files if filename.endswith("~"): return False max_bytes = 128 try: with open(filename, 'rb') as f: text = f.read(max_bytes) if not text: return False except IOError: return False first_line = text.splitlines()[0] return PYTHON_SHEBANG_REGEX.match(first_line) def iterSourceCode(paths): """ Iterate over all Python source files in C{paths}. @param paths: A list of paths. Directories will be recursed into and any .py files found will be yielded. Any non-directories will be yielded as-is. """ for path in paths: if os.path.isdir(path): for dirpath, dirnames, filenames in os.walk(path): for filename in filenames: full_path = os.path.join(dirpath, filename) if isPythonFile(full_path): yield full_path else: yield path def _exitOnSignal(sigName, message): """Handles a signal with sys.exit. Some of these signals (SIGPIPE, for example) don't exist or are invalid on Windows. So, ignore errors that might arise. """ import signal try: sigNumber = getattr(signal, sigName) except AttributeError: # the signal constants defined in the signal module are defined by # whether the C library supports them or not. So, SIGPIPE might not # even be defined. return def handler(sig, f): sys.exit(message) try: signal.signal(sigNumber, handler) except ValueError: # It's also possible the signal is defined, but then it's invalid. In # this case, signal.signal raises ValueError. pass def _get_version(): """ Retrieve and format package version along with python version & OS used """ return ('%s Python %s on %s' % (__version__, platform.python_version(), platform.system())) def main(prog=None, args=None): """Entry point for the script "pyflakes".""" import optparse # Handle "Keyboard Interrupt" and "Broken pipe" gracefully _exitOnSignal('SIGINT', '... stopped') _exitOnSignal('SIGPIPE', 1) parser = optparse.OptionParser(prog=prog, version=_get_version()) (__, args) = parser.parse_args(args=args) reporter = modReporter._makeDefaultReporter() if args: warnings = checkRecursive(args, reporter) else: warnings = check(sys.stdin.read(), '<stdin>', reporter) raise SystemExit(warnings > 0)
PyCQA/pyflakes
pyflakes/api.py
_exitOnSignal
python
def _exitOnSignal(sigName, message): import signal try: sigNumber = getattr(signal, sigName) except AttributeError: # the signal constants defined in the signal module are defined by # whether the C library supports them or not. So, SIGPIPE might not # even be defined. return def handler(sig, f): sys.exit(message) try: signal.signal(sigNumber, handler) except ValueError: # It's also possible the signal is defined, but then it's invalid. In # this case, signal.signal raises ValueError. pass
Handles a signal with sys.exit. Some of these signals (SIGPIPE, for example) don't exist or are invalid on Windows. So, ignore errors that might arise.
train
https://github.com/PyCQA/pyflakes/blob/232cb1d27ee134bf96adc8f37e53589dc259b159/pyflakes/api.py#L160-L184
null
""" API for the command-line I{pyflakes} tool. """ from __future__ import with_statement import ast import os import platform import re import sys from pyflakes import checker, __version__ from pyflakes import reporter as modReporter __all__ = ['check', 'checkPath', 'checkRecursive', 'iterSourceCode', 'main'] PYTHON_SHEBANG_REGEX = re.compile(br'^#!.*\bpython[23w]?\b\s*$') def check(codeString, filename, reporter=None): """ Check the Python source given by C{codeString} for flakes. @param codeString: The Python source to check. @type codeString: C{str} @param filename: The name of the file the source came from, used to report errors. @type filename: C{str} @param reporter: A L{Reporter} instance, where errors and warnings will be reported. @return: The number of warnings emitted. @rtype: C{int} """ if reporter is None: reporter = modReporter._makeDefaultReporter() # First, compile into an AST and handle syntax errors. try: tree = ast.parse(codeString, filename=filename) except SyntaxError: value = sys.exc_info()[1] msg = value.args[0] (lineno, offset, text) = value.lineno, value.offset, value.text if checker.PYPY: if text is None: lines = codeString.splitlines() if len(lines) >= lineno: text = lines[lineno - 1] if sys.version_info >= (3, ) and isinstance(text, bytes): try: text = text.decode('ascii') except UnicodeDecodeError: text = None offset -= 1 # If there's an encoding problem with the file, the text is None. if text is None: # Avoid using msg, since for the only known case, it contains a # bogus message that claims the encoding the file declared was # unknown. reporter.unexpectedError(filename, 'problem decoding source') else: reporter.syntaxError(filename, msg, lineno, offset, text) return 1 except Exception: reporter.unexpectedError(filename, 'problem decoding source') return 1 # Okay, it's syntactically valid. Now check it. file_tokens = checker.make_tokens(codeString) w = checker.Checker(tree, file_tokens=file_tokens, filename=filename) w.messages.sort(key=lambda m: m.lineno) for warning in w.messages: reporter.flake(warning) return len(w.messages) def checkPath(filename, reporter=None): """ Check the given path, printing out any warnings detected. @param reporter: A L{Reporter} instance, where errors and warnings will be reported. @return: the number of warnings printed """ if reporter is None: reporter = modReporter._makeDefaultReporter() try: with open(filename, 'rb') as f: codestr = f.read() except IOError: msg = sys.exc_info()[1] reporter.unexpectedError(filename, msg.args[1]) return 1 return check(codestr, filename, reporter) def isPythonFile(filename): """Return True if filename points to a Python file.""" if filename.endswith('.py'): return True # Avoid obvious Emacs backup files if filename.endswith("~"): return False max_bytes = 128 try: with open(filename, 'rb') as f: text = f.read(max_bytes) if not text: return False except IOError: return False first_line = text.splitlines()[0] return PYTHON_SHEBANG_REGEX.match(first_line) def iterSourceCode(paths): """ Iterate over all Python source files in C{paths}. @param paths: A list of paths. Directories will be recursed into and any .py files found will be yielded. Any non-directories will be yielded as-is. """ for path in paths: if os.path.isdir(path): for dirpath, dirnames, filenames in os.walk(path): for filename in filenames: full_path = os.path.join(dirpath, filename) if isPythonFile(full_path): yield full_path else: yield path def checkRecursive(paths, reporter): """ Recursively check all source files in C{paths}. @param paths: A list of paths to Python source files and directories containing Python source files. @param reporter: A L{Reporter} where all of the warnings and errors will be reported to. @return: The number of warnings found. """ warnings = 0 for sourcePath in iterSourceCode(paths): warnings += checkPath(sourcePath, reporter) return warnings def _get_version(): """ Retrieve and format package version along with python version & OS used """ return ('%s Python %s on %s' % (__version__, platform.python_version(), platform.system())) def main(prog=None, args=None): """Entry point for the script "pyflakes".""" import optparse # Handle "Keyboard Interrupt" and "Broken pipe" gracefully _exitOnSignal('SIGINT', '... stopped') _exitOnSignal('SIGPIPE', 1) parser = optparse.OptionParser(prog=prog, version=_get_version()) (__, args) = parser.parse_args(args=args) reporter = modReporter._makeDefaultReporter() if args: warnings = checkRecursive(args, reporter) else: warnings = check(sys.stdin.read(), '<stdin>', reporter) raise SystemExit(warnings > 0)
nicodv/kmodes
kmodes/kmodes.py
init_huang
python
def init_huang(X, n_clusters, dissim, random_state): n_attrs = X.shape[1] centroids = np.empty((n_clusters, n_attrs), dtype='object') # determine frequencies of attributes for iattr in range(n_attrs): freq = defaultdict(int) for curattr in X[:, iattr]: freq[curattr] += 1 # Sample centroids using the probabilities of attributes. # (I assume that's what's meant in the Huang [1998] paper; it works, # at least) # Note: sampling using population in static list with as many choices # as frequency counts. Since the counts are small integers, # memory consumption is low. choices = [chc for chc, wght in freq.items() for _ in range(wght)] # So that we are consistent between Python versions, # each with different dict ordering. choices = sorted(choices) centroids[:, iattr] = random_state.choice(choices, n_clusters) # The previously chosen centroids could result in empty clusters, # so set centroid to closest point in X. for ik in range(n_clusters): ndx = np.argsort(dissim(X, centroids[ik])) # We want the centroid to be unique, if possible. while np.all(X[ndx[0]] == centroids, axis=1).any() and ndx.shape[0] > 1: ndx = np.delete(ndx, 0) centroids[ik] = X[ndx[0]] return centroids
Initialize centroids according to method by Huang [1997].
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/kmodes.py#L21-L50
null
""" K-modes clustering for categorical data """ # pylint: disable=unused-argument,attribute-defined-outside-init from collections import defaultdict import numpy as np from joblib import Parallel, delayed from scipy import sparse from sklearn.base import BaseEstimator, ClusterMixin from sklearn.utils import check_random_state from sklearn.utils.validation import check_array from .util import get_max_value_key, encode_features, get_unique_rows, \ decode_centroids, pandas_to_numpy from .util.dissim import matching_dissim, ng_dissim def init_cao(X, n_clusters, dissim): """Initialize centroids according to method by Cao et al. [2009]. Note: O(N * attr * n_clusters**2), so watch out with large n_clusters """ n_points, n_attrs = X.shape centroids = np.empty((n_clusters, n_attrs), dtype='object') # Method is based on determining density of points. dens = np.zeros(n_points) for iattr in range(n_attrs): freq = defaultdict(int) for val in X[:, iattr]: freq[val] += 1 for ipoint in range(n_points): dens[ipoint] += freq[X[ipoint, iattr]] / float(n_points) / float(n_attrs) # Choose initial centroids based on distance and density. centroids[0] = X[np.argmax(dens)] if n_clusters > 1: # For the remaining centroids, choose maximum dens * dissim to the # (already assigned) centroid with the lowest dens * dissim. for ik in range(1, n_clusters): dd = np.empty((ik, n_points)) for ikk in range(ik): dd[ikk] = dissim(X, centroids[ikk]) * dens centroids[ik] = X[np.argmax(np.min(dd, axis=0))] return centroids def move_point_cat(point, ipoint, to_clust, from_clust, cl_attr_freq, membship, centroids): """Move point between clusters, categorical attributes.""" membship[to_clust, ipoint] = 1 membship[from_clust, ipoint] = 0 # Update frequencies of attributes in cluster. for iattr, curattr in enumerate(point): to_attr_counts = cl_attr_freq[to_clust][iattr] from_attr_counts = cl_attr_freq[from_clust][iattr] # Increment the attribute count for the new "to" cluster to_attr_counts[curattr] += 1 current_attribute_value_freq = to_attr_counts[curattr] current_centroid_value = centroids[to_clust][iattr] current_centroid_freq = to_attr_counts[current_centroid_value] if current_centroid_freq < current_attribute_value_freq: # We have incremented this value to the new mode. Update the centroid. centroids[to_clust][iattr] = curattr # Decrement the attribute count for the old "from" cluster from_attr_counts[curattr] -= 1 old_centroid_value = centroids[from_clust][iattr] if old_centroid_value == curattr: # We have just removed a count from the old centroid value. We need to # recalculate the centroid as it may no longer be the maximum centroids[from_clust][iattr] = get_max_value_key(from_attr_counts) return cl_attr_freq, membship, centroids def _labels_cost(X, centroids, dissim, membship=None): """Calculate labels and cost function given a matrix of points and a list of centroids for the k-modes algorithm. """ X = check_array(X) n_points = X.shape[0] cost = 0. labels = np.empty(n_points, dtype=np.uint16) for ipoint, curpoint in enumerate(X): diss = dissim(centroids, curpoint, X=X, membship=membship) clust = np.argmin(diss) labels[ipoint] = clust cost += diss[clust] return labels, cost def _k_modes_iter(X, centroids, cl_attr_freq, membship, dissim, random_state): """Single iteration of k-modes clustering algorithm""" moves = 0 for ipoint, curpoint in enumerate(X): clust = np.argmin(dissim(centroids, curpoint, X=X, membship=membship)) if membship[clust, ipoint]: # Point is already in its right place. continue # Move point, and update old/new cluster frequencies and centroids. moves += 1 old_clust = np.argwhere(membship[:, ipoint])[0][0] cl_attr_freq, membship, centroids = move_point_cat( curpoint, ipoint, clust, old_clust, cl_attr_freq, membship, centroids ) # In case of an empty cluster, reinitialize with a random point # from the largest cluster. if not membship[old_clust, :].any(): from_clust = membship.sum(axis=1).argmax() choices = [ii for ii, ch in enumerate(membship[from_clust, :]) if ch] rindx = random_state.choice(choices) cl_attr_freq, membship, centroids = move_point_cat( X[rindx], rindx, old_clust, from_clust, cl_attr_freq, membship, centroids ) return centroids, moves def k_modes_single(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, random_state): random_state = check_random_state(random_state) # _____ INIT _____ if verbose: print("Init: initializing centroids") if isinstance(init, str) and init.lower() == 'huang': centroids = init_huang(X, n_clusters, dissim, random_state) elif isinstance(init, str) and init.lower() == 'cao': centroids = init_cao(X, n_clusters, dissim) elif isinstance(init, str) and init.lower() == 'random': seeds = random_state.choice(range(n_points), n_clusters) centroids = X[seeds] elif hasattr(init, '__array__'): # Make sure init is a 2D array. if len(init.shape) == 1: init = np.atleast_2d(init).T assert init.shape[0] == n_clusters, \ "Wrong number of initial centroids in init ({}, should be {})." \ .format(init.shape[0], n_clusters) assert init.shape[1] == n_attrs, \ "Wrong number of attributes in init ({}, should be {})." \ .format(init.shape[1], n_attrs) centroids = np.asarray(init, dtype=np.uint16) else: raise NotImplementedError if verbose: print("Init: initializing clusters") membship = np.zeros((n_clusters, n_points), dtype=np.uint8) # cl_attr_freq is a list of lists with dictionaries that contain the # frequencies of values per cluster and attribute. cl_attr_freq = [[defaultdict(int) for _ in range(n_attrs)] for _ in range(n_clusters)] for ipoint, curpoint in enumerate(X): # Initial assignment to clusters clust = np.argmin(dissim(centroids, curpoint, X=X, membship=membship)) membship[clust, ipoint] = 1 # Count attribute values per cluster. for iattr, curattr in enumerate(curpoint): cl_attr_freq[clust][iattr][curattr] += 1 # Perform an initial centroid update. for ik in range(n_clusters): for iattr in range(n_attrs): if sum(membship[ik]) == 0: # Empty centroid, choose randomly centroids[ik, iattr] = random_state.choice(X[:, iattr]) else: centroids[ik, iattr] = get_max_value_key(cl_attr_freq[ik][iattr]) # _____ ITERATION _____ if verbose: print("Starting iterations...") itr = 0 labels = None converged = False cost = np.Inf while itr <= max_iter and not converged: itr += 1 centroids, moves = _k_modes_iter( X, centroids, cl_attr_freq, membship, dissim, random_state ) # All points seen in this iteration labels, ncost = _labels_cost(X, centroids, dissim, membship) converged = (moves == 0) or (ncost >= cost) cost = ncost if verbose: print("Run {}, iteration: {}/{}, moves: {}, cost: {}" .format(init_no + 1, itr, max_iter, moves, cost)) return centroids, labels, cost, itr def k_modes(X, n_clusters, max_iter, dissim, init, n_init, verbose, random_state, n_jobs): """k-modes algorithm""" random_state = check_random_state(random_state) if sparse.issparse(X): raise TypeError("k-modes does not support sparse data.") X = check_array(X, dtype=None) # Convert the categorical values in X to integers for speed. # Based on the unique values in X, we can make a mapping to achieve this. X, enc_map = encode_features(X) n_points, n_attrs = X.shape assert n_clusters <= n_points, "Cannot have more clusters ({}) " \ "than data points ({}).".format(n_clusters, n_points) # Are there more n_clusters than unique rows? Then set the unique # rows as initial values and skip iteration. unique = get_unique_rows(X) n_unique = unique.shape[0] if n_unique <= n_clusters: max_iter = 0 n_init = 1 n_clusters = n_unique init = unique results = [] seeds = random_state.randint(np.iinfo(np.int32).max, size=n_init) if n_jobs == 1: for init_no in range(n_init): results.append(k_modes_single(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, seeds[init_no])) else: results = Parallel(n_jobs=n_jobs, verbose=0)( delayed(k_modes_single)(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, seed) for init_no, seed in enumerate(seeds)) all_centroids, all_labels, all_costs, all_n_iters = zip(*results) best = np.argmin(all_costs) if n_init > 1 and verbose: print("Best run was number {}".format(best + 1)) return all_centroids[best], enc_map, all_labels[best], \ all_costs[best], all_n_iters[best] class KModes(BaseEstimator, ClusterMixin): """k-modes clustering algorithm for categorical data. Parameters ----------- n_clusters : int, optional, default: 8 The number of clusters to form as well as the number of centroids to generate. max_iter : int, default: 300 Maximum number of iterations of the k-modes algorithm for a single run. cat_dissim : func, default: matching_dissim Dissimilarity function used by the k-modes algorithm for categorical variables. Defaults to the matching dissimilarity function. init : {'Huang', 'Cao', 'random' or an ndarray}, default: 'Cao' Method for initialization: 'Huang': Method in Huang [1997, 1998] 'Cao': Method in Cao et al. [2009] 'random': choose 'n_clusters' observations (rows) at random from data for the initial centroids. If an ndarray is passed, it should be of shape (n_clusters, n_features) and gives the initial centroids. n_init : int, default: 10 Number of time the k-modes algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of cost. verbose : int, optional Verbosity mode. random_state : int, RandomState instance or None, optional, default: None If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. n_jobs : int, default: 1 The number of jobs to use for the computation. This works by computing each of the n_init runs in parallel. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Attributes ---------- cluster_centroids_ : array, [n_clusters, n_features] Categories of cluster centroids labels_ : Labels of each point cost_ : float Clustering cost, defined as the sum distance of all points to their respective cluster centroids. n_iter_ : int The number of iterations the algorithm ran for. Notes ----- See: Huang, Z.: Extensions to the k-modes algorithm for clustering large data sets with categorical values, Data Mining and Knowledge Discovery 2(3), 1998. """ def __init__(self, n_clusters=8, max_iter=100, cat_dissim=matching_dissim, init='Cao', n_init=1, verbose=0, random_state=None, n_jobs=1): self.n_clusters = n_clusters self.max_iter = max_iter self.cat_dissim = cat_dissim self.init = init self.n_init = n_init self.verbose = verbose self.random_state = random_state self.n_jobs = n_jobs if ((isinstance(self.init, str) and self.init == 'Cao') or hasattr(self.init, '__array__')) and self.n_init > 1: if self.verbose: print("Initialization method and algorithm are deterministic. " "Setting n_init to 1.") self.n_init = 1 def fit(self, X, y=None, **kwargs): """Compute k-modes clustering. Parameters ---------- X : array-like, shape=[n_samples, n_features] """ X = pandas_to_numpy(X) random_state = check_random_state(self.random_state) self._enc_cluster_centroids, self._enc_map, self.labels_,\ self.cost_, self.n_iter_ = k_modes(X, self.n_clusters, self.max_iter, self.cat_dissim, self.init, self.n_init, self.verbose, random_state, self.n_jobs) return self def fit_predict(self, X, y=None, **kwargs): """Compute cluster centroids and predict cluster index for each sample. Convenience method; equivalent to calling fit(X) followed by predict(X). """ return self.fit(X, **kwargs).predict(X, **kwargs) def predict(self, X, **kwargs): """Predict the closest cluster each sample in X belongs to. Parameters ---------- X : array-like, shape = [n_samples, n_features] New data to predict. Returns ------- labels : array, shape [n_samples,] Index of the cluster each sample belongs to. """ assert hasattr(self, '_enc_cluster_centroids'), "Model not yet fitted." if self.verbose and self.cat_dissim == ng_dissim: print("Ng's dissimilarity measure was used to train this model, " "but now that it is predicting the model will fall back to " "using simple matching dissimilarity.") X = pandas_to_numpy(X) X = check_array(X, dtype=None) X, _ = encode_features(X, enc_map=self._enc_map) return _labels_cost(X, self._enc_cluster_centroids, self.cat_dissim)[0] @property def cluster_centroids_(self): if hasattr(self, '_enc_cluster_centroids'): return decode_centroids(self._enc_cluster_centroids, self._enc_map) else: raise AttributeError("'{}' object has no attribute 'cluster_centroids_' " "because the model is not yet fitted.")
nicodv/kmodes
kmodes/kmodes.py
init_cao
python
def init_cao(X, n_clusters, dissim): n_points, n_attrs = X.shape centroids = np.empty((n_clusters, n_attrs), dtype='object') # Method is based on determining density of points. dens = np.zeros(n_points) for iattr in range(n_attrs): freq = defaultdict(int) for val in X[:, iattr]: freq[val] += 1 for ipoint in range(n_points): dens[ipoint] += freq[X[ipoint, iattr]] / float(n_points) / float(n_attrs) # Choose initial centroids based on distance and density. centroids[0] = X[np.argmax(dens)] if n_clusters > 1: # For the remaining centroids, choose maximum dens * dissim to the # (already assigned) centroid with the lowest dens * dissim. for ik in range(1, n_clusters): dd = np.empty((ik, n_points)) for ikk in range(ik): dd[ikk] = dissim(X, centroids[ikk]) * dens centroids[ik] = X[np.argmax(np.min(dd, axis=0))] return centroids
Initialize centroids according to method by Cao et al. [2009]. Note: O(N * attr * n_clusters**2), so watch out with large n_clusters
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/kmodes.py#L53-L80
null
""" K-modes clustering for categorical data """ # pylint: disable=unused-argument,attribute-defined-outside-init from collections import defaultdict import numpy as np from joblib import Parallel, delayed from scipy import sparse from sklearn.base import BaseEstimator, ClusterMixin from sklearn.utils import check_random_state from sklearn.utils.validation import check_array from .util import get_max_value_key, encode_features, get_unique_rows, \ decode_centroids, pandas_to_numpy from .util.dissim import matching_dissim, ng_dissim def init_huang(X, n_clusters, dissim, random_state): """Initialize centroids according to method by Huang [1997].""" n_attrs = X.shape[1] centroids = np.empty((n_clusters, n_attrs), dtype='object') # determine frequencies of attributes for iattr in range(n_attrs): freq = defaultdict(int) for curattr in X[:, iattr]: freq[curattr] += 1 # Sample centroids using the probabilities of attributes. # (I assume that's what's meant in the Huang [1998] paper; it works, # at least) # Note: sampling using population in static list with as many choices # as frequency counts. Since the counts are small integers, # memory consumption is low. choices = [chc for chc, wght in freq.items() for _ in range(wght)] # So that we are consistent between Python versions, # each with different dict ordering. choices = sorted(choices) centroids[:, iattr] = random_state.choice(choices, n_clusters) # The previously chosen centroids could result in empty clusters, # so set centroid to closest point in X. for ik in range(n_clusters): ndx = np.argsort(dissim(X, centroids[ik])) # We want the centroid to be unique, if possible. while np.all(X[ndx[0]] == centroids, axis=1).any() and ndx.shape[0] > 1: ndx = np.delete(ndx, 0) centroids[ik] = X[ndx[0]] return centroids def move_point_cat(point, ipoint, to_clust, from_clust, cl_attr_freq, membship, centroids): """Move point between clusters, categorical attributes.""" membship[to_clust, ipoint] = 1 membship[from_clust, ipoint] = 0 # Update frequencies of attributes in cluster. for iattr, curattr in enumerate(point): to_attr_counts = cl_attr_freq[to_clust][iattr] from_attr_counts = cl_attr_freq[from_clust][iattr] # Increment the attribute count for the new "to" cluster to_attr_counts[curattr] += 1 current_attribute_value_freq = to_attr_counts[curattr] current_centroid_value = centroids[to_clust][iattr] current_centroid_freq = to_attr_counts[current_centroid_value] if current_centroid_freq < current_attribute_value_freq: # We have incremented this value to the new mode. Update the centroid. centroids[to_clust][iattr] = curattr # Decrement the attribute count for the old "from" cluster from_attr_counts[curattr] -= 1 old_centroid_value = centroids[from_clust][iattr] if old_centroid_value == curattr: # We have just removed a count from the old centroid value. We need to # recalculate the centroid as it may no longer be the maximum centroids[from_clust][iattr] = get_max_value_key(from_attr_counts) return cl_attr_freq, membship, centroids def _labels_cost(X, centroids, dissim, membship=None): """Calculate labels and cost function given a matrix of points and a list of centroids for the k-modes algorithm. """ X = check_array(X) n_points = X.shape[0] cost = 0. labels = np.empty(n_points, dtype=np.uint16) for ipoint, curpoint in enumerate(X): diss = dissim(centroids, curpoint, X=X, membship=membship) clust = np.argmin(diss) labels[ipoint] = clust cost += diss[clust] return labels, cost def _k_modes_iter(X, centroids, cl_attr_freq, membship, dissim, random_state): """Single iteration of k-modes clustering algorithm""" moves = 0 for ipoint, curpoint in enumerate(X): clust = np.argmin(dissim(centroids, curpoint, X=X, membship=membship)) if membship[clust, ipoint]: # Point is already in its right place. continue # Move point, and update old/new cluster frequencies and centroids. moves += 1 old_clust = np.argwhere(membship[:, ipoint])[0][0] cl_attr_freq, membship, centroids = move_point_cat( curpoint, ipoint, clust, old_clust, cl_attr_freq, membship, centroids ) # In case of an empty cluster, reinitialize with a random point # from the largest cluster. if not membship[old_clust, :].any(): from_clust = membship.sum(axis=1).argmax() choices = [ii for ii, ch in enumerate(membship[from_clust, :]) if ch] rindx = random_state.choice(choices) cl_attr_freq, membship, centroids = move_point_cat( X[rindx], rindx, old_clust, from_clust, cl_attr_freq, membship, centroids ) return centroids, moves def k_modes_single(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, random_state): random_state = check_random_state(random_state) # _____ INIT _____ if verbose: print("Init: initializing centroids") if isinstance(init, str) and init.lower() == 'huang': centroids = init_huang(X, n_clusters, dissim, random_state) elif isinstance(init, str) and init.lower() == 'cao': centroids = init_cao(X, n_clusters, dissim) elif isinstance(init, str) and init.lower() == 'random': seeds = random_state.choice(range(n_points), n_clusters) centroids = X[seeds] elif hasattr(init, '__array__'): # Make sure init is a 2D array. if len(init.shape) == 1: init = np.atleast_2d(init).T assert init.shape[0] == n_clusters, \ "Wrong number of initial centroids in init ({}, should be {})." \ .format(init.shape[0], n_clusters) assert init.shape[1] == n_attrs, \ "Wrong number of attributes in init ({}, should be {})." \ .format(init.shape[1], n_attrs) centroids = np.asarray(init, dtype=np.uint16) else: raise NotImplementedError if verbose: print("Init: initializing clusters") membship = np.zeros((n_clusters, n_points), dtype=np.uint8) # cl_attr_freq is a list of lists with dictionaries that contain the # frequencies of values per cluster and attribute. cl_attr_freq = [[defaultdict(int) for _ in range(n_attrs)] for _ in range(n_clusters)] for ipoint, curpoint in enumerate(X): # Initial assignment to clusters clust = np.argmin(dissim(centroids, curpoint, X=X, membship=membship)) membship[clust, ipoint] = 1 # Count attribute values per cluster. for iattr, curattr in enumerate(curpoint): cl_attr_freq[clust][iattr][curattr] += 1 # Perform an initial centroid update. for ik in range(n_clusters): for iattr in range(n_attrs): if sum(membship[ik]) == 0: # Empty centroid, choose randomly centroids[ik, iattr] = random_state.choice(X[:, iattr]) else: centroids[ik, iattr] = get_max_value_key(cl_attr_freq[ik][iattr]) # _____ ITERATION _____ if verbose: print("Starting iterations...") itr = 0 labels = None converged = False cost = np.Inf while itr <= max_iter and not converged: itr += 1 centroids, moves = _k_modes_iter( X, centroids, cl_attr_freq, membship, dissim, random_state ) # All points seen in this iteration labels, ncost = _labels_cost(X, centroids, dissim, membship) converged = (moves == 0) or (ncost >= cost) cost = ncost if verbose: print("Run {}, iteration: {}/{}, moves: {}, cost: {}" .format(init_no + 1, itr, max_iter, moves, cost)) return centroids, labels, cost, itr def k_modes(X, n_clusters, max_iter, dissim, init, n_init, verbose, random_state, n_jobs): """k-modes algorithm""" random_state = check_random_state(random_state) if sparse.issparse(X): raise TypeError("k-modes does not support sparse data.") X = check_array(X, dtype=None) # Convert the categorical values in X to integers for speed. # Based on the unique values in X, we can make a mapping to achieve this. X, enc_map = encode_features(X) n_points, n_attrs = X.shape assert n_clusters <= n_points, "Cannot have more clusters ({}) " \ "than data points ({}).".format(n_clusters, n_points) # Are there more n_clusters than unique rows? Then set the unique # rows as initial values and skip iteration. unique = get_unique_rows(X) n_unique = unique.shape[0] if n_unique <= n_clusters: max_iter = 0 n_init = 1 n_clusters = n_unique init = unique results = [] seeds = random_state.randint(np.iinfo(np.int32).max, size=n_init) if n_jobs == 1: for init_no in range(n_init): results.append(k_modes_single(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, seeds[init_no])) else: results = Parallel(n_jobs=n_jobs, verbose=0)( delayed(k_modes_single)(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, seed) for init_no, seed in enumerate(seeds)) all_centroids, all_labels, all_costs, all_n_iters = zip(*results) best = np.argmin(all_costs) if n_init > 1 and verbose: print("Best run was number {}".format(best + 1)) return all_centroids[best], enc_map, all_labels[best], \ all_costs[best], all_n_iters[best] class KModes(BaseEstimator, ClusterMixin): """k-modes clustering algorithm for categorical data. Parameters ----------- n_clusters : int, optional, default: 8 The number of clusters to form as well as the number of centroids to generate. max_iter : int, default: 300 Maximum number of iterations of the k-modes algorithm for a single run. cat_dissim : func, default: matching_dissim Dissimilarity function used by the k-modes algorithm for categorical variables. Defaults to the matching dissimilarity function. init : {'Huang', 'Cao', 'random' or an ndarray}, default: 'Cao' Method for initialization: 'Huang': Method in Huang [1997, 1998] 'Cao': Method in Cao et al. [2009] 'random': choose 'n_clusters' observations (rows) at random from data for the initial centroids. If an ndarray is passed, it should be of shape (n_clusters, n_features) and gives the initial centroids. n_init : int, default: 10 Number of time the k-modes algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of cost. verbose : int, optional Verbosity mode. random_state : int, RandomState instance or None, optional, default: None If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. n_jobs : int, default: 1 The number of jobs to use for the computation. This works by computing each of the n_init runs in parallel. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Attributes ---------- cluster_centroids_ : array, [n_clusters, n_features] Categories of cluster centroids labels_ : Labels of each point cost_ : float Clustering cost, defined as the sum distance of all points to their respective cluster centroids. n_iter_ : int The number of iterations the algorithm ran for. Notes ----- See: Huang, Z.: Extensions to the k-modes algorithm for clustering large data sets with categorical values, Data Mining and Knowledge Discovery 2(3), 1998. """ def __init__(self, n_clusters=8, max_iter=100, cat_dissim=matching_dissim, init='Cao', n_init=1, verbose=0, random_state=None, n_jobs=1): self.n_clusters = n_clusters self.max_iter = max_iter self.cat_dissim = cat_dissim self.init = init self.n_init = n_init self.verbose = verbose self.random_state = random_state self.n_jobs = n_jobs if ((isinstance(self.init, str) and self.init == 'Cao') or hasattr(self.init, '__array__')) and self.n_init > 1: if self.verbose: print("Initialization method and algorithm are deterministic. " "Setting n_init to 1.") self.n_init = 1 def fit(self, X, y=None, **kwargs): """Compute k-modes clustering. Parameters ---------- X : array-like, shape=[n_samples, n_features] """ X = pandas_to_numpy(X) random_state = check_random_state(self.random_state) self._enc_cluster_centroids, self._enc_map, self.labels_,\ self.cost_, self.n_iter_ = k_modes(X, self.n_clusters, self.max_iter, self.cat_dissim, self.init, self.n_init, self.verbose, random_state, self.n_jobs) return self def fit_predict(self, X, y=None, **kwargs): """Compute cluster centroids and predict cluster index for each sample. Convenience method; equivalent to calling fit(X) followed by predict(X). """ return self.fit(X, **kwargs).predict(X, **kwargs) def predict(self, X, **kwargs): """Predict the closest cluster each sample in X belongs to. Parameters ---------- X : array-like, shape = [n_samples, n_features] New data to predict. Returns ------- labels : array, shape [n_samples,] Index of the cluster each sample belongs to. """ assert hasattr(self, '_enc_cluster_centroids'), "Model not yet fitted." if self.verbose and self.cat_dissim == ng_dissim: print("Ng's dissimilarity measure was used to train this model, " "but now that it is predicting the model will fall back to " "using simple matching dissimilarity.") X = pandas_to_numpy(X) X = check_array(X, dtype=None) X, _ = encode_features(X, enc_map=self._enc_map) return _labels_cost(X, self._enc_cluster_centroids, self.cat_dissim)[0] @property def cluster_centroids_(self): if hasattr(self, '_enc_cluster_centroids'): return decode_centroids(self._enc_cluster_centroids, self._enc_map) else: raise AttributeError("'{}' object has no attribute 'cluster_centroids_' " "because the model is not yet fitted.")
nicodv/kmodes
kmodes/kmodes.py
move_point_cat
python
def move_point_cat(point, ipoint, to_clust, from_clust, cl_attr_freq, membship, centroids): membship[to_clust, ipoint] = 1 membship[from_clust, ipoint] = 0 # Update frequencies of attributes in cluster. for iattr, curattr in enumerate(point): to_attr_counts = cl_attr_freq[to_clust][iattr] from_attr_counts = cl_attr_freq[from_clust][iattr] # Increment the attribute count for the new "to" cluster to_attr_counts[curattr] += 1 current_attribute_value_freq = to_attr_counts[curattr] current_centroid_value = centroids[to_clust][iattr] current_centroid_freq = to_attr_counts[current_centroid_value] if current_centroid_freq < current_attribute_value_freq: # We have incremented this value to the new mode. Update the centroid. centroids[to_clust][iattr] = curattr # Decrement the attribute count for the old "from" cluster from_attr_counts[curattr] -= 1 old_centroid_value = centroids[from_clust][iattr] if old_centroid_value == curattr: # We have just removed a count from the old centroid value. We need to # recalculate the centroid as it may no longer be the maximum centroids[from_clust][iattr] = get_max_value_key(from_attr_counts) return cl_attr_freq, membship, centroids
Move point between clusters, categorical attributes.
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/kmodes.py#L83-L112
[ "def get_max_value_key(dic):\n \"\"\"Gets the key for the maximum value in a dict.\"\"\"\n v = np.array(list(dic.values()))\n k = np.array(list(dic.keys()))\n\n maxima = np.where(v == np.max(v))[0]\n if len(maxima) == 1:\n return k[maxima[0]]\n else:\n # In order to be consistent, always selects the minimum key\n # (guaranteed to be unique) when there are multiple maximum values.\n return k[maxima[np.argmin(k[maxima])]]\n" ]
""" K-modes clustering for categorical data """ # pylint: disable=unused-argument,attribute-defined-outside-init from collections import defaultdict import numpy as np from joblib import Parallel, delayed from scipy import sparse from sklearn.base import BaseEstimator, ClusterMixin from sklearn.utils import check_random_state from sklearn.utils.validation import check_array from .util import get_max_value_key, encode_features, get_unique_rows, \ decode_centroids, pandas_to_numpy from .util.dissim import matching_dissim, ng_dissim def init_huang(X, n_clusters, dissim, random_state): """Initialize centroids according to method by Huang [1997].""" n_attrs = X.shape[1] centroids = np.empty((n_clusters, n_attrs), dtype='object') # determine frequencies of attributes for iattr in range(n_attrs): freq = defaultdict(int) for curattr in X[:, iattr]: freq[curattr] += 1 # Sample centroids using the probabilities of attributes. # (I assume that's what's meant in the Huang [1998] paper; it works, # at least) # Note: sampling using population in static list with as many choices # as frequency counts. Since the counts are small integers, # memory consumption is low. choices = [chc for chc, wght in freq.items() for _ in range(wght)] # So that we are consistent between Python versions, # each with different dict ordering. choices = sorted(choices) centroids[:, iattr] = random_state.choice(choices, n_clusters) # The previously chosen centroids could result in empty clusters, # so set centroid to closest point in X. for ik in range(n_clusters): ndx = np.argsort(dissim(X, centroids[ik])) # We want the centroid to be unique, if possible. while np.all(X[ndx[0]] == centroids, axis=1).any() and ndx.shape[0] > 1: ndx = np.delete(ndx, 0) centroids[ik] = X[ndx[0]] return centroids def init_cao(X, n_clusters, dissim): """Initialize centroids according to method by Cao et al. [2009]. Note: O(N * attr * n_clusters**2), so watch out with large n_clusters """ n_points, n_attrs = X.shape centroids = np.empty((n_clusters, n_attrs), dtype='object') # Method is based on determining density of points. dens = np.zeros(n_points) for iattr in range(n_attrs): freq = defaultdict(int) for val in X[:, iattr]: freq[val] += 1 for ipoint in range(n_points): dens[ipoint] += freq[X[ipoint, iattr]] / float(n_points) / float(n_attrs) # Choose initial centroids based on distance and density. centroids[0] = X[np.argmax(dens)] if n_clusters > 1: # For the remaining centroids, choose maximum dens * dissim to the # (already assigned) centroid with the lowest dens * dissim. for ik in range(1, n_clusters): dd = np.empty((ik, n_points)) for ikk in range(ik): dd[ikk] = dissim(X, centroids[ikk]) * dens centroids[ik] = X[np.argmax(np.min(dd, axis=0))] return centroids def _labels_cost(X, centroids, dissim, membship=None): """Calculate labels and cost function given a matrix of points and a list of centroids for the k-modes algorithm. """ X = check_array(X) n_points = X.shape[0] cost = 0. labels = np.empty(n_points, dtype=np.uint16) for ipoint, curpoint in enumerate(X): diss = dissim(centroids, curpoint, X=X, membship=membship) clust = np.argmin(diss) labels[ipoint] = clust cost += diss[clust] return labels, cost def _k_modes_iter(X, centroids, cl_attr_freq, membship, dissim, random_state): """Single iteration of k-modes clustering algorithm""" moves = 0 for ipoint, curpoint in enumerate(X): clust = np.argmin(dissim(centroids, curpoint, X=X, membship=membship)) if membship[clust, ipoint]: # Point is already in its right place. continue # Move point, and update old/new cluster frequencies and centroids. moves += 1 old_clust = np.argwhere(membship[:, ipoint])[0][0] cl_attr_freq, membship, centroids = move_point_cat( curpoint, ipoint, clust, old_clust, cl_attr_freq, membship, centroids ) # In case of an empty cluster, reinitialize with a random point # from the largest cluster. if not membship[old_clust, :].any(): from_clust = membship.sum(axis=1).argmax() choices = [ii for ii, ch in enumerate(membship[from_clust, :]) if ch] rindx = random_state.choice(choices) cl_attr_freq, membship, centroids = move_point_cat( X[rindx], rindx, old_clust, from_clust, cl_attr_freq, membship, centroids ) return centroids, moves def k_modes_single(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, random_state): random_state = check_random_state(random_state) # _____ INIT _____ if verbose: print("Init: initializing centroids") if isinstance(init, str) and init.lower() == 'huang': centroids = init_huang(X, n_clusters, dissim, random_state) elif isinstance(init, str) and init.lower() == 'cao': centroids = init_cao(X, n_clusters, dissim) elif isinstance(init, str) and init.lower() == 'random': seeds = random_state.choice(range(n_points), n_clusters) centroids = X[seeds] elif hasattr(init, '__array__'): # Make sure init is a 2D array. if len(init.shape) == 1: init = np.atleast_2d(init).T assert init.shape[0] == n_clusters, \ "Wrong number of initial centroids in init ({}, should be {})." \ .format(init.shape[0], n_clusters) assert init.shape[1] == n_attrs, \ "Wrong number of attributes in init ({}, should be {})." \ .format(init.shape[1], n_attrs) centroids = np.asarray(init, dtype=np.uint16) else: raise NotImplementedError if verbose: print("Init: initializing clusters") membship = np.zeros((n_clusters, n_points), dtype=np.uint8) # cl_attr_freq is a list of lists with dictionaries that contain the # frequencies of values per cluster and attribute. cl_attr_freq = [[defaultdict(int) for _ in range(n_attrs)] for _ in range(n_clusters)] for ipoint, curpoint in enumerate(X): # Initial assignment to clusters clust = np.argmin(dissim(centroids, curpoint, X=X, membship=membship)) membship[clust, ipoint] = 1 # Count attribute values per cluster. for iattr, curattr in enumerate(curpoint): cl_attr_freq[clust][iattr][curattr] += 1 # Perform an initial centroid update. for ik in range(n_clusters): for iattr in range(n_attrs): if sum(membship[ik]) == 0: # Empty centroid, choose randomly centroids[ik, iattr] = random_state.choice(X[:, iattr]) else: centroids[ik, iattr] = get_max_value_key(cl_attr_freq[ik][iattr]) # _____ ITERATION _____ if verbose: print("Starting iterations...") itr = 0 labels = None converged = False cost = np.Inf while itr <= max_iter and not converged: itr += 1 centroids, moves = _k_modes_iter( X, centroids, cl_attr_freq, membship, dissim, random_state ) # All points seen in this iteration labels, ncost = _labels_cost(X, centroids, dissim, membship) converged = (moves == 0) or (ncost >= cost) cost = ncost if verbose: print("Run {}, iteration: {}/{}, moves: {}, cost: {}" .format(init_no + 1, itr, max_iter, moves, cost)) return centroids, labels, cost, itr def k_modes(X, n_clusters, max_iter, dissim, init, n_init, verbose, random_state, n_jobs): """k-modes algorithm""" random_state = check_random_state(random_state) if sparse.issparse(X): raise TypeError("k-modes does not support sparse data.") X = check_array(X, dtype=None) # Convert the categorical values in X to integers for speed. # Based on the unique values in X, we can make a mapping to achieve this. X, enc_map = encode_features(X) n_points, n_attrs = X.shape assert n_clusters <= n_points, "Cannot have more clusters ({}) " \ "than data points ({}).".format(n_clusters, n_points) # Are there more n_clusters than unique rows? Then set the unique # rows as initial values and skip iteration. unique = get_unique_rows(X) n_unique = unique.shape[0] if n_unique <= n_clusters: max_iter = 0 n_init = 1 n_clusters = n_unique init = unique results = [] seeds = random_state.randint(np.iinfo(np.int32).max, size=n_init) if n_jobs == 1: for init_no in range(n_init): results.append(k_modes_single(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, seeds[init_no])) else: results = Parallel(n_jobs=n_jobs, verbose=0)( delayed(k_modes_single)(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, seed) for init_no, seed in enumerate(seeds)) all_centroids, all_labels, all_costs, all_n_iters = zip(*results) best = np.argmin(all_costs) if n_init > 1 and verbose: print("Best run was number {}".format(best + 1)) return all_centroids[best], enc_map, all_labels[best], \ all_costs[best], all_n_iters[best] class KModes(BaseEstimator, ClusterMixin): """k-modes clustering algorithm for categorical data. Parameters ----------- n_clusters : int, optional, default: 8 The number of clusters to form as well as the number of centroids to generate. max_iter : int, default: 300 Maximum number of iterations of the k-modes algorithm for a single run. cat_dissim : func, default: matching_dissim Dissimilarity function used by the k-modes algorithm for categorical variables. Defaults to the matching dissimilarity function. init : {'Huang', 'Cao', 'random' or an ndarray}, default: 'Cao' Method for initialization: 'Huang': Method in Huang [1997, 1998] 'Cao': Method in Cao et al. [2009] 'random': choose 'n_clusters' observations (rows) at random from data for the initial centroids. If an ndarray is passed, it should be of shape (n_clusters, n_features) and gives the initial centroids. n_init : int, default: 10 Number of time the k-modes algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of cost. verbose : int, optional Verbosity mode. random_state : int, RandomState instance or None, optional, default: None If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. n_jobs : int, default: 1 The number of jobs to use for the computation. This works by computing each of the n_init runs in parallel. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Attributes ---------- cluster_centroids_ : array, [n_clusters, n_features] Categories of cluster centroids labels_ : Labels of each point cost_ : float Clustering cost, defined as the sum distance of all points to their respective cluster centroids. n_iter_ : int The number of iterations the algorithm ran for. Notes ----- See: Huang, Z.: Extensions to the k-modes algorithm for clustering large data sets with categorical values, Data Mining and Knowledge Discovery 2(3), 1998. """ def __init__(self, n_clusters=8, max_iter=100, cat_dissim=matching_dissim, init='Cao', n_init=1, verbose=0, random_state=None, n_jobs=1): self.n_clusters = n_clusters self.max_iter = max_iter self.cat_dissim = cat_dissim self.init = init self.n_init = n_init self.verbose = verbose self.random_state = random_state self.n_jobs = n_jobs if ((isinstance(self.init, str) and self.init == 'Cao') or hasattr(self.init, '__array__')) and self.n_init > 1: if self.verbose: print("Initialization method and algorithm are deterministic. " "Setting n_init to 1.") self.n_init = 1 def fit(self, X, y=None, **kwargs): """Compute k-modes clustering. Parameters ---------- X : array-like, shape=[n_samples, n_features] """ X = pandas_to_numpy(X) random_state = check_random_state(self.random_state) self._enc_cluster_centroids, self._enc_map, self.labels_,\ self.cost_, self.n_iter_ = k_modes(X, self.n_clusters, self.max_iter, self.cat_dissim, self.init, self.n_init, self.verbose, random_state, self.n_jobs) return self def fit_predict(self, X, y=None, **kwargs): """Compute cluster centroids and predict cluster index for each sample. Convenience method; equivalent to calling fit(X) followed by predict(X). """ return self.fit(X, **kwargs).predict(X, **kwargs) def predict(self, X, **kwargs): """Predict the closest cluster each sample in X belongs to. Parameters ---------- X : array-like, shape = [n_samples, n_features] New data to predict. Returns ------- labels : array, shape [n_samples,] Index of the cluster each sample belongs to. """ assert hasattr(self, '_enc_cluster_centroids'), "Model not yet fitted." if self.verbose and self.cat_dissim == ng_dissim: print("Ng's dissimilarity measure was used to train this model, " "but now that it is predicting the model will fall back to " "using simple matching dissimilarity.") X = pandas_to_numpy(X) X = check_array(X, dtype=None) X, _ = encode_features(X, enc_map=self._enc_map) return _labels_cost(X, self._enc_cluster_centroids, self.cat_dissim)[0] @property def cluster_centroids_(self): if hasattr(self, '_enc_cluster_centroids'): return decode_centroids(self._enc_cluster_centroids, self._enc_map) else: raise AttributeError("'{}' object has no attribute 'cluster_centroids_' " "because the model is not yet fitted.")
nicodv/kmodes
kmodes/kmodes.py
_labels_cost
python
def _labels_cost(X, centroids, dissim, membship=None): X = check_array(X) n_points = X.shape[0] cost = 0. labels = np.empty(n_points, dtype=np.uint16) for ipoint, curpoint in enumerate(X): diss = dissim(centroids, curpoint, X=X, membship=membship) clust = np.argmin(diss) labels[ipoint] = clust cost += diss[clust] return labels, cost
Calculate labels and cost function given a matrix of points and a list of centroids for the k-modes algorithm.
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/kmodes.py#L115-L131
null
""" K-modes clustering for categorical data """ # pylint: disable=unused-argument,attribute-defined-outside-init from collections import defaultdict import numpy as np from joblib import Parallel, delayed from scipy import sparse from sklearn.base import BaseEstimator, ClusterMixin from sklearn.utils import check_random_state from sklearn.utils.validation import check_array from .util import get_max_value_key, encode_features, get_unique_rows, \ decode_centroids, pandas_to_numpy from .util.dissim import matching_dissim, ng_dissim def init_huang(X, n_clusters, dissim, random_state): """Initialize centroids according to method by Huang [1997].""" n_attrs = X.shape[1] centroids = np.empty((n_clusters, n_attrs), dtype='object') # determine frequencies of attributes for iattr in range(n_attrs): freq = defaultdict(int) for curattr in X[:, iattr]: freq[curattr] += 1 # Sample centroids using the probabilities of attributes. # (I assume that's what's meant in the Huang [1998] paper; it works, # at least) # Note: sampling using population in static list with as many choices # as frequency counts. Since the counts are small integers, # memory consumption is low. choices = [chc for chc, wght in freq.items() for _ in range(wght)] # So that we are consistent between Python versions, # each with different dict ordering. choices = sorted(choices) centroids[:, iattr] = random_state.choice(choices, n_clusters) # The previously chosen centroids could result in empty clusters, # so set centroid to closest point in X. for ik in range(n_clusters): ndx = np.argsort(dissim(X, centroids[ik])) # We want the centroid to be unique, if possible. while np.all(X[ndx[0]] == centroids, axis=1).any() and ndx.shape[0] > 1: ndx = np.delete(ndx, 0) centroids[ik] = X[ndx[0]] return centroids def init_cao(X, n_clusters, dissim): """Initialize centroids according to method by Cao et al. [2009]. Note: O(N * attr * n_clusters**2), so watch out with large n_clusters """ n_points, n_attrs = X.shape centroids = np.empty((n_clusters, n_attrs), dtype='object') # Method is based on determining density of points. dens = np.zeros(n_points) for iattr in range(n_attrs): freq = defaultdict(int) for val in X[:, iattr]: freq[val] += 1 for ipoint in range(n_points): dens[ipoint] += freq[X[ipoint, iattr]] / float(n_points) / float(n_attrs) # Choose initial centroids based on distance and density. centroids[0] = X[np.argmax(dens)] if n_clusters > 1: # For the remaining centroids, choose maximum dens * dissim to the # (already assigned) centroid with the lowest dens * dissim. for ik in range(1, n_clusters): dd = np.empty((ik, n_points)) for ikk in range(ik): dd[ikk] = dissim(X, centroids[ikk]) * dens centroids[ik] = X[np.argmax(np.min(dd, axis=0))] return centroids def move_point_cat(point, ipoint, to_clust, from_clust, cl_attr_freq, membship, centroids): """Move point between clusters, categorical attributes.""" membship[to_clust, ipoint] = 1 membship[from_clust, ipoint] = 0 # Update frequencies of attributes in cluster. for iattr, curattr in enumerate(point): to_attr_counts = cl_attr_freq[to_clust][iattr] from_attr_counts = cl_attr_freq[from_clust][iattr] # Increment the attribute count for the new "to" cluster to_attr_counts[curattr] += 1 current_attribute_value_freq = to_attr_counts[curattr] current_centroid_value = centroids[to_clust][iattr] current_centroid_freq = to_attr_counts[current_centroid_value] if current_centroid_freq < current_attribute_value_freq: # We have incremented this value to the new mode. Update the centroid. centroids[to_clust][iattr] = curattr # Decrement the attribute count for the old "from" cluster from_attr_counts[curattr] -= 1 old_centroid_value = centroids[from_clust][iattr] if old_centroid_value == curattr: # We have just removed a count from the old centroid value. We need to # recalculate the centroid as it may no longer be the maximum centroids[from_clust][iattr] = get_max_value_key(from_attr_counts) return cl_attr_freq, membship, centroids def _k_modes_iter(X, centroids, cl_attr_freq, membship, dissim, random_state): """Single iteration of k-modes clustering algorithm""" moves = 0 for ipoint, curpoint in enumerate(X): clust = np.argmin(dissim(centroids, curpoint, X=X, membship=membship)) if membship[clust, ipoint]: # Point is already in its right place. continue # Move point, and update old/new cluster frequencies and centroids. moves += 1 old_clust = np.argwhere(membship[:, ipoint])[0][0] cl_attr_freq, membship, centroids = move_point_cat( curpoint, ipoint, clust, old_clust, cl_attr_freq, membship, centroids ) # In case of an empty cluster, reinitialize with a random point # from the largest cluster. if not membship[old_clust, :].any(): from_clust = membship.sum(axis=1).argmax() choices = [ii for ii, ch in enumerate(membship[from_clust, :]) if ch] rindx = random_state.choice(choices) cl_attr_freq, membship, centroids = move_point_cat( X[rindx], rindx, old_clust, from_clust, cl_attr_freq, membship, centroids ) return centroids, moves def k_modes_single(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, random_state): random_state = check_random_state(random_state) # _____ INIT _____ if verbose: print("Init: initializing centroids") if isinstance(init, str) and init.lower() == 'huang': centroids = init_huang(X, n_clusters, dissim, random_state) elif isinstance(init, str) and init.lower() == 'cao': centroids = init_cao(X, n_clusters, dissim) elif isinstance(init, str) and init.lower() == 'random': seeds = random_state.choice(range(n_points), n_clusters) centroids = X[seeds] elif hasattr(init, '__array__'): # Make sure init is a 2D array. if len(init.shape) == 1: init = np.atleast_2d(init).T assert init.shape[0] == n_clusters, \ "Wrong number of initial centroids in init ({}, should be {})." \ .format(init.shape[0], n_clusters) assert init.shape[1] == n_attrs, \ "Wrong number of attributes in init ({}, should be {})." \ .format(init.shape[1], n_attrs) centroids = np.asarray(init, dtype=np.uint16) else: raise NotImplementedError if verbose: print("Init: initializing clusters") membship = np.zeros((n_clusters, n_points), dtype=np.uint8) # cl_attr_freq is a list of lists with dictionaries that contain the # frequencies of values per cluster and attribute. cl_attr_freq = [[defaultdict(int) for _ in range(n_attrs)] for _ in range(n_clusters)] for ipoint, curpoint in enumerate(X): # Initial assignment to clusters clust = np.argmin(dissim(centroids, curpoint, X=X, membship=membship)) membship[clust, ipoint] = 1 # Count attribute values per cluster. for iattr, curattr in enumerate(curpoint): cl_attr_freq[clust][iattr][curattr] += 1 # Perform an initial centroid update. for ik in range(n_clusters): for iattr in range(n_attrs): if sum(membship[ik]) == 0: # Empty centroid, choose randomly centroids[ik, iattr] = random_state.choice(X[:, iattr]) else: centroids[ik, iattr] = get_max_value_key(cl_attr_freq[ik][iattr]) # _____ ITERATION _____ if verbose: print("Starting iterations...") itr = 0 labels = None converged = False cost = np.Inf while itr <= max_iter and not converged: itr += 1 centroids, moves = _k_modes_iter( X, centroids, cl_attr_freq, membship, dissim, random_state ) # All points seen in this iteration labels, ncost = _labels_cost(X, centroids, dissim, membship) converged = (moves == 0) or (ncost >= cost) cost = ncost if verbose: print("Run {}, iteration: {}/{}, moves: {}, cost: {}" .format(init_no + 1, itr, max_iter, moves, cost)) return centroids, labels, cost, itr def k_modes(X, n_clusters, max_iter, dissim, init, n_init, verbose, random_state, n_jobs): """k-modes algorithm""" random_state = check_random_state(random_state) if sparse.issparse(X): raise TypeError("k-modes does not support sparse data.") X = check_array(X, dtype=None) # Convert the categorical values in X to integers for speed. # Based on the unique values in X, we can make a mapping to achieve this. X, enc_map = encode_features(X) n_points, n_attrs = X.shape assert n_clusters <= n_points, "Cannot have more clusters ({}) " \ "than data points ({}).".format(n_clusters, n_points) # Are there more n_clusters than unique rows? Then set the unique # rows as initial values and skip iteration. unique = get_unique_rows(X) n_unique = unique.shape[0] if n_unique <= n_clusters: max_iter = 0 n_init = 1 n_clusters = n_unique init = unique results = [] seeds = random_state.randint(np.iinfo(np.int32).max, size=n_init) if n_jobs == 1: for init_no in range(n_init): results.append(k_modes_single(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, seeds[init_no])) else: results = Parallel(n_jobs=n_jobs, verbose=0)( delayed(k_modes_single)(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, seed) for init_no, seed in enumerate(seeds)) all_centroids, all_labels, all_costs, all_n_iters = zip(*results) best = np.argmin(all_costs) if n_init > 1 and verbose: print("Best run was number {}".format(best + 1)) return all_centroids[best], enc_map, all_labels[best], \ all_costs[best], all_n_iters[best] class KModes(BaseEstimator, ClusterMixin): """k-modes clustering algorithm for categorical data. Parameters ----------- n_clusters : int, optional, default: 8 The number of clusters to form as well as the number of centroids to generate. max_iter : int, default: 300 Maximum number of iterations of the k-modes algorithm for a single run. cat_dissim : func, default: matching_dissim Dissimilarity function used by the k-modes algorithm for categorical variables. Defaults to the matching dissimilarity function. init : {'Huang', 'Cao', 'random' or an ndarray}, default: 'Cao' Method for initialization: 'Huang': Method in Huang [1997, 1998] 'Cao': Method in Cao et al. [2009] 'random': choose 'n_clusters' observations (rows) at random from data for the initial centroids. If an ndarray is passed, it should be of shape (n_clusters, n_features) and gives the initial centroids. n_init : int, default: 10 Number of time the k-modes algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of cost. verbose : int, optional Verbosity mode. random_state : int, RandomState instance or None, optional, default: None If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. n_jobs : int, default: 1 The number of jobs to use for the computation. This works by computing each of the n_init runs in parallel. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Attributes ---------- cluster_centroids_ : array, [n_clusters, n_features] Categories of cluster centroids labels_ : Labels of each point cost_ : float Clustering cost, defined as the sum distance of all points to their respective cluster centroids. n_iter_ : int The number of iterations the algorithm ran for. Notes ----- See: Huang, Z.: Extensions to the k-modes algorithm for clustering large data sets with categorical values, Data Mining and Knowledge Discovery 2(3), 1998. """ def __init__(self, n_clusters=8, max_iter=100, cat_dissim=matching_dissim, init='Cao', n_init=1, verbose=0, random_state=None, n_jobs=1): self.n_clusters = n_clusters self.max_iter = max_iter self.cat_dissim = cat_dissim self.init = init self.n_init = n_init self.verbose = verbose self.random_state = random_state self.n_jobs = n_jobs if ((isinstance(self.init, str) and self.init == 'Cao') or hasattr(self.init, '__array__')) and self.n_init > 1: if self.verbose: print("Initialization method and algorithm are deterministic. " "Setting n_init to 1.") self.n_init = 1 def fit(self, X, y=None, **kwargs): """Compute k-modes clustering. Parameters ---------- X : array-like, shape=[n_samples, n_features] """ X = pandas_to_numpy(X) random_state = check_random_state(self.random_state) self._enc_cluster_centroids, self._enc_map, self.labels_,\ self.cost_, self.n_iter_ = k_modes(X, self.n_clusters, self.max_iter, self.cat_dissim, self.init, self.n_init, self.verbose, random_state, self.n_jobs) return self def fit_predict(self, X, y=None, **kwargs): """Compute cluster centroids and predict cluster index for each sample. Convenience method; equivalent to calling fit(X) followed by predict(X). """ return self.fit(X, **kwargs).predict(X, **kwargs) def predict(self, X, **kwargs): """Predict the closest cluster each sample in X belongs to. Parameters ---------- X : array-like, shape = [n_samples, n_features] New data to predict. Returns ------- labels : array, shape [n_samples,] Index of the cluster each sample belongs to. """ assert hasattr(self, '_enc_cluster_centroids'), "Model not yet fitted." if self.verbose and self.cat_dissim == ng_dissim: print("Ng's dissimilarity measure was used to train this model, " "but now that it is predicting the model will fall back to " "using simple matching dissimilarity.") X = pandas_to_numpy(X) X = check_array(X, dtype=None) X, _ = encode_features(X, enc_map=self._enc_map) return _labels_cost(X, self._enc_cluster_centroids, self.cat_dissim)[0] @property def cluster_centroids_(self): if hasattr(self, '_enc_cluster_centroids'): return decode_centroids(self._enc_cluster_centroids, self._enc_map) else: raise AttributeError("'{}' object has no attribute 'cluster_centroids_' " "because the model is not yet fitted.")
nicodv/kmodes
kmodes/kmodes.py
_k_modes_iter
python
def _k_modes_iter(X, centroids, cl_attr_freq, membship, dissim, random_state): moves = 0 for ipoint, curpoint in enumerate(X): clust = np.argmin(dissim(centroids, curpoint, X=X, membship=membship)) if membship[clust, ipoint]: # Point is already in its right place. continue # Move point, and update old/new cluster frequencies and centroids. moves += 1 old_clust = np.argwhere(membship[:, ipoint])[0][0] cl_attr_freq, membship, centroids = move_point_cat( curpoint, ipoint, clust, old_clust, cl_attr_freq, membship, centroids ) # In case of an empty cluster, reinitialize with a random point # from the largest cluster. if not membship[old_clust, :].any(): from_clust = membship.sum(axis=1).argmax() choices = [ii for ii, ch in enumerate(membship[from_clust, :]) if ch] rindx = random_state.choice(choices) cl_attr_freq, membship, centroids = move_point_cat( X[rindx], rindx, old_clust, from_clust, cl_attr_freq, membship, centroids ) return centroids, moves
Single iteration of k-modes clustering algorithm
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/kmodes.py#L134-L162
null
""" K-modes clustering for categorical data """ # pylint: disable=unused-argument,attribute-defined-outside-init from collections import defaultdict import numpy as np from joblib import Parallel, delayed from scipy import sparse from sklearn.base import BaseEstimator, ClusterMixin from sklearn.utils import check_random_state from sklearn.utils.validation import check_array from .util import get_max_value_key, encode_features, get_unique_rows, \ decode_centroids, pandas_to_numpy from .util.dissim import matching_dissim, ng_dissim def init_huang(X, n_clusters, dissim, random_state): """Initialize centroids according to method by Huang [1997].""" n_attrs = X.shape[1] centroids = np.empty((n_clusters, n_attrs), dtype='object') # determine frequencies of attributes for iattr in range(n_attrs): freq = defaultdict(int) for curattr in X[:, iattr]: freq[curattr] += 1 # Sample centroids using the probabilities of attributes. # (I assume that's what's meant in the Huang [1998] paper; it works, # at least) # Note: sampling using population in static list with as many choices # as frequency counts. Since the counts are small integers, # memory consumption is low. choices = [chc for chc, wght in freq.items() for _ in range(wght)] # So that we are consistent between Python versions, # each with different dict ordering. choices = sorted(choices) centroids[:, iattr] = random_state.choice(choices, n_clusters) # The previously chosen centroids could result in empty clusters, # so set centroid to closest point in X. for ik in range(n_clusters): ndx = np.argsort(dissim(X, centroids[ik])) # We want the centroid to be unique, if possible. while np.all(X[ndx[0]] == centroids, axis=1).any() and ndx.shape[0] > 1: ndx = np.delete(ndx, 0) centroids[ik] = X[ndx[0]] return centroids def init_cao(X, n_clusters, dissim): """Initialize centroids according to method by Cao et al. [2009]. Note: O(N * attr * n_clusters**2), so watch out with large n_clusters """ n_points, n_attrs = X.shape centroids = np.empty((n_clusters, n_attrs), dtype='object') # Method is based on determining density of points. dens = np.zeros(n_points) for iattr in range(n_attrs): freq = defaultdict(int) for val in X[:, iattr]: freq[val] += 1 for ipoint in range(n_points): dens[ipoint] += freq[X[ipoint, iattr]] / float(n_points) / float(n_attrs) # Choose initial centroids based on distance and density. centroids[0] = X[np.argmax(dens)] if n_clusters > 1: # For the remaining centroids, choose maximum dens * dissim to the # (already assigned) centroid with the lowest dens * dissim. for ik in range(1, n_clusters): dd = np.empty((ik, n_points)) for ikk in range(ik): dd[ikk] = dissim(X, centroids[ikk]) * dens centroids[ik] = X[np.argmax(np.min(dd, axis=0))] return centroids def move_point_cat(point, ipoint, to_clust, from_clust, cl_attr_freq, membship, centroids): """Move point between clusters, categorical attributes.""" membship[to_clust, ipoint] = 1 membship[from_clust, ipoint] = 0 # Update frequencies of attributes in cluster. for iattr, curattr in enumerate(point): to_attr_counts = cl_attr_freq[to_clust][iattr] from_attr_counts = cl_attr_freq[from_clust][iattr] # Increment the attribute count for the new "to" cluster to_attr_counts[curattr] += 1 current_attribute_value_freq = to_attr_counts[curattr] current_centroid_value = centroids[to_clust][iattr] current_centroid_freq = to_attr_counts[current_centroid_value] if current_centroid_freq < current_attribute_value_freq: # We have incremented this value to the new mode. Update the centroid. centroids[to_clust][iattr] = curattr # Decrement the attribute count for the old "from" cluster from_attr_counts[curattr] -= 1 old_centroid_value = centroids[from_clust][iattr] if old_centroid_value == curattr: # We have just removed a count from the old centroid value. We need to # recalculate the centroid as it may no longer be the maximum centroids[from_clust][iattr] = get_max_value_key(from_attr_counts) return cl_attr_freq, membship, centroids def _labels_cost(X, centroids, dissim, membship=None): """Calculate labels and cost function given a matrix of points and a list of centroids for the k-modes algorithm. """ X = check_array(X) n_points = X.shape[0] cost = 0. labels = np.empty(n_points, dtype=np.uint16) for ipoint, curpoint in enumerate(X): diss = dissim(centroids, curpoint, X=X, membship=membship) clust = np.argmin(diss) labels[ipoint] = clust cost += diss[clust] return labels, cost def k_modes_single(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, random_state): random_state = check_random_state(random_state) # _____ INIT _____ if verbose: print("Init: initializing centroids") if isinstance(init, str) and init.lower() == 'huang': centroids = init_huang(X, n_clusters, dissim, random_state) elif isinstance(init, str) and init.lower() == 'cao': centroids = init_cao(X, n_clusters, dissim) elif isinstance(init, str) and init.lower() == 'random': seeds = random_state.choice(range(n_points), n_clusters) centroids = X[seeds] elif hasattr(init, '__array__'): # Make sure init is a 2D array. if len(init.shape) == 1: init = np.atleast_2d(init).T assert init.shape[0] == n_clusters, \ "Wrong number of initial centroids in init ({}, should be {})." \ .format(init.shape[0], n_clusters) assert init.shape[1] == n_attrs, \ "Wrong number of attributes in init ({}, should be {})." \ .format(init.shape[1], n_attrs) centroids = np.asarray(init, dtype=np.uint16) else: raise NotImplementedError if verbose: print("Init: initializing clusters") membship = np.zeros((n_clusters, n_points), dtype=np.uint8) # cl_attr_freq is a list of lists with dictionaries that contain the # frequencies of values per cluster and attribute. cl_attr_freq = [[defaultdict(int) for _ in range(n_attrs)] for _ in range(n_clusters)] for ipoint, curpoint in enumerate(X): # Initial assignment to clusters clust = np.argmin(dissim(centroids, curpoint, X=X, membship=membship)) membship[clust, ipoint] = 1 # Count attribute values per cluster. for iattr, curattr in enumerate(curpoint): cl_attr_freq[clust][iattr][curattr] += 1 # Perform an initial centroid update. for ik in range(n_clusters): for iattr in range(n_attrs): if sum(membship[ik]) == 0: # Empty centroid, choose randomly centroids[ik, iattr] = random_state.choice(X[:, iattr]) else: centroids[ik, iattr] = get_max_value_key(cl_attr_freq[ik][iattr]) # _____ ITERATION _____ if verbose: print("Starting iterations...") itr = 0 labels = None converged = False cost = np.Inf while itr <= max_iter and not converged: itr += 1 centroids, moves = _k_modes_iter( X, centroids, cl_attr_freq, membship, dissim, random_state ) # All points seen in this iteration labels, ncost = _labels_cost(X, centroids, dissim, membship) converged = (moves == 0) or (ncost >= cost) cost = ncost if verbose: print("Run {}, iteration: {}/{}, moves: {}, cost: {}" .format(init_no + 1, itr, max_iter, moves, cost)) return centroids, labels, cost, itr def k_modes(X, n_clusters, max_iter, dissim, init, n_init, verbose, random_state, n_jobs): """k-modes algorithm""" random_state = check_random_state(random_state) if sparse.issparse(X): raise TypeError("k-modes does not support sparse data.") X = check_array(X, dtype=None) # Convert the categorical values in X to integers for speed. # Based on the unique values in X, we can make a mapping to achieve this. X, enc_map = encode_features(X) n_points, n_attrs = X.shape assert n_clusters <= n_points, "Cannot have more clusters ({}) " \ "than data points ({}).".format(n_clusters, n_points) # Are there more n_clusters than unique rows? Then set the unique # rows as initial values and skip iteration. unique = get_unique_rows(X) n_unique = unique.shape[0] if n_unique <= n_clusters: max_iter = 0 n_init = 1 n_clusters = n_unique init = unique results = [] seeds = random_state.randint(np.iinfo(np.int32).max, size=n_init) if n_jobs == 1: for init_no in range(n_init): results.append(k_modes_single(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, seeds[init_no])) else: results = Parallel(n_jobs=n_jobs, verbose=0)( delayed(k_modes_single)(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, seed) for init_no, seed in enumerate(seeds)) all_centroids, all_labels, all_costs, all_n_iters = zip(*results) best = np.argmin(all_costs) if n_init > 1 and verbose: print("Best run was number {}".format(best + 1)) return all_centroids[best], enc_map, all_labels[best], \ all_costs[best], all_n_iters[best] class KModes(BaseEstimator, ClusterMixin): """k-modes clustering algorithm for categorical data. Parameters ----------- n_clusters : int, optional, default: 8 The number of clusters to form as well as the number of centroids to generate. max_iter : int, default: 300 Maximum number of iterations of the k-modes algorithm for a single run. cat_dissim : func, default: matching_dissim Dissimilarity function used by the k-modes algorithm for categorical variables. Defaults to the matching dissimilarity function. init : {'Huang', 'Cao', 'random' or an ndarray}, default: 'Cao' Method for initialization: 'Huang': Method in Huang [1997, 1998] 'Cao': Method in Cao et al. [2009] 'random': choose 'n_clusters' observations (rows) at random from data for the initial centroids. If an ndarray is passed, it should be of shape (n_clusters, n_features) and gives the initial centroids. n_init : int, default: 10 Number of time the k-modes algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of cost. verbose : int, optional Verbosity mode. random_state : int, RandomState instance or None, optional, default: None If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. n_jobs : int, default: 1 The number of jobs to use for the computation. This works by computing each of the n_init runs in parallel. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Attributes ---------- cluster_centroids_ : array, [n_clusters, n_features] Categories of cluster centroids labels_ : Labels of each point cost_ : float Clustering cost, defined as the sum distance of all points to their respective cluster centroids. n_iter_ : int The number of iterations the algorithm ran for. Notes ----- See: Huang, Z.: Extensions to the k-modes algorithm for clustering large data sets with categorical values, Data Mining and Knowledge Discovery 2(3), 1998. """ def __init__(self, n_clusters=8, max_iter=100, cat_dissim=matching_dissim, init='Cao', n_init=1, verbose=0, random_state=None, n_jobs=1): self.n_clusters = n_clusters self.max_iter = max_iter self.cat_dissim = cat_dissim self.init = init self.n_init = n_init self.verbose = verbose self.random_state = random_state self.n_jobs = n_jobs if ((isinstance(self.init, str) and self.init == 'Cao') or hasattr(self.init, '__array__')) and self.n_init > 1: if self.verbose: print("Initialization method and algorithm are deterministic. " "Setting n_init to 1.") self.n_init = 1 def fit(self, X, y=None, **kwargs): """Compute k-modes clustering. Parameters ---------- X : array-like, shape=[n_samples, n_features] """ X = pandas_to_numpy(X) random_state = check_random_state(self.random_state) self._enc_cluster_centroids, self._enc_map, self.labels_,\ self.cost_, self.n_iter_ = k_modes(X, self.n_clusters, self.max_iter, self.cat_dissim, self.init, self.n_init, self.verbose, random_state, self.n_jobs) return self def fit_predict(self, X, y=None, **kwargs): """Compute cluster centroids and predict cluster index for each sample. Convenience method; equivalent to calling fit(X) followed by predict(X). """ return self.fit(X, **kwargs).predict(X, **kwargs) def predict(self, X, **kwargs): """Predict the closest cluster each sample in X belongs to. Parameters ---------- X : array-like, shape = [n_samples, n_features] New data to predict. Returns ------- labels : array, shape [n_samples,] Index of the cluster each sample belongs to. """ assert hasattr(self, '_enc_cluster_centroids'), "Model not yet fitted." if self.verbose and self.cat_dissim == ng_dissim: print("Ng's dissimilarity measure was used to train this model, " "but now that it is predicting the model will fall back to " "using simple matching dissimilarity.") X = pandas_to_numpy(X) X = check_array(X, dtype=None) X, _ = encode_features(X, enc_map=self._enc_map) return _labels_cost(X, self._enc_cluster_centroids, self.cat_dissim)[0] @property def cluster_centroids_(self): if hasattr(self, '_enc_cluster_centroids'): return decode_centroids(self._enc_cluster_centroids, self._enc_map) else: raise AttributeError("'{}' object has no attribute 'cluster_centroids_' " "because the model is not yet fitted.")
nicodv/kmodes
kmodes/kmodes.py
k_modes
python
def k_modes(X, n_clusters, max_iter, dissim, init, n_init, verbose, random_state, n_jobs): random_state = check_random_state(random_state) if sparse.issparse(X): raise TypeError("k-modes does not support sparse data.") X = check_array(X, dtype=None) # Convert the categorical values in X to integers for speed. # Based on the unique values in X, we can make a mapping to achieve this. X, enc_map = encode_features(X) n_points, n_attrs = X.shape assert n_clusters <= n_points, "Cannot have more clusters ({}) " \ "than data points ({}).".format(n_clusters, n_points) # Are there more n_clusters than unique rows? Then set the unique # rows as initial values and skip iteration. unique = get_unique_rows(X) n_unique = unique.shape[0] if n_unique <= n_clusters: max_iter = 0 n_init = 1 n_clusters = n_unique init = unique results = [] seeds = random_state.randint(np.iinfo(np.int32).max, size=n_init) if n_jobs == 1: for init_no in range(n_init): results.append(k_modes_single(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, seeds[init_no])) else: results = Parallel(n_jobs=n_jobs, verbose=0)( delayed(k_modes_single)(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, seed) for init_no, seed in enumerate(seeds)) all_centroids, all_labels, all_costs, all_n_iters = zip(*results) best = np.argmin(all_costs) if n_init > 1 and verbose: print("Best run was number {}".format(best + 1)) return all_centroids[best], enc_map, all_labels[best], \ all_costs[best], all_n_iters[best]
k-modes algorithm
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/kmodes.py#L243-L287
[ "def encode_features(X, enc_map=None):\n \"\"\"Converts categorical values in each column of X to integers in the range\n [0, n_unique_values_in_column - 1], if X is not already of integer type.\n\n If mapping is not provided, it is calculated based on the values in X.\n\n Unknown values during prediction get a value of -1. np.NaNs are ignored\n during encoding, and get treated as unknowns during prediction.\n \"\"\"\n if enc_map is None:\n fit = True\n # We will calculate enc_map, so initialize the list of column mappings.\n enc_map = []\n else:\n fit = False\n\n Xenc = np.zeros(X.shape).astype('int')\n for ii in range(X.shape[1]):\n if fit:\n col_enc = {val: jj for jj, val in enumerate(np.unique(X[:, ii]))\n if not (isinstance(val, float) and np.isnan(val))}\n enc_map.append(col_enc)\n # Unknown categories (including np.NaNs) all get a value of -1.\n Xenc[:, ii] = np.array([enc_map[ii].get(x, -1) for x in X[:, ii]])\n\n return Xenc, enc_map\n", "def get_unique_rows(a):\n \"\"\"Gets the unique rows in a numpy array.\"\"\"\n return np.vstack(list({tuple(row) for row in a}))\n", "def k_modes_single(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no,\n verbose, random_state):\n random_state = check_random_state(random_state)\n # _____ INIT _____\n if verbose:\n print(\"Init: initializing centroids\")\n if isinstance(init, str) and init.lower() == 'huang':\n centroids = init_huang(X, n_clusters, dissim, random_state)\n elif isinstance(init, str) and init.lower() == 'cao':\n centroids = init_cao(X, n_clusters, dissim)\n elif isinstance(init, str) and init.lower() == 'random':\n seeds = random_state.choice(range(n_points), n_clusters)\n centroids = X[seeds]\n elif hasattr(init, '__array__'):\n # Make sure init is a 2D array.\n if len(init.shape) == 1:\n init = np.atleast_2d(init).T\n assert init.shape[0] == n_clusters, \\\n \"Wrong number of initial centroids in init ({}, should be {}).\" \\\n .format(init.shape[0], n_clusters)\n assert init.shape[1] == n_attrs, \\\n \"Wrong number of attributes in init ({}, should be {}).\" \\\n .format(init.shape[1], n_attrs)\n centroids = np.asarray(init, dtype=np.uint16)\n else:\n raise NotImplementedError\n\n if verbose:\n print(\"Init: initializing clusters\")\n membship = np.zeros((n_clusters, n_points), dtype=np.uint8)\n # cl_attr_freq is a list of lists with dictionaries that contain the\n # frequencies of values per cluster and attribute.\n cl_attr_freq = [[defaultdict(int) for _ in range(n_attrs)]\n for _ in range(n_clusters)]\n for ipoint, curpoint in enumerate(X):\n # Initial assignment to clusters\n clust = np.argmin(dissim(centroids, curpoint, X=X, membship=membship))\n membship[clust, ipoint] = 1\n # Count attribute values per cluster.\n for iattr, curattr in enumerate(curpoint):\n cl_attr_freq[clust][iattr][curattr] += 1\n # Perform an initial centroid update.\n for ik in range(n_clusters):\n for iattr in range(n_attrs):\n if sum(membship[ik]) == 0:\n # Empty centroid, choose randomly\n centroids[ik, iattr] = random_state.choice(X[:, iattr])\n else:\n centroids[ik, iattr] = get_max_value_key(cl_attr_freq[ik][iattr])\n\n # _____ ITERATION _____\n if verbose:\n print(\"Starting iterations...\")\n itr = 0\n labels = None\n converged = False\n cost = np.Inf\n while itr <= max_iter and not converged:\n itr += 1\n centroids, moves = _k_modes_iter(\n X,\n centroids,\n cl_attr_freq,\n membship,\n dissim,\n random_state\n )\n # All points seen in this iteration\n labels, ncost = _labels_cost(X, centroids, dissim, membship)\n converged = (moves == 0) or (ncost >= cost)\n cost = ncost\n if verbose:\n print(\"Run {}, iteration: {}/{}, moves: {}, cost: {}\"\n .format(init_no + 1, itr, max_iter, moves, cost))\n\n return centroids, labels, cost, itr\n" ]
""" K-modes clustering for categorical data """ # pylint: disable=unused-argument,attribute-defined-outside-init from collections import defaultdict import numpy as np from joblib import Parallel, delayed from scipy import sparse from sklearn.base import BaseEstimator, ClusterMixin from sklearn.utils import check_random_state from sklearn.utils.validation import check_array from .util import get_max_value_key, encode_features, get_unique_rows, \ decode_centroids, pandas_to_numpy from .util.dissim import matching_dissim, ng_dissim def init_huang(X, n_clusters, dissim, random_state): """Initialize centroids according to method by Huang [1997].""" n_attrs = X.shape[1] centroids = np.empty((n_clusters, n_attrs), dtype='object') # determine frequencies of attributes for iattr in range(n_attrs): freq = defaultdict(int) for curattr in X[:, iattr]: freq[curattr] += 1 # Sample centroids using the probabilities of attributes. # (I assume that's what's meant in the Huang [1998] paper; it works, # at least) # Note: sampling using population in static list with as many choices # as frequency counts. Since the counts are small integers, # memory consumption is low. choices = [chc for chc, wght in freq.items() for _ in range(wght)] # So that we are consistent between Python versions, # each with different dict ordering. choices = sorted(choices) centroids[:, iattr] = random_state.choice(choices, n_clusters) # The previously chosen centroids could result in empty clusters, # so set centroid to closest point in X. for ik in range(n_clusters): ndx = np.argsort(dissim(X, centroids[ik])) # We want the centroid to be unique, if possible. while np.all(X[ndx[0]] == centroids, axis=1).any() and ndx.shape[0] > 1: ndx = np.delete(ndx, 0) centroids[ik] = X[ndx[0]] return centroids def init_cao(X, n_clusters, dissim): """Initialize centroids according to method by Cao et al. [2009]. Note: O(N * attr * n_clusters**2), so watch out with large n_clusters """ n_points, n_attrs = X.shape centroids = np.empty((n_clusters, n_attrs), dtype='object') # Method is based on determining density of points. dens = np.zeros(n_points) for iattr in range(n_attrs): freq = defaultdict(int) for val in X[:, iattr]: freq[val] += 1 for ipoint in range(n_points): dens[ipoint] += freq[X[ipoint, iattr]] / float(n_points) / float(n_attrs) # Choose initial centroids based on distance and density. centroids[0] = X[np.argmax(dens)] if n_clusters > 1: # For the remaining centroids, choose maximum dens * dissim to the # (already assigned) centroid with the lowest dens * dissim. for ik in range(1, n_clusters): dd = np.empty((ik, n_points)) for ikk in range(ik): dd[ikk] = dissim(X, centroids[ikk]) * dens centroids[ik] = X[np.argmax(np.min(dd, axis=0))] return centroids def move_point_cat(point, ipoint, to_clust, from_clust, cl_attr_freq, membship, centroids): """Move point between clusters, categorical attributes.""" membship[to_clust, ipoint] = 1 membship[from_clust, ipoint] = 0 # Update frequencies of attributes in cluster. for iattr, curattr in enumerate(point): to_attr_counts = cl_attr_freq[to_clust][iattr] from_attr_counts = cl_attr_freq[from_clust][iattr] # Increment the attribute count for the new "to" cluster to_attr_counts[curattr] += 1 current_attribute_value_freq = to_attr_counts[curattr] current_centroid_value = centroids[to_clust][iattr] current_centroid_freq = to_attr_counts[current_centroid_value] if current_centroid_freq < current_attribute_value_freq: # We have incremented this value to the new mode. Update the centroid. centroids[to_clust][iattr] = curattr # Decrement the attribute count for the old "from" cluster from_attr_counts[curattr] -= 1 old_centroid_value = centroids[from_clust][iattr] if old_centroid_value == curattr: # We have just removed a count from the old centroid value. We need to # recalculate the centroid as it may no longer be the maximum centroids[from_clust][iattr] = get_max_value_key(from_attr_counts) return cl_attr_freq, membship, centroids def _labels_cost(X, centroids, dissim, membship=None): """Calculate labels and cost function given a matrix of points and a list of centroids for the k-modes algorithm. """ X = check_array(X) n_points = X.shape[0] cost = 0. labels = np.empty(n_points, dtype=np.uint16) for ipoint, curpoint in enumerate(X): diss = dissim(centroids, curpoint, X=X, membship=membship) clust = np.argmin(diss) labels[ipoint] = clust cost += diss[clust] return labels, cost def _k_modes_iter(X, centroids, cl_attr_freq, membship, dissim, random_state): """Single iteration of k-modes clustering algorithm""" moves = 0 for ipoint, curpoint in enumerate(X): clust = np.argmin(dissim(centroids, curpoint, X=X, membship=membship)) if membship[clust, ipoint]: # Point is already in its right place. continue # Move point, and update old/new cluster frequencies and centroids. moves += 1 old_clust = np.argwhere(membship[:, ipoint])[0][0] cl_attr_freq, membship, centroids = move_point_cat( curpoint, ipoint, clust, old_clust, cl_attr_freq, membship, centroids ) # In case of an empty cluster, reinitialize with a random point # from the largest cluster. if not membship[old_clust, :].any(): from_clust = membship.sum(axis=1).argmax() choices = [ii for ii, ch in enumerate(membship[from_clust, :]) if ch] rindx = random_state.choice(choices) cl_attr_freq, membship, centroids = move_point_cat( X[rindx], rindx, old_clust, from_clust, cl_attr_freq, membship, centroids ) return centroids, moves def k_modes_single(X, n_clusters, n_points, n_attrs, max_iter, dissim, init, init_no, verbose, random_state): random_state = check_random_state(random_state) # _____ INIT _____ if verbose: print("Init: initializing centroids") if isinstance(init, str) and init.lower() == 'huang': centroids = init_huang(X, n_clusters, dissim, random_state) elif isinstance(init, str) and init.lower() == 'cao': centroids = init_cao(X, n_clusters, dissim) elif isinstance(init, str) and init.lower() == 'random': seeds = random_state.choice(range(n_points), n_clusters) centroids = X[seeds] elif hasattr(init, '__array__'): # Make sure init is a 2D array. if len(init.shape) == 1: init = np.atleast_2d(init).T assert init.shape[0] == n_clusters, \ "Wrong number of initial centroids in init ({}, should be {})." \ .format(init.shape[0], n_clusters) assert init.shape[1] == n_attrs, \ "Wrong number of attributes in init ({}, should be {})." \ .format(init.shape[1], n_attrs) centroids = np.asarray(init, dtype=np.uint16) else: raise NotImplementedError if verbose: print("Init: initializing clusters") membship = np.zeros((n_clusters, n_points), dtype=np.uint8) # cl_attr_freq is a list of lists with dictionaries that contain the # frequencies of values per cluster and attribute. cl_attr_freq = [[defaultdict(int) for _ in range(n_attrs)] for _ in range(n_clusters)] for ipoint, curpoint in enumerate(X): # Initial assignment to clusters clust = np.argmin(dissim(centroids, curpoint, X=X, membship=membship)) membship[clust, ipoint] = 1 # Count attribute values per cluster. for iattr, curattr in enumerate(curpoint): cl_attr_freq[clust][iattr][curattr] += 1 # Perform an initial centroid update. for ik in range(n_clusters): for iattr in range(n_attrs): if sum(membship[ik]) == 0: # Empty centroid, choose randomly centroids[ik, iattr] = random_state.choice(X[:, iattr]) else: centroids[ik, iattr] = get_max_value_key(cl_attr_freq[ik][iattr]) # _____ ITERATION _____ if verbose: print("Starting iterations...") itr = 0 labels = None converged = False cost = np.Inf while itr <= max_iter and not converged: itr += 1 centroids, moves = _k_modes_iter( X, centroids, cl_attr_freq, membship, dissim, random_state ) # All points seen in this iteration labels, ncost = _labels_cost(X, centroids, dissim, membship) converged = (moves == 0) or (ncost >= cost) cost = ncost if verbose: print("Run {}, iteration: {}/{}, moves: {}, cost: {}" .format(init_no + 1, itr, max_iter, moves, cost)) return centroids, labels, cost, itr class KModes(BaseEstimator, ClusterMixin): """k-modes clustering algorithm for categorical data. Parameters ----------- n_clusters : int, optional, default: 8 The number of clusters to form as well as the number of centroids to generate. max_iter : int, default: 300 Maximum number of iterations of the k-modes algorithm for a single run. cat_dissim : func, default: matching_dissim Dissimilarity function used by the k-modes algorithm for categorical variables. Defaults to the matching dissimilarity function. init : {'Huang', 'Cao', 'random' or an ndarray}, default: 'Cao' Method for initialization: 'Huang': Method in Huang [1997, 1998] 'Cao': Method in Cao et al. [2009] 'random': choose 'n_clusters' observations (rows) at random from data for the initial centroids. If an ndarray is passed, it should be of shape (n_clusters, n_features) and gives the initial centroids. n_init : int, default: 10 Number of time the k-modes algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of cost. verbose : int, optional Verbosity mode. random_state : int, RandomState instance or None, optional, default: None If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. n_jobs : int, default: 1 The number of jobs to use for the computation. This works by computing each of the n_init runs in parallel. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Attributes ---------- cluster_centroids_ : array, [n_clusters, n_features] Categories of cluster centroids labels_ : Labels of each point cost_ : float Clustering cost, defined as the sum distance of all points to their respective cluster centroids. n_iter_ : int The number of iterations the algorithm ran for. Notes ----- See: Huang, Z.: Extensions to the k-modes algorithm for clustering large data sets with categorical values, Data Mining and Knowledge Discovery 2(3), 1998. """ def __init__(self, n_clusters=8, max_iter=100, cat_dissim=matching_dissim, init='Cao', n_init=1, verbose=0, random_state=None, n_jobs=1): self.n_clusters = n_clusters self.max_iter = max_iter self.cat_dissim = cat_dissim self.init = init self.n_init = n_init self.verbose = verbose self.random_state = random_state self.n_jobs = n_jobs if ((isinstance(self.init, str) and self.init == 'Cao') or hasattr(self.init, '__array__')) and self.n_init > 1: if self.verbose: print("Initialization method and algorithm are deterministic. " "Setting n_init to 1.") self.n_init = 1 def fit(self, X, y=None, **kwargs): """Compute k-modes clustering. Parameters ---------- X : array-like, shape=[n_samples, n_features] """ X = pandas_to_numpy(X) random_state = check_random_state(self.random_state) self._enc_cluster_centroids, self._enc_map, self.labels_,\ self.cost_, self.n_iter_ = k_modes(X, self.n_clusters, self.max_iter, self.cat_dissim, self.init, self.n_init, self.verbose, random_state, self.n_jobs) return self def fit_predict(self, X, y=None, **kwargs): """Compute cluster centroids and predict cluster index for each sample. Convenience method; equivalent to calling fit(X) followed by predict(X). """ return self.fit(X, **kwargs).predict(X, **kwargs) def predict(self, X, **kwargs): """Predict the closest cluster each sample in X belongs to. Parameters ---------- X : array-like, shape = [n_samples, n_features] New data to predict. Returns ------- labels : array, shape [n_samples,] Index of the cluster each sample belongs to. """ assert hasattr(self, '_enc_cluster_centroids'), "Model not yet fitted." if self.verbose and self.cat_dissim == ng_dissim: print("Ng's dissimilarity measure was used to train this model, " "but now that it is predicting the model will fall back to " "using simple matching dissimilarity.") X = pandas_to_numpy(X) X = check_array(X, dtype=None) X, _ = encode_features(X, enc_map=self._enc_map) return _labels_cost(X, self._enc_cluster_centroids, self.cat_dissim)[0] @property def cluster_centroids_(self): if hasattr(self, '_enc_cluster_centroids'): return decode_centroids(self._enc_cluster_centroids, self._enc_map) else: raise AttributeError("'{}' object has no attribute 'cluster_centroids_' " "because the model is not yet fitted.")
nicodv/kmodes
kmodes/kmodes.py
KModes.fit
python
def fit(self, X, y=None, **kwargs): X = pandas_to_numpy(X) random_state = check_random_state(self.random_state) self._enc_cluster_centroids, self._enc_map, self.labels_,\ self.cost_, self.n_iter_ = k_modes(X, self.n_clusters, self.max_iter, self.cat_dissim, self.init, self.n_init, self.verbose, random_state, self.n_jobs) return self
Compute k-modes clustering. Parameters ---------- X : array-like, shape=[n_samples, n_features]
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/kmodes.py#L381-L401
[ "def pandas_to_numpy(x):\n return x.values if 'pandas' in str(x.__class__) else x\n", "def k_modes(X, n_clusters, max_iter, dissim, init, n_init, verbose, random_state, n_jobs):\n \"\"\"k-modes algorithm\"\"\"\n random_state = check_random_state(random_state)\n if sparse.issparse(X):\n raise TypeError(\"k-modes does not support sparse data.\")\n\n X = check_array(X, dtype=None)\n\n # Convert the categorical values in X to integers for speed.\n # Based on the unique values in X, we can make a mapping to achieve this.\n X, enc_map = encode_features(X)\n\n n_points, n_attrs = X.shape\n assert n_clusters <= n_points, \"Cannot have more clusters ({}) \" \\\n \"than data points ({}).\".format(n_clusters, n_points)\n\n # Are there more n_clusters than unique rows? Then set the unique\n # rows as initial values and skip iteration.\n unique = get_unique_rows(X)\n n_unique = unique.shape[0]\n if n_unique <= n_clusters:\n max_iter = 0\n n_init = 1\n n_clusters = n_unique\n init = unique\n\n results = []\n seeds = random_state.randint(np.iinfo(np.int32).max, size=n_init)\n if n_jobs == 1:\n for init_no in range(n_init):\n results.append(k_modes_single(X, n_clusters, n_points, n_attrs, max_iter,\n dissim, init, init_no, verbose, seeds[init_no]))\n else:\n results = Parallel(n_jobs=n_jobs, verbose=0)(\n delayed(k_modes_single)(X, n_clusters, n_points, n_attrs, max_iter,\n dissim, init, init_no, verbose, seed)\n for init_no, seed in enumerate(seeds))\n all_centroids, all_labels, all_costs, all_n_iters = zip(*results)\n\n best = np.argmin(all_costs)\n if n_init > 1 and verbose:\n print(\"Best run was number {}\".format(best + 1))\n\n return all_centroids[best], enc_map, all_labels[best], \\\n all_costs[best], all_n_iters[best]\n" ]
class KModes(BaseEstimator, ClusterMixin): """k-modes clustering algorithm for categorical data. Parameters ----------- n_clusters : int, optional, default: 8 The number of clusters to form as well as the number of centroids to generate. max_iter : int, default: 300 Maximum number of iterations of the k-modes algorithm for a single run. cat_dissim : func, default: matching_dissim Dissimilarity function used by the k-modes algorithm for categorical variables. Defaults to the matching dissimilarity function. init : {'Huang', 'Cao', 'random' or an ndarray}, default: 'Cao' Method for initialization: 'Huang': Method in Huang [1997, 1998] 'Cao': Method in Cao et al. [2009] 'random': choose 'n_clusters' observations (rows) at random from data for the initial centroids. If an ndarray is passed, it should be of shape (n_clusters, n_features) and gives the initial centroids. n_init : int, default: 10 Number of time the k-modes algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of cost. verbose : int, optional Verbosity mode. random_state : int, RandomState instance or None, optional, default: None If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. n_jobs : int, default: 1 The number of jobs to use for the computation. This works by computing each of the n_init runs in parallel. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Attributes ---------- cluster_centroids_ : array, [n_clusters, n_features] Categories of cluster centroids labels_ : Labels of each point cost_ : float Clustering cost, defined as the sum distance of all points to their respective cluster centroids. n_iter_ : int The number of iterations the algorithm ran for. Notes ----- See: Huang, Z.: Extensions to the k-modes algorithm for clustering large data sets with categorical values, Data Mining and Knowledge Discovery 2(3), 1998. """ def __init__(self, n_clusters=8, max_iter=100, cat_dissim=matching_dissim, init='Cao', n_init=1, verbose=0, random_state=None, n_jobs=1): self.n_clusters = n_clusters self.max_iter = max_iter self.cat_dissim = cat_dissim self.init = init self.n_init = n_init self.verbose = verbose self.random_state = random_state self.n_jobs = n_jobs if ((isinstance(self.init, str) and self.init == 'Cao') or hasattr(self.init, '__array__')) and self.n_init > 1: if self.verbose: print("Initialization method and algorithm are deterministic. " "Setting n_init to 1.") self.n_init = 1 def fit_predict(self, X, y=None, **kwargs): """Compute cluster centroids and predict cluster index for each sample. Convenience method; equivalent to calling fit(X) followed by predict(X). """ return self.fit(X, **kwargs).predict(X, **kwargs) def predict(self, X, **kwargs): """Predict the closest cluster each sample in X belongs to. Parameters ---------- X : array-like, shape = [n_samples, n_features] New data to predict. Returns ------- labels : array, shape [n_samples,] Index of the cluster each sample belongs to. """ assert hasattr(self, '_enc_cluster_centroids'), "Model not yet fitted." if self.verbose and self.cat_dissim == ng_dissim: print("Ng's dissimilarity measure was used to train this model, " "but now that it is predicting the model will fall back to " "using simple matching dissimilarity.") X = pandas_to_numpy(X) X = check_array(X, dtype=None) X, _ = encode_features(X, enc_map=self._enc_map) return _labels_cost(X, self._enc_cluster_centroids, self.cat_dissim)[0] @property def cluster_centroids_(self): if hasattr(self, '_enc_cluster_centroids'): return decode_centroids(self._enc_cluster_centroids, self._enc_map) else: raise AttributeError("'{}' object has no attribute 'cluster_centroids_' " "because the model is not yet fitted.")
nicodv/kmodes
kmodes/kmodes.py
KModes.fit_predict
python
def fit_predict(self, X, y=None, **kwargs): return self.fit(X, **kwargs).predict(X, **kwargs)
Compute cluster centroids and predict cluster index for each sample. Convenience method; equivalent to calling fit(X) followed by predict(X).
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/kmodes.py#L403-L409
null
class KModes(BaseEstimator, ClusterMixin): """k-modes clustering algorithm for categorical data. Parameters ----------- n_clusters : int, optional, default: 8 The number of clusters to form as well as the number of centroids to generate. max_iter : int, default: 300 Maximum number of iterations of the k-modes algorithm for a single run. cat_dissim : func, default: matching_dissim Dissimilarity function used by the k-modes algorithm for categorical variables. Defaults to the matching dissimilarity function. init : {'Huang', 'Cao', 'random' or an ndarray}, default: 'Cao' Method for initialization: 'Huang': Method in Huang [1997, 1998] 'Cao': Method in Cao et al. [2009] 'random': choose 'n_clusters' observations (rows) at random from data for the initial centroids. If an ndarray is passed, it should be of shape (n_clusters, n_features) and gives the initial centroids. n_init : int, default: 10 Number of time the k-modes algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of cost. verbose : int, optional Verbosity mode. random_state : int, RandomState instance or None, optional, default: None If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. n_jobs : int, default: 1 The number of jobs to use for the computation. This works by computing each of the n_init runs in parallel. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Attributes ---------- cluster_centroids_ : array, [n_clusters, n_features] Categories of cluster centroids labels_ : Labels of each point cost_ : float Clustering cost, defined as the sum distance of all points to their respective cluster centroids. n_iter_ : int The number of iterations the algorithm ran for. Notes ----- See: Huang, Z.: Extensions to the k-modes algorithm for clustering large data sets with categorical values, Data Mining and Knowledge Discovery 2(3), 1998. """ def __init__(self, n_clusters=8, max_iter=100, cat_dissim=matching_dissim, init='Cao', n_init=1, verbose=0, random_state=None, n_jobs=1): self.n_clusters = n_clusters self.max_iter = max_iter self.cat_dissim = cat_dissim self.init = init self.n_init = n_init self.verbose = verbose self.random_state = random_state self.n_jobs = n_jobs if ((isinstance(self.init, str) and self.init == 'Cao') or hasattr(self.init, '__array__')) and self.n_init > 1: if self.verbose: print("Initialization method and algorithm are deterministic. " "Setting n_init to 1.") self.n_init = 1 def fit(self, X, y=None, **kwargs): """Compute k-modes clustering. Parameters ---------- X : array-like, shape=[n_samples, n_features] """ X = pandas_to_numpy(X) random_state = check_random_state(self.random_state) self._enc_cluster_centroids, self._enc_map, self.labels_,\ self.cost_, self.n_iter_ = k_modes(X, self.n_clusters, self.max_iter, self.cat_dissim, self.init, self.n_init, self.verbose, random_state, self.n_jobs) return self def predict(self, X, **kwargs): """Predict the closest cluster each sample in X belongs to. Parameters ---------- X : array-like, shape = [n_samples, n_features] New data to predict. Returns ------- labels : array, shape [n_samples,] Index of the cluster each sample belongs to. """ assert hasattr(self, '_enc_cluster_centroids'), "Model not yet fitted." if self.verbose and self.cat_dissim == ng_dissim: print("Ng's dissimilarity measure was used to train this model, " "but now that it is predicting the model will fall back to " "using simple matching dissimilarity.") X = pandas_to_numpy(X) X = check_array(X, dtype=None) X, _ = encode_features(X, enc_map=self._enc_map) return _labels_cost(X, self._enc_cluster_centroids, self.cat_dissim)[0] @property def cluster_centroids_(self): if hasattr(self, '_enc_cluster_centroids'): return decode_centroids(self._enc_cluster_centroids, self._enc_map) else: raise AttributeError("'{}' object has no attribute 'cluster_centroids_' " "because the model is not yet fitted.")
nicodv/kmodes
kmodes/kmodes.py
KModes.predict
python
def predict(self, X, **kwargs): assert hasattr(self, '_enc_cluster_centroids'), "Model not yet fitted." if self.verbose and self.cat_dissim == ng_dissim: print("Ng's dissimilarity measure was used to train this model, " "but now that it is predicting the model will fall back to " "using simple matching dissimilarity.") X = pandas_to_numpy(X) X = check_array(X, dtype=None) X, _ = encode_features(X, enc_map=self._enc_map) return _labels_cost(X, self._enc_cluster_centroids, self.cat_dissim)[0]
Predict the closest cluster each sample in X belongs to. Parameters ---------- X : array-like, shape = [n_samples, n_features] New data to predict. Returns ------- labels : array, shape [n_samples,] Index of the cluster each sample belongs to.
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/kmodes.py#L411-L435
[ "def pandas_to_numpy(x):\n return x.values if 'pandas' in str(x.__class__) else x\n", "def encode_features(X, enc_map=None):\n \"\"\"Converts categorical values in each column of X to integers in the range\n [0, n_unique_values_in_column - 1], if X is not already of integer type.\n\n If mapping is not provided, it is calculated based on the values in X.\n\n Unknown values during prediction get a value of -1. np.NaNs are ignored\n during encoding, and get treated as unknowns during prediction.\n \"\"\"\n if enc_map is None:\n fit = True\n # We will calculate enc_map, so initialize the list of column mappings.\n enc_map = []\n else:\n fit = False\n\n Xenc = np.zeros(X.shape).astype('int')\n for ii in range(X.shape[1]):\n if fit:\n col_enc = {val: jj for jj, val in enumerate(np.unique(X[:, ii]))\n if not (isinstance(val, float) and np.isnan(val))}\n enc_map.append(col_enc)\n # Unknown categories (including np.NaNs) all get a value of -1.\n Xenc[:, ii] = np.array([enc_map[ii].get(x, -1) for x in X[:, ii]])\n\n return Xenc, enc_map\n", "def _labels_cost(X, centroids, dissim, membship=None):\n \"\"\"Calculate labels and cost function given a matrix of points and\n a list of centroids for the k-modes algorithm.\n \"\"\"\n\n X = check_array(X)\n\n n_points = X.shape[0]\n cost = 0.\n labels = np.empty(n_points, dtype=np.uint16)\n for ipoint, curpoint in enumerate(X):\n diss = dissim(centroids, curpoint, X=X, membship=membship)\n clust = np.argmin(diss)\n labels[ipoint] = clust\n cost += diss[clust]\n\n return labels, cost\n" ]
class KModes(BaseEstimator, ClusterMixin): """k-modes clustering algorithm for categorical data. Parameters ----------- n_clusters : int, optional, default: 8 The number of clusters to form as well as the number of centroids to generate. max_iter : int, default: 300 Maximum number of iterations of the k-modes algorithm for a single run. cat_dissim : func, default: matching_dissim Dissimilarity function used by the k-modes algorithm for categorical variables. Defaults to the matching dissimilarity function. init : {'Huang', 'Cao', 'random' or an ndarray}, default: 'Cao' Method for initialization: 'Huang': Method in Huang [1997, 1998] 'Cao': Method in Cao et al. [2009] 'random': choose 'n_clusters' observations (rows) at random from data for the initial centroids. If an ndarray is passed, it should be of shape (n_clusters, n_features) and gives the initial centroids. n_init : int, default: 10 Number of time the k-modes algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of cost. verbose : int, optional Verbosity mode. random_state : int, RandomState instance or None, optional, default: None If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. n_jobs : int, default: 1 The number of jobs to use for the computation. This works by computing each of the n_init runs in parallel. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Attributes ---------- cluster_centroids_ : array, [n_clusters, n_features] Categories of cluster centroids labels_ : Labels of each point cost_ : float Clustering cost, defined as the sum distance of all points to their respective cluster centroids. n_iter_ : int The number of iterations the algorithm ran for. Notes ----- See: Huang, Z.: Extensions to the k-modes algorithm for clustering large data sets with categorical values, Data Mining and Knowledge Discovery 2(3), 1998. """ def __init__(self, n_clusters=8, max_iter=100, cat_dissim=matching_dissim, init='Cao', n_init=1, verbose=0, random_state=None, n_jobs=1): self.n_clusters = n_clusters self.max_iter = max_iter self.cat_dissim = cat_dissim self.init = init self.n_init = n_init self.verbose = verbose self.random_state = random_state self.n_jobs = n_jobs if ((isinstance(self.init, str) and self.init == 'Cao') or hasattr(self.init, '__array__')) and self.n_init > 1: if self.verbose: print("Initialization method and algorithm are deterministic. " "Setting n_init to 1.") self.n_init = 1 def fit(self, X, y=None, **kwargs): """Compute k-modes clustering. Parameters ---------- X : array-like, shape=[n_samples, n_features] """ X = pandas_to_numpy(X) random_state = check_random_state(self.random_state) self._enc_cluster_centroids, self._enc_map, self.labels_,\ self.cost_, self.n_iter_ = k_modes(X, self.n_clusters, self.max_iter, self.cat_dissim, self.init, self.n_init, self.verbose, random_state, self.n_jobs) return self def fit_predict(self, X, y=None, **kwargs): """Compute cluster centroids and predict cluster index for each sample. Convenience method; equivalent to calling fit(X) followed by predict(X). """ return self.fit(X, **kwargs).predict(X, **kwargs) @property def cluster_centroids_(self): if hasattr(self, '_enc_cluster_centroids'): return decode_centroids(self._enc_cluster_centroids, self._enc_map) else: raise AttributeError("'{}' object has no attribute 'cluster_centroids_' " "because the model is not yet fitted.")
nicodv/kmodes
kmodes/util/__init__.py
get_max_value_key
python
def get_max_value_key(dic): v = np.array(list(dic.values())) k = np.array(list(dic.keys())) maxima = np.where(v == np.max(v))[0] if len(maxima) == 1: return k[maxima[0]] else: # In order to be consistent, always selects the minimum key # (guaranteed to be unique) when there are multiple maximum values. return k[maxima[np.argmin(k[maxima])]]
Gets the key for the maximum value in a dict.
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/util/__init__.py#L12-L23
null
""" Generic utilities for clustering """ import numpy as np def pandas_to_numpy(x): return x.values if 'pandas' in str(x.__class__) else x def encode_features(X, enc_map=None): """Converts categorical values in each column of X to integers in the range [0, n_unique_values_in_column - 1], if X is not already of integer type. If mapping is not provided, it is calculated based on the values in X. Unknown values during prediction get a value of -1. np.NaNs are ignored during encoding, and get treated as unknowns during prediction. """ if enc_map is None: fit = True # We will calculate enc_map, so initialize the list of column mappings. enc_map = [] else: fit = False Xenc = np.zeros(X.shape).astype('int') for ii in range(X.shape[1]): if fit: col_enc = {val: jj for jj, val in enumerate(np.unique(X[:, ii])) if not (isinstance(val, float) and np.isnan(val))} enc_map.append(col_enc) # Unknown categories (including np.NaNs) all get a value of -1. Xenc[:, ii] = np.array([enc_map[ii].get(x, -1) for x in X[:, ii]]) return Xenc, enc_map def decode_centroids(encoded, mapping): """Decodes the encoded centroids array back to the original data labels using a list of mappings. """ decoded = [] for ii in range(encoded.shape[1]): # Invert the mapping so that we can decode. inv_mapping = {v: k for k, v in mapping[ii].items()} decoded.append(np.vectorize(inv_mapping.__getitem__)(encoded[:, ii])) return np.atleast_2d(np.array(decoded)).T def get_unique_rows(a): """Gets the unique rows in a numpy array.""" return np.vstack(list({tuple(row) for row in a}))
nicodv/kmodes
kmodes/util/__init__.py
encode_features
python
def encode_features(X, enc_map=None): if enc_map is None: fit = True # We will calculate enc_map, so initialize the list of column mappings. enc_map = [] else: fit = False Xenc = np.zeros(X.shape).astype('int') for ii in range(X.shape[1]): if fit: col_enc = {val: jj for jj, val in enumerate(np.unique(X[:, ii])) if not (isinstance(val, float) and np.isnan(val))} enc_map.append(col_enc) # Unknown categories (including np.NaNs) all get a value of -1. Xenc[:, ii] = np.array([enc_map[ii].get(x, -1) for x in X[:, ii]]) return Xenc, enc_map
Converts categorical values in each column of X to integers in the range [0, n_unique_values_in_column - 1], if X is not already of integer type. If mapping is not provided, it is calculated based on the values in X. Unknown values during prediction get a value of -1. np.NaNs are ignored during encoding, and get treated as unknowns during prediction.
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/util/__init__.py#L26-L51
null
""" Generic utilities for clustering """ import numpy as np def pandas_to_numpy(x): return x.values if 'pandas' in str(x.__class__) else x def get_max_value_key(dic): """Gets the key for the maximum value in a dict.""" v = np.array(list(dic.values())) k = np.array(list(dic.keys())) maxima = np.where(v == np.max(v))[0] if len(maxima) == 1: return k[maxima[0]] else: # In order to be consistent, always selects the minimum key # (guaranteed to be unique) when there are multiple maximum values. return k[maxima[np.argmin(k[maxima])]] def decode_centroids(encoded, mapping): """Decodes the encoded centroids array back to the original data labels using a list of mappings. """ decoded = [] for ii in range(encoded.shape[1]): # Invert the mapping so that we can decode. inv_mapping = {v: k for k, v in mapping[ii].items()} decoded.append(np.vectorize(inv_mapping.__getitem__)(encoded[:, ii])) return np.atleast_2d(np.array(decoded)).T def get_unique_rows(a): """Gets the unique rows in a numpy array.""" return np.vstack(list({tuple(row) for row in a}))
nicodv/kmodes
kmodes/util/__init__.py
decode_centroids
python
def decode_centroids(encoded, mapping): decoded = [] for ii in range(encoded.shape[1]): # Invert the mapping so that we can decode. inv_mapping = {v: k for k, v in mapping[ii].items()} decoded.append(np.vectorize(inv_mapping.__getitem__)(encoded[:, ii])) return np.atleast_2d(np.array(decoded)).T
Decodes the encoded centroids array back to the original data labels using a list of mappings.
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/util/__init__.py#L54-L63
null
""" Generic utilities for clustering """ import numpy as np def pandas_to_numpy(x): return x.values if 'pandas' in str(x.__class__) else x def get_max_value_key(dic): """Gets the key for the maximum value in a dict.""" v = np.array(list(dic.values())) k = np.array(list(dic.keys())) maxima = np.where(v == np.max(v))[0] if len(maxima) == 1: return k[maxima[0]] else: # In order to be consistent, always selects the minimum key # (guaranteed to be unique) when there are multiple maximum values. return k[maxima[np.argmin(k[maxima])]] def encode_features(X, enc_map=None): """Converts categorical values in each column of X to integers in the range [0, n_unique_values_in_column - 1], if X is not already of integer type. If mapping is not provided, it is calculated based on the values in X. Unknown values during prediction get a value of -1. np.NaNs are ignored during encoding, and get treated as unknowns during prediction. """ if enc_map is None: fit = True # We will calculate enc_map, so initialize the list of column mappings. enc_map = [] else: fit = False Xenc = np.zeros(X.shape).astype('int') for ii in range(X.shape[1]): if fit: col_enc = {val: jj for jj, val in enumerate(np.unique(X[:, ii])) if not (isinstance(val, float) and np.isnan(val))} enc_map.append(col_enc) # Unknown categories (including np.NaNs) all get a value of -1. Xenc[:, ii] = np.array([enc_map[ii].get(x, -1) for x in X[:, ii]]) return Xenc, enc_map def get_unique_rows(a): """Gets the unique rows in a numpy array.""" return np.vstack(list({tuple(row) for row in a}))
nicodv/kmodes
kmodes/kprototypes.py
move_point_num
python
def move_point_num(point, to_clust, from_clust, cl_attr_sum, cl_memb_sum): # Update sum of attributes in cluster. for iattr, curattr in enumerate(point): cl_attr_sum[to_clust][iattr] += curattr cl_attr_sum[from_clust][iattr] -= curattr # Update sums of memberships in cluster cl_memb_sum[to_clust] += 1 cl_memb_sum[from_clust] -= 1 return cl_attr_sum, cl_memb_sum
Move point between clusters, numerical attributes.
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/kprototypes.py#L28-L37
null
""" K-prototypes clustering for mixed categorical and numerical data """ # pylint: disable=super-on-old-class,unused-argument,attribute-defined-outside-init from collections import defaultdict import numpy as np from joblib import Parallel, delayed from scipy import sparse from sklearn.utils import check_random_state from sklearn.utils.validation import check_array from . import kmodes from .util import get_max_value_key, encode_features, get_unique_rows, \ decode_centroids, pandas_to_numpy from .util.dissim import matching_dissim, euclidean_dissim # Number of tries we give the initialization methods to find non-empty # clusters before we switch to random initialization. MAX_INIT_TRIES = 20 # Number of tries we give the initialization before we raise an # initialization error. RAISE_INIT_TRIES = 100 def _split_num_cat(X, categorical): """Extract numerical and categorical columns. Convert to numpy arrays, if needed. :param X: Feature matrix :param categorical: Indices of categorical columns """ Xnum = np.asanyarray(X[:, [ii for ii in range(X.shape[1]) if ii not in categorical]]).astype(np.float64) Xcat = np.asanyarray(X[:, categorical]) return Xnum, Xcat def _labels_cost(Xnum, Xcat, centroids, num_dissim, cat_dissim, gamma, membship=None): """Calculate labels and cost function given a matrix of points and a list of centroids for the k-prototypes algorithm. """ n_points = Xnum.shape[0] Xnum = check_array(Xnum) cost = 0. labels = np.empty(n_points, dtype=np.uint16) for ipoint in range(n_points): # Numerical cost = sum of Euclidean distances num_costs = num_dissim(centroids[0], Xnum[ipoint]) cat_costs = cat_dissim(centroids[1], Xcat[ipoint], X=Xcat, membship=membship) # Gamma relates the categorical cost to the numerical cost. tot_costs = num_costs + gamma * cat_costs clust = np.argmin(tot_costs) labels[ipoint] = clust cost += tot_costs[clust] return labels, cost def _k_prototypes_iter(Xnum, Xcat, centroids, cl_attr_sum, cl_memb_sum, cl_attr_freq, membship, num_dissim, cat_dissim, gamma, random_state): """Single iteration of the k-prototypes algorithm""" moves = 0 for ipoint in range(Xnum.shape[0]): clust = np.argmin( num_dissim(centroids[0], Xnum[ipoint]) + gamma * cat_dissim(centroids[1], Xcat[ipoint], X=Xcat, membship=membship) ) if membship[clust, ipoint]: # Point is already in its right place. continue # Move point, and update old/new cluster frequencies and centroids. moves += 1 old_clust = np.argwhere(membship[:, ipoint])[0][0] # Note that membship gets updated by kmodes.move_point_cat. # move_point_num only updates things specific to the k-means part. cl_attr_sum, cl_memb_sum = move_point_num( Xnum[ipoint], clust, old_clust, cl_attr_sum, cl_memb_sum ) cl_attr_freq, membship, centroids[1] = kmodes.move_point_cat( Xcat[ipoint], ipoint, clust, old_clust, cl_attr_freq, membship, centroids[1] ) # Update old and new centroids for numerical attributes using # the means and sums of all values for iattr in range(len(Xnum[ipoint])): for curc in (clust, old_clust): if cl_memb_sum[curc]: centroids[0][curc, iattr] = cl_attr_sum[curc, iattr] / cl_memb_sum[curc] else: centroids[0][curc, iattr] = 0. # In case of an empty cluster, reinitialize with a random point # from largest cluster. if not cl_memb_sum[old_clust]: from_clust = membship.sum(axis=1).argmax() choices = [ii for ii, ch in enumerate(membship[from_clust, :]) if ch] rindx = random_state.choice(choices) cl_attr_sum, cl_memb_sum = move_point_num( Xnum[rindx], old_clust, from_clust, cl_attr_sum, cl_memb_sum ) cl_attr_freq, membship, centroids[1] = kmodes.move_point_cat( Xcat[rindx], rindx, old_clust, from_clust, cl_attr_freq, membship, centroids[1] ) return centroids, moves def k_prototypes_single(Xnum, Xcat, nnumattrs, ncatattrs, n_clusters, n_points, max_iter, num_dissim, cat_dissim, gamma, init, init_no, verbose, random_state): # For numerical part of initialization, we don't have a guarantee # that there is not an empty cluster, so we need to retry until # there is none. random_state = check_random_state(random_state) init_tries = 0 while True: init_tries += 1 # _____ INIT _____ if verbose: print("Init: initializing centroids") if isinstance(init, str) and init.lower() == 'huang': centroids = kmodes.init_huang(Xcat, n_clusters, cat_dissim, random_state) elif isinstance(init, str) and init.lower() == 'cao': centroids = kmodes.init_cao(Xcat, n_clusters, cat_dissim) elif isinstance(init, str) and init.lower() == 'random': seeds = random_state.choice(range(n_points), n_clusters) centroids = Xcat[seeds] elif isinstance(init, list): # Make sure inits are 2D arrays. init = [np.atleast_2d(cur_init).T if len(cur_init.shape) == 1 else cur_init for cur_init in init] assert init[0].shape[0] == n_clusters, \ "Wrong number of initial numerical centroids in init " \ "({}, should be {}).".format(init[0].shape[0], n_clusters) assert init[0].shape[1] == nnumattrs, \ "Wrong number of numerical attributes in init ({}, should be {})." \ .format(init[0].shape[1], nnumattrs) assert init[1].shape[0] == n_clusters, \ "Wrong number of initial categorical centroids in init ({}, " \ "should be {}).".format(init[1].shape[0], n_clusters) assert init[1].shape[1] == ncatattrs, \ "Wrong number of categorical attributes in init ({}, should be {})." \ .format(init[1].shape[1], ncatattrs) centroids = [np.asarray(init[0], dtype=np.float64), np.asarray(init[1], dtype=np.uint16)] else: raise NotImplementedError("Initialization method not supported.") if not isinstance(init, list): # Numerical is initialized by drawing from normal distribution, # categorical following the k-modes methods. meanx = np.mean(Xnum, axis=0) stdx = np.std(Xnum, axis=0) centroids = [ meanx + random_state.randn(n_clusters, nnumattrs) * stdx, centroids ] if verbose: print("Init: initializing clusters") membship = np.zeros((n_clusters, n_points), dtype=np.uint8) # Keep track of the sum of attribute values per cluster so that we # can do k-means on the numerical attributes. cl_attr_sum = np.zeros((n_clusters, nnumattrs), dtype=np.float64) # Same for the membership sum per cluster cl_memb_sum = np.zeros(n_clusters, dtype=int) # cl_attr_freq is a list of lists with dictionaries that contain # the frequencies of values per cluster and attribute. cl_attr_freq = [[defaultdict(int) for _ in range(ncatattrs)] for _ in range(n_clusters)] for ipoint in range(n_points): # Initial assignment to clusters clust = np.argmin( num_dissim(centroids[0], Xnum[ipoint]) + gamma * cat_dissim(centroids[1], Xcat[ipoint], X=Xcat, membship=membship) ) membship[clust, ipoint] = 1 cl_memb_sum[clust] += 1 # Count attribute values per cluster. for iattr, curattr in enumerate(Xnum[ipoint]): cl_attr_sum[clust, iattr] += curattr for iattr, curattr in enumerate(Xcat[ipoint]): cl_attr_freq[clust][iattr][curattr] += 1 # If no empty clusters, then consider initialization finalized. if membship.sum(axis=1).min() > 0: break if init_tries == MAX_INIT_TRIES: # Could not get rid of empty clusters. Randomly # initialize instead. init = 'random' elif init_tries == RAISE_INIT_TRIES: raise ValueError( "Clustering algorithm could not initialize. " "Consider assigning the initial clusters manually." ) # Perform an initial centroid update. for ik in range(n_clusters): for iattr in range(nnumattrs): centroids[0][ik, iattr] = cl_attr_sum[ik, iattr] / cl_memb_sum[ik] for iattr in range(ncatattrs): centroids[1][ik, iattr] = get_max_value_key(cl_attr_freq[ik][iattr]) # _____ ITERATION _____ if verbose: print("Starting iterations...") itr = 0 labels = None converged = False cost = np.Inf while itr <= max_iter and not converged: itr += 1 centroids, moves = _k_prototypes_iter(Xnum, Xcat, centroids, cl_attr_sum, cl_memb_sum, cl_attr_freq, membship, num_dissim, cat_dissim, gamma, random_state) # All points seen in this iteration labels, ncost = _labels_cost(Xnum, Xcat, centroids, num_dissim, cat_dissim, gamma, membship) converged = (moves == 0) or (ncost >= cost) cost = ncost if verbose: print("Run: {}, iteration: {}/{}, moves: {}, ncost: {}" .format(init_no + 1, itr, max_iter, moves, ncost)) return centroids, labels, cost, itr def k_prototypes(X, categorical, n_clusters, max_iter, num_dissim, cat_dissim, gamma, init, n_init, verbose, random_state, n_jobs): """k-prototypes algorithm""" random_state = check_random_state(random_state) if sparse.issparse(X): raise TypeError("k-prototypes does not support sparse data.") if categorical is None or not categorical: raise NotImplementedError( "No categorical data selected, effectively doing k-means. " "Present a list of categorical columns, or use scikit-learn's " "KMeans instead." ) if isinstance(categorical, int): categorical = [categorical] assert len(categorical) != X.shape[1], \ "All columns are categorical, use k-modes instead of k-prototypes." assert max(categorical) < X.shape[1], \ "Categorical index larger than number of columns." ncatattrs = len(categorical) nnumattrs = X.shape[1] - ncatattrs n_points = X.shape[0] assert n_clusters <= n_points, "Cannot have more clusters ({}) " \ "than data points ({}).".format(n_clusters, n_points) Xnum, Xcat = _split_num_cat(X, categorical) Xnum, Xcat = check_array(Xnum), check_array(Xcat, dtype=None) # Convert the categorical values in Xcat to integers for speed. # Based on the unique values in Xcat, we can make a mapping to achieve this. Xcat, enc_map = encode_features(Xcat) # Are there more n_clusters than unique rows? Then set the unique # rows as initial values and skip iteration. unique = get_unique_rows(X) n_unique = unique.shape[0] if n_unique <= n_clusters: max_iter = 0 n_init = 1 n_clusters = n_unique init = list(_split_num_cat(unique, categorical)) init[1], _ = encode_features(init[1], enc_map) # Estimate a good value for gamma, which determines the weighing of # categorical values in clusters (see Huang [1997]). if gamma is None: gamma = 0.5 * Xnum.std() results = [] seeds = random_state.randint(np.iinfo(np.int32).max, size=n_init) if n_jobs == 1: for init_no in range(n_init): results.append(k_prototypes_single(Xnum, Xcat, nnumattrs, ncatattrs, n_clusters, n_points, max_iter, num_dissim, cat_dissim, gamma, init, init_no, verbose, seeds[init_no])) else: results = Parallel(n_jobs=n_jobs, verbose=0)( delayed(k_prototypes_single)(Xnum, Xcat, nnumattrs, ncatattrs, n_clusters, n_points, max_iter, num_dissim, cat_dissim, gamma, init, init_no, verbose, seed) for init_no, seed in enumerate(seeds)) all_centroids, all_labels, all_costs, all_n_iters = zip(*results) best = np.argmin(all_costs) if n_init > 1 and verbose: print("Best run was number {}".format(best + 1)) # Note: return gamma in case it was automatically determined. return all_centroids[best], enc_map, all_labels[best], \ all_costs[best], all_n_iters[best], gamma class KPrototypes(kmodes.KModes): """k-protoypes clustering algorithm for mixed numerical/categorical data. Parameters ----------- n_clusters : int, optional, default: 8 The number of clusters to form as well as the number of centroids to generate. max_iter : int, default: 300 Maximum number of iterations of the k-modes algorithm for a single run. num_dissim : func, default: euclidian_dissim Dissimilarity function used by the algorithm for numerical variables. Defaults to the Euclidian dissimilarity function. cat_dissim : func, default: matching_dissim Dissimilarity function used by the kmodes algorithm for categorical variables. Defaults to the matching dissimilarity function. n_init : int, default: 10 Number of time the k-modes algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of cost. init : {'Huang', 'Cao', 'random' or a list of ndarrays}, default: 'Cao' Method for initialization: 'Huang': Method in Huang [1997, 1998] 'Cao': Method in Cao et al. [2009] 'random': choose 'n_clusters' observations (rows) at random from data for the initial centroids. If a list of ndarrays is passed, it should be of length 2, with shapes (n_clusters, n_features) for numerical and categorical data respectively. These are the initial centroids. gamma : float, default: None Weighing factor that determines relative importance of numerical vs. categorical attributes (see discussion in Huang [1997]). By default, automatically calculated from data. verbose : integer, optional Verbosity mode. random_state : int, RandomState instance or None, optional, default: None If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. n_jobs : int, default: 1 The number of jobs to use for the computation. This works by computing each of the n_init runs in parallel. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Attributes ---------- cluster_centroids_ : array, [n_clusters, n_features] Categories of cluster centroids labels_ : Labels of each point cost_ : float Clustering cost, defined as the sum distance of all points to their respective cluster centroids. n_iter_ : int The number of iterations the algorithm ran for. gamma : float The (potentially calculated) weighing factor. Notes ----- See: Huang, Z.: Extensions to the k-modes algorithm for clustering large data sets with categorical values, Data Mining and Knowledge Discovery 2(3), 1998. """ def __init__(self, n_clusters=8, max_iter=100, num_dissim=euclidean_dissim, cat_dissim=matching_dissim, init='Huang', n_init=10, gamma=None, verbose=0, random_state=None, n_jobs=1): super(KPrototypes, self).__init__(n_clusters, max_iter, cat_dissim, init, verbose=verbose, random_state=random_state, n_jobs=n_jobs) self.num_dissim = num_dissim self.gamma = gamma self.n_init = n_init if isinstance(self.init, list) and self.n_init > 1: if self.verbose: print("Initialization method is deterministic. " "Setting n_init to 1.") self.n_init = 1 def fit(self, X, y=None, categorical=None): """Compute k-prototypes clustering. Parameters ---------- X : array-like, shape=[n_samples, n_features] categorical : Index of columns that contain categorical data """ if categorical is not None: assert isinstance(categorical, (int, list, tuple)), "The 'categorical' \ argument needs to be an integer with the index of the categorical \ column in your data, or a list or tuple of several of them, \ but it is a {}.".format(type(categorical)) X = pandas_to_numpy(X) random_state = check_random_state(self.random_state) # If self.gamma is None, gamma will be automatically determined from # the data. The function below returns its value. self._enc_cluster_centroids, self._enc_map, self.labels_, self.cost_,\ self.n_iter_, self.gamma = k_prototypes(X, categorical, self.n_clusters, self.max_iter, self.num_dissim, self.cat_dissim, self.gamma, self.init, self.n_init, self.verbose, random_state, self.n_jobs) return self def predict(self, X, categorical=None): """Predict the closest cluster each sample in X belongs to. Parameters ---------- X : array-like, shape = [n_samples, n_features] New data to predict. categorical : Indices of columns that contain categorical data Returns ------- labels : array, shape [n_samples,] Index of the cluster each sample belongs to. """ assert hasattr(self, '_enc_cluster_centroids'), "Model not yet fitted." if categorical is not None: assert isinstance(categorical, (int, list, tuple)), "The 'categorical' \ argument needs to be an integer with the index of the categorical \ column in your data, or a list or tuple of several of them, \ but it is a {}.".format(type(categorical)) X = pandas_to_numpy(X) Xnum, Xcat = _split_num_cat(X, categorical) Xnum, Xcat = check_array(Xnum), check_array(Xcat, dtype=None) Xcat, _ = encode_features(Xcat, enc_map=self._enc_map) return _labels_cost(Xnum, Xcat, self._enc_cluster_centroids, self.num_dissim, self.cat_dissim, self.gamma)[0] @property def cluster_centroids_(self): if hasattr(self, '_enc_cluster_centroids'): return [ self._enc_cluster_centroids[0], decode_centroids(self._enc_cluster_centroids[1], self._enc_map) ] else: raise AttributeError("'{}' object has no attribute 'cluster_centroids_' " "because the model is not yet fitted.")
nicodv/kmodes
kmodes/kprototypes.py
_split_num_cat
python
def _split_num_cat(X, categorical): Xnum = np.asanyarray(X[:, [ii for ii in range(X.shape[1]) if ii not in categorical]]).astype(np.float64) Xcat = np.asanyarray(X[:, categorical]) return Xnum, Xcat
Extract numerical and categorical columns. Convert to numpy arrays, if needed. :param X: Feature matrix :param categorical: Indices of categorical columns
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/kprototypes.py#L40-L50
null
""" K-prototypes clustering for mixed categorical and numerical data """ # pylint: disable=super-on-old-class,unused-argument,attribute-defined-outside-init from collections import defaultdict import numpy as np from joblib import Parallel, delayed from scipy import sparse from sklearn.utils import check_random_state from sklearn.utils.validation import check_array from . import kmodes from .util import get_max_value_key, encode_features, get_unique_rows, \ decode_centroids, pandas_to_numpy from .util.dissim import matching_dissim, euclidean_dissim # Number of tries we give the initialization methods to find non-empty # clusters before we switch to random initialization. MAX_INIT_TRIES = 20 # Number of tries we give the initialization before we raise an # initialization error. RAISE_INIT_TRIES = 100 def move_point_num(point, to_clust, from_clust, cl_attr_sum, cl_memb_sum): """Move point between clusters, numerical attributes.""" # Update sum of attributes in cluster. for iattr, curattr in enumerate(point): cl_attr_sum[to_clust][iattr] += curattr cl_attr_sum[from_clust][iattr] -= curattr # Update sums of memberships in cluster cl_memb_sum[to_clust] += 1 cl_memb_sum[from_clust] -= 1 return cl_attr_sum, cl_memb_sum def _labels_cost(Xnum, Xcat, centroids, num_dissim, cat_dissim, gamma, membship=None): """Calculate labels and cost function given a matrix of points and a list of centroids for the k-prototypes algorithm. """ n_points = Xnum.shape[0] Xnum = check_array(Xnum) cost = 0. labels = np.empty(n_points, dtype=np.uint16) for ipoint in range(n_points): # Numerical cost = sum of Euclidean distances num_costs = num_dissim(centroids[0], Xnum[ipoint]) cat_costs = cat_dissim(centroids[1], Xcat[ipoint], X=Xcat, membship=membship) # Gamma relates the categorical cost to the numerical cost. tot_costs = num_costs + gamma * cat_costs clust = np.argmin(tot_costs) labels[ipoint] = clust cost += tot_costs[clust] return labels, cost def _k_prototypes_iter(Xnum, Xcat, centroids, cl_attr_sum, cl_memb_sum, cl_attr_freq, membship, num_dissim, cat_dissim, gamma, random_state): """Single iteration of the k-prototypes algorithm""" moves = 0 for ipoint in range(Xnum.shape[0]): clust = np.argmin( num_dissim(centroids[0], Xnum[ipoint]) + gamma * cat_dissim(centroids[1], Xcat[ipoint], X=Xcat, membship=membship) ) if membship[clust, ipoint]: # Point is already in its right place. continue # Move point, and update old/new cluster frequencies and centroids. moves += 1 old_clust = np.argwhere(membship[:, ipoint])[0][0] # Note that membship gets updated by kmodes.move_point_cat. # move_point_num only updates things specific to the k-means part. cl_attr_sum, cl_memb_sum = move_point_num( Xnum[ipoint], clust, old_clust, cl_attr_sum, cl_memb_sum ) cl_attr_freq, membship, centroids[1] = kmodes.move_point_cat( Xcat[ipoint], ipoint, clust, old_clust, cl_attr_freq, membship, centroids[1] ) # Update old and new centroids for numerical attributes using # the means and sums of all values for iattr in range(len(Xnum[ipoint])): for curc in (clust, old_clust): if cl_memb_sum[curc]: centroids[0][curc, iattr] = cl_attr_sum[curc, iattr] / cl_memb_sum[curc] else: centroids[0][curc, iattr] = 0. # In case of an empty cluster, reinitialize with a random point # from largest cluster. if not cl_memb_sum[old_clust]: from_clust = membship.sum(axis=1).argmax() choices = [ii for ii, ch in enumerate(membship[from_clust, :]) if ch] rindx = random_state.choice(choices) cl_attr_sum, cl_memb_sum = move_point_num( Xnum[rindx], old_clust, from_clust, cl_attr_sum, cl_memb_sum ) cl_attr_freq, membship, centroids[1] = kmodes.move_point_cat( Xcat[rindx], rindx, old_clust, from_clust, cl_attr_freq, membship, centroids[1] ) return centroids, moves def k_prototypes_single(Xnum, Xcat, nnumattrs, ncatattrs, n_clusters, n_points, max_iter, num_dissim, cat_dissim, gamma, init, init_no, verbose, random_state): # For numerical part of initialization, we don't have a guarantee # that there is not an empty cluster, so we need to retry until # there is none. random_state = check_random_state(random_state) init_tries = 0 while True: init_tries += 1 # _____ INIT _____ if verbose: print("Init: initializing centroids") if isinstance(init, str) and init.lower() == 'huang': centroids = kmodes.init_huang(Xcat, n_clusters, cat_dissim, random_state) elif isinstance(init, str) and init.lower() == 'cao': centroids = kmodes.init_cao(Xcat, n_clusters, cat_dissim) elif isinstance(init, str) and init.lower() == 'random': seeds = random_state.choice(range(n_points), n_clusters) centroids = Xcat[seeds] elif isinstance(init, list): # Make sure inits are 2D arrays. init = [np.atleast_2d(cur_init).T if len(cur_init.shape) == 1 else cur_init for cur_init in init] assert init[0].shape[0] == n_clusters, \ "Wrong number of initial numerical centroids in init " \ "({}, should be {}).".format(init[0].shape[0], n_clusters) assert init[0].shape[1] == nnumattrs, \ "Wrong number of numerical attributes in init ({}, should be {})." \ .format(init[0].shape[1], nnumattrs) assert init[1].shape[0] == n_clusters, \ "Wrong number of initial categorical centroids in init ({}, " \ "should be {}).".format(init[1].shape[0], n_clusters) assert init[1].shape[1] == ncatattrs, \ "Wrong number of categorical attributes in init ({}, should be {})." \ .format(init[1].shape[1], ncatattrs) centroids = [np.asarray(init[0], dtype=np.float64), np.asarray(init[1], dtype=np.uint16)] else: raise NotImplementedError("Initialization method not supported.") if not isinstance(init, list): # Numerical is initialized by drawing from normal distribution, # categorical following the k-modes methods. meanx = np.mean(Xnum, axis=0) stdx = np.std(Xnum, axis=0) centroids = [ meanx + random_state.randn(n_clusters, nnumattrs) * stdx, centroids ] if verbose: print("Init: initializing clusters") membship = np.zeros((n_clusters, n_points), dtype=np.uint8) # Keep track of the sum of attribute values per cluster so that we # can do k-means on the numerical attributes. cl_attr_sum = np.zeros((n_clusters, nnumattrs), dtype=np.float64) # Same for the membership sum per cluster cl_memb_sum = np.zeros(n_clusters, dtype=int) # cl_attr_freq is a list of lists with dictionaries that contain # the frequencies of values per cluster and attribute. cl_attr_freq = [[defaultdict(int) for _ in range(ncatattrs)] for _ in range(n_clusters)] for ipoint in range(n_points): # Initial assignment to clusters clust = np.argmin( num_dissim(centroids[0], Xnum[ipoint]) + gamma * cat_dissim(centroids[1], Xcat[ipoint], X=Xcat, membship=membship) ) membship[clust, ipoint] = 1 cl_memb_sum[clust] += 1 # Count attribute values per cluster. for iattr, curattr in enumerate(Xnum[ipoint]): cl_attr_sum[clust, iattr] += curattr for iattr, curattr in enumerate(Xcat[ipoint]): cl_attr_freq[clust][iattr][curattr] += 1 # If no empty clusters, then consider initialization finalized. if membship.sum(axis=1).min() > 0: break if init_tries == MAX_INIT_TRIES: # Could not get rid of empty clusters. Randomly # initialize instead. init = 'random' elif init_tries == RAISE_INIT_TRIES: raise ValueError( "Clustering algorithm could not initialize. " "Consider assigning the initial clusters manually." ) # Perform an initial centroid update. for ik in range(n_clusters): for iattr in range(nnumattrs): centroids[0][ik, iattr] = cl_attr_sum[ik, iattr] / cl_memb_sum[ik] for iattr in range(ncatattrs): centroids[1][ik, iattr] = get_max_value_key(cl_attr_freq[ik][iattr]) # _____ ITERATION _____ if verbose: print("Starting iterations...") itr = 0 labels = None converged = False cost = np.Inf while itr <= max_iter and not converged: itr += 1 centroids, moves = _k_prototypes_iter(Xnum, Xcat, centroids, cl_attr_sum, cl_memb_sum, cl_attr_freq, membship, num_dissim, cat_dissim, gamma, random_state) # All points seen in this iteration labels, ncost = _labels_cost(Xnum, Xcat, centroids, num_dissim, cat_dissim, gamma, membship) converged = (moves == 0) or (ncost >= cost) cost = ncost if verbose: print("Run: {}, iteration: {}/{}, moves: {}, ncost: {}" .format(init_no + 1, itr, max_iter, moves, ncost)) return centroids, labels, cost, itr def k_prototypes(X, categorical, n_clusters, max_iter, num_dissim, cat_dissim, gamma, init, n_init, verbose, random_state, n_jobs): """k-prototypes algorithm""" random_state = check_random_state(random_state) if sparse.issparse(X): raise TypeError("k-prototypes does not support sparse data.") if categorical is None or not categorical: raise NotImplementedError( "No categorical data selected, effectively doing k-means. " "Present a list of categorical columns, or use scikit-learn's " "KMeans instead." ) if isinstance(categorical, int): categorical = [categorical] assert len(categorical) != X.shape[1], \ "All columns are categorical, use k-modes instead of k-prototypes." assert max(categorical) < X.shape[1], \ "Categorical index larger than number of columns." ncatattrs = len(categorical) nnumattrs = X.shape[1] - ncatattrs n_points = X.shape[0] assert n_clusters <= n_points, "Cannot have more clusters ({}) " \ "than data points ({}).".format(n_clusters, n_points) Xnum, Xcat = _split_num_cat(X, categorical) Xnum, Xcat = check_array(Xnum), check_array(Xcat, dtype=None) # Convert the categorical values in Xcat to integers for speed. # Based on the unique values in Xcat, we can make a mapping to achieve this. Xcat, enc_map = encode_features(Xcat) # Are there more n_clusters than unique rows? Then set the unique # rows as initial values and skip iteration. unique = get_unique_rows(X) n_unique = unique.shape[0] if n_unique <= n_clusters: max_iter = 0 n_init = 1 n_clusters = n_unique init = list(_split_num_cat(unique, categorical)) init[1], _ = encode_features(init[1], enc_map) # Estimate a good value for gamma, which determines the weighing of # categorical values in clusters (see Huang [1997]). if gamma is None: gamma = 0.5 * Xnum.std() results = [] seeds = random_state.randint(np.iinfo(np.int32).max, size=n_init) if n_jobs == 1: for init_no in range(n_init): results.append(k_prototypes_single(Xnum, Xcat, nnumattrs, ncatattrs, n_clusters, n_points, max_iter, num_dissim, cat_dissim, gamma, init, init_no, verbose, seeds[init_no])) else: results = Parallel(n_jobs=n_jobs, verbose=0)( delayed(k_prototypes_single)(Xnum, Xcat, nnumattrs, ncatattrs, n_clusters, n_points, max_iter, num_dissim, cat_dissim, gamma, init, init_no, verbose, seed) for init_no, seed in enumerate(seeds)) all_centroids, all_labels, all_costs, all_n_iters = zip(*results) best = np.argmin(all_costs) if n_init > 1 and verbose: print("Best run was number {}".format(best + 1)) # Note: return gamma in case it was automatically determined. return all_centroids[best], enc_map, all_labels[best], \ all_costs[best], all_n_iters[best], gamma class KPrototypes(kmodes.KModes): """k-protoypes clustering algorithm for mixed numerical/categorical data. Parameters ----------- n_clusters : int, optional, default: 8 The number of clusters to form as well as the number of centroids to generate. max_iter : int, default: 300 Maximum number of iterations of the k-modes algorithm for a single run. num_dissim : func, default: euclidian_dissim Dissimilarity function used by the algorithm for numerical variables. Defaults to the Euclidian dissimilarity function. cat_dissim : func, default: matching_dissim Dissimilarity function used by the kmodes algorithm for categorical variables. Defaults to the matching dissimilarity function. n_init : int, default: 10 Number of time the k-modes algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of cost. init : {'Huang', 'Cao', 'random' or a list of ndarrays}, default: 'Cao' Method for initialization: 'Huang': Method in Huang [1997, 1998] 'Cao': Method in Cao et al. [2009] 'random': choose 'n_clusters' observations (rows) at random from data for the initial centroids. If a list of ndarrays is passed, it should be of length 2, with shapes (n_clusters, n_features) for numerical and categorical data respectively. These are the initial centroids. gamma : float, default: None Weighing factor that determines relative importance of numerical vs. categorical attributes (see discussion in Huang [1997]). By default, automatically calculated from data. verbose : integer, optional Verbosity mode. random_state : int, RandomState instance or None, optional, default: None If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. n_jobs : int, default: 1 The number of jobs to use for the computation. This works by computing each of the n_init runs in parallel. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Attributes ---------- cluster_centroids_ : array, [n_clusters, n_features] Categories of cluster centroids labels_ : Labels of each point cost_ : float Clustering cost, defined as the sum distance of all points to their respective cluster centroids. n_iter_ : int The number of iterations the algorithm ran for. gamma : float The (potentially calculated) weighing factor. Notes ----- See: Huang, Z.: Extensions to the k-modes algorithm for clustering large data sets with categorical values, Data Mining and Knowledge Discovery 2(3), 1998. """ def __init__(self, n_clusters=8, max_iter=100, num_dissim=euclidean_dissim, cat_dissim=matching_dissim, init='Huang', n_init=10, gamma=None, verbose=0, random_state=None, n_jobs=1): super(KPrototypes, self).__init__(n_clusters, max_iter, cat_dissim, init, verbose=verbose, random_state=random_state, n_jobs=n_jobs) self.num_dissim = num_dissim self.gamma = gamma self.n_init = n_init if isinstance(self.init, list) and self.n_init > 1: if self.verbose: print("Initialization method is deterministic. " "Setting n_init to 1.") self.n_init = 1 def fit(self, X, y=None, categorical=None): """Compute k-prototypes clustering. Parameters ---------- X : array-like, shape=[n_samples, n_features] categorical : Index of columns that contain categorical data """ if categorical is not None: assert isinstance(categorical, (int, list, tuple)), "The 'categorical' \ argument needs to be an integer with the index of the categorical \ column in your data, or a list or tuple of several of them, \ but it is a {}.".format(type(categorical)) X = pandas_to_numpy(X) random_state = check_random_state(self.random_state) # If self.gamma is None, gamma will be automatically determined from # the data. The function below returns its value. self._enc_cluster_centroids, self._enc_map, self.labels_, self.cost_,\ self.n_iter_, self.gamma = k_prototypes(X, categorical, self.n_clusters, self.max_iter, self.num_dissim, self.cat_dissim, self.gamma, self.init, self.n_init, self.verbose, random_state, self.n_jobs) return self def predict(self, X, categorical=None): """Predict the closest cluster each sample in X belongs to. Parameters ---------- X : array-like, shape = [n_samples, n_features] New data to predict. categorical : Indices of columns that contain categorical data Returns ------- labels : array, shape [n_samples,] Index of the cluster each sample belongs to. """ assert hasattr(self, '_enc_cluster_centroids'), "Model not yet fitted." if categorical is not None: assert isinstance(categorical, (int, list, tuple)), "The 'categorical' \ argument needs to be an integer with the index of the categorical \ column in your data, or a list or tuple of several of them, \ but it is a {}.".format(type(categorical)) X = pandas_to_numpy(X) Xnum, Xcat = _split_num_cat(X, categorical) Xnum, Xcat = check_array(Xnum), check_array(Xcat, dtype=None) Xcat, _ = encode_features(Xcat, enc_map=self._enc_map) return _labels_cost(Xnum, Xcat, self._enc_cluster_centroids, self.num_dissim, self.cat_dissim, self.gamma)[0] @property def cluster_centroids_(self): if hasattr(self, '_enc_cluster_centroids'): return [ self._enc_cluster_centroids[0], decode_centroids(self._enc_cluster_centroids[1], self._enc_map) ] else: raise AttributeError("'{}' object has no attribute 'cluster_centroids_' " "because the model is not yet fitted.")
nicodv/kmodes
kmodes/kprototypes.py
_labels_cost
python
def _labels_cost(Xnum, Xcat, centroids, num_dissim, cat_dissim, gamma, membship=None): n_points = Xnum.shape[0] Xnum = check_array(Xnum) cost = 0. labels = np.empty(n_points, dtype=np.uint16) for ipoint in range(n_points): # Numerical cost = sum of Euclidean distances num_costs = num_dissim(centroids[0], Xnum[ipoint]) cat_costs = cat_dissim(centroids[1], Xcat[ipoint], X=Xcat, membship=membship) # Gamma relates the categorical cost to the numerical cost. tot_costs = num_costs + gamma * cat_costs clust = np.argmin(tot_costs) labels[ipoint] = clust cost += tot_costs[clust] return labels, cost
Calculate labels and cost function given a matrix of points and a list of centroids for the k-prototypes algorithm.
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/kprototypes.py#L53-L73
null
""" K-prototypes clustering for mixed categorical and numerical data """ # pylint: disable=super-on-old-class,unused-argument,attribute-defined-outside-init from collections import defaultdict import numpy as np from joblib import Parallel, delayed from scipy import sparse from sklearn.utils import check_random_state from sklearn.utils.validation import check_array from . import kmodes from .util import get_max_value_key, encode_features, get_unique_rows, \ decode_centroids, pandas_to_numpy from .util.dissim import matching_dissim, euclidean_dissim # Number of tries we give the initialization methods to find non-empty # clusters before we switch to random initialization. MAX_INIT_TRIES = 20 # Number of tries we give the initialization before we raise an # initialization error. RAISE_INIT_TRIES = 100 def move_point_num(point, to_clust, from_clust, cl_attr_sum, cl_memb_sum): """Move point between clusters, numerical attributes.""" # Update sum of attributes in cluster. for iattr, curattr in enumerate(point): cl_attr_sum[to_clust][iattr] += curattr cl_attr_sum[from_clust][iattr] -= curattr # Update sums of memberships in cluster cl_memb_sum[to_clust] += 1 cl_memb_sum[from_clust] -= 1 return cl_attr_sum, cl_memb_sum def _split_num_cat(X, categorical): """Extract numerical and categorical columns. Convert to numpy arrays, if needed. :param X: Feature matrix :param categorical: Indices of categorical columns """ Xnum = np.asanyarray(X[:, [ii for ii in range(X.shape[1]) if ii not in categorical]]).astype(np.float64) Xcat = np.asanyarray(X[:, categorical]) return Xnum, Xcat def _k_prototypes_iter(Xnum, Xcat, centroids, cl_attr_sum, cl_memb_sum, cl_attr_freq, membship, num_dissim, cat_dissim, gamma, random_state): """Single iteration of the k-prototypes algorithm""" moves = 0 for ipoint in range(Xnum.shape[0]): clust = np.argmin( num_dissim(centroids[0], Xnum[ipoint]) + gamma * cat_dissim(centroids[1], Xcat[ipoint], X=Xcat, membship=membship) ) if membship[clust, ipoint]: # Point is already in its right place. continue # Move point, and update old/new cluster frequencies and centroids. moves += 1 old_clust = np.argwhere(membship[:, ipoint])[0][0] # Note that membship gets updated by kmodes.move_point_cat. # move_point_num only updates things specific to the k-means part. cl_attr_sum, cl_memb_sum = move_point_num( Xnum[ipoint], clust, old_clust, cl_attr_sum, cl_memb_sum ) cl_attr_freq, membship, centroids[1] = kmodes.move_point_cat( Xcat[ipoint], ipoint, clust, old_clust, cl_attr_freq, membship, centroids[1] ) # Update old and new centroids for numerical attributes using # the means and sums of all values for iattr in range(len(Xnum[ipoint])): for curc in (clust, old_clust): if cl_memb_sum[curc]: centroids[0][curc, iattr] = cl_attr_sum[curc, iattr] / cl_memb_sum[curc] else: centroids[0][curc, iattr] = 0. # In case of an empty cluster, reinitialize with a random point # from largest cluster. if not cl_memb_sum[old_clust]: from_clust = membship.sum(axis=1).argmax() choices = [ii for ii, ch in enumerate(membship[from_clust, :]) if ch] rindx = random_state.choice(choices) cl_attr_sum, cl_memb_sum = move_point_num( Xnum[rindx], old_clust, from_clust, cl_attr_sum, cl_memb_sum ) cl_attr_freq, membship, centroids[1] = kmodes.move_point_cat( Xcat[rindx], rindx, old_clust, from_clust, cl_attr_freq, membship, centroids[1] ) return centroids, moves def k_prototypes_single(Xnum, Xcat, nnumattrs, ncatattrs, n_clusters, n_points, max_iter, num_dissim, cat_dissim, gamma, init, init_no, verbose, random_state): # For numerical part of initialization, we don't have a guarantee # that there is not an empty cluster, so we need to retry until # there is none. random_state = check_random_state(random_state) init_tries = 0 while True: init_tries += 1 # _____ INIT _____ if verbose: print("Init: initializing centroids") if isinstance(init, str) and init.lower() == 'huang': centroids = kmodes.init_huang(Xcat, n_clusters, cat_dissim, random_state) elif isinstance(init, str) and init.lower() == 'cao': centroids = kmodes.init_cao(Xcat, n_clusters, cat_dissim) elif isinstance(init, str) and init.lower() == 'random': seeds = random_state.choice(range(n_points), n_clusters) centroids = Xcat[seeds] elif isinstance(init, list): # Make sure inits are 2D arrays. init = [np.atleast_2d(cur_init).T if len(cur_init.shape) == 1 else cur_init for cur_init in init] assert init[0].shape[0] == n_clusters, \ "Wrong number of initial numerical centroids in init " \ "({}, should be {}).".format(init[0].shape[0], n_clusters) assert init[0].shape[1] == nnumattrs, \ "Wrong number of numerical attributes in init ({}, should be {})." \ .format(init[0].shape[1], nnumattrs) assert init[1].shape[0] == n_clusters, \ "Wrong number of initial categorical centroids in init ({}, " \ "should be {}).".format(init[1].shape[0], n_clusters) assert init[1].shape[1] == ncatattrs, \ "Wrong number of categorical attributes in init ({}, should be {})." \ .format(init[1].shape[1], ncatattrs) centroids = [np.asarray(init[0], dtype=np.float64), np.asarray(init[1], dtype=np.uint16)] else: raise NotImplementedError("Initialization method not supported.") if not isinstance(init, list): # Numerical is initialized by drawing from normal distribution, # categorical following the k-modes methods. meanx = np.mean(Xnum, axis=0) stdx = np.std(Xnum, axis=0) centroids = [ meanx + random_state.randn(n_clusters, nnumattrs) * stdx, centroids ] if verbose: print("Init: initializing clusters") membship = np.zeros((n_clusters, n_points), dtype=np.uint8) # Keep track of the sum of attribute values per cluster so that we # can do k-means on the numerical attributes. cl_attr_sum = np.zeros((n_clusters, nnumattrs), dtype=np.float64) # Same for the membership sum per cluster cl_memb_sum = np.zeros(n_clusters, dtype=int) # cl_attr_freq is a list of lists with dictionaries that contain # the frequencies of values per cluster and attribute. cl_attr_freq = [[defaultdict(int) for _ in range(ncatattrs)] for _ in range(n_clusters)] for ipoint in range(n_points): # Initial assignment to clusters clust = np.argmin( num_dissim(centroids[0], Xnum[ipoint]) + gamma * cat_dissim(centroids[1], Xcat[ipoint], X=Xcat, membship=membship) ) membship[clust, ipoint] = 1 cl_memb_sum[clust] += 1 # Count attribute values per cluster. for iattr, curattr in enumerate(Xnum[ipoint]): cl_attr_sum[clust, iattr] += curattr for iattr, curattr in enumerate(Xcat[ipoint]): cl_attr_freq[clust][iattr][curattr] += 1 # If no empty clusters, then consider initialization finalized. if membship.sum(axis=1).min() > 0: break if init_tries == MAX_INIT_TRIES: # Could not get rid of empty clusters. Randomly # initialize instead. init = 'random' elif init_tries == RAISE_INIT_TRIES: raise ValueError( "Clustering algorithm could not initialize. " "Consider assigning the initial clusters manually." ) # Perform an initial centroid update. for ik in range(n_clusters): for iattr in range(nnumattrs): centroids[0][ik, iattr] = cl_attr_sum[ik, iattr] / cl_memb_sum[ik] for iattr in range(ncatattrs): centroids[1][ik, iattr] = get_max_value_key(cl_attr_freq[ik][iattr]) # _____ ITERATION _____ if verbose: print("Starting iterations...") itr = 0 labels = None converged = False cost = np.Inf while itr <= max_iter and not converged: itr += 1 centroids, moves = _k_prototypes_iter(Xnum, Xcat, centroids, cl_attr_sum, cl_memb_sum, cl_attr_freq, membship, num_dissim, cat_dissim, gamma, random_state) # All points seen in this iteration labels, ncost = _labels_cost(Xnum, Xcat, centroids, num_dissim, cat_dissim, gamma, membship) converged = (moves == 0) or (ncost >= cost) cost = ncost if verbose: print("Run: {}, iteration: {}/{}, moves: {}, ncost: {}" .format(init_no + 1, itr, max_iter, moves, ncost)) return centroids, labels, cost, itr def k_prototypes(X, categorical, n_clusters, max_iter, num_dissim, cat_dissim, gamma, init, n_init, verbose, random_state, n_jobs): """k-prototypes algorithm""" random_state = check_random_state(random_state) if sparse.issparse(X): raise TypeError("k-prototypes does not support sparse data.") if categorical is None or not categorical: raise NotImplementedError( "No categorical data selected, effectively doing k-means. " "Present a list of categorical columns, or use scikit-learn's " "KMeans instead." ) if isinstance(categorical, int): categorical = [categorical] assert len(categorical) != X.shape[1], \ "All columns are categorical, use k-modes instead of k-prototypes." assert max(categorical) < X.shape[1], \ "Categorical index larger than number of columns." ncatattrs = len(categorical) nnumattrs = X.shape[1] - ncatattrs n_points = X.shape[0] assert n_clusters <= n_points, "Cannot have more clusters ({}) " \ "than data points ({}).".format(n_clusters, n_points) Xnum, Xcat = _split_num_cat(X, categorical) Xnum, Xcat = check_array(Xnum), check_array(Xcat, dtype=None) # Convert the categorical values in Xcat to integers for speed. # Based on the unique values in Xcat, we can make a mapping to achieve this. Xcat, enc_map = encode_features(Xcat) # Are there more n_clusters than unique rows? Then set the unique # rows as initial values and skip iteration. unique = get_unique_rows(X) n_unique = unique.shape[0] if n_unique <= n_clusters: max_iter = 0 n_init = 1 n_clusters = n_unique init = list(_split_num_cat(unique, categorical)) init[1], _ = encode_features(init[1], enc_map) # Estimate a good value for gamma, which determines the weighing of # categorical values in clusters (see Huang [1997]). if gamma is None: gamma = 0.5 * Xnum.std() results = [] seeds = random_state.randint(np.iinfo(np.int32).max, size=n_init) if n_jobs == 1: for init_no in range(n_init): results.append(k_prototypes_single(Xnum, Xcat, nnumattrs, ncatattrs, n_clusters, n_points, max_iter, num_dissim, cat_dissim, gamma, init, init_no, verbose, seeds[init_no])) else: results = Parallel(n_jobs=n_jobs, verbose=0)( delayed(k_prototypes_single)(Xnum, Xcat, nnumattrs, ncatattrs, n_clusters, n_points, max_iter, num_dissim, cat_dissim, gamma, init, init_no, verbose, seed) for init_no, seed in enumerate(seeds)) all_centroids, all_labels, all_costs, all_n_iters = zip(*results) best = np.argmin(all_costs) if n_init > 1 and verbose: print("Best run was number {}".format(best + 1)) # Note: return gamma in case it was automatically determined. return all_centroids[best], enc_map, all_labels[best], \ all_costs[best], all_n_iters[best], gamma class KPrototypes(kmodes.KModes): """k-protoypes clustering algorithm for mixed numerical/categorical data. Parameters ----------- n_clusters : int, optional, default: 8 The number of clusters to form as well as the number of centroids to generate. max_iter : int, default: 300 Maximum number of iterations of the k-modes algorithm for a single run. num_dissim : func, default: euclidian_dissim Dissimilarity function used by the algorithm for numerical variables. Defaults to the Euclidian dissimilarity function. cat_dissim : func, default: matching_dissim Dissimilarity function used by the kmodes algorithm for categorical variables. Defaults to the matching dissimilarity function. n_init : int, default: 10 Number of time the k-modes algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of cost. init : {'Huang', 'Cao', 'random' or a list of ndarrays}, default: 'Cao' Method for initialization: 'Huang': Method in Huang [1997, 1998] 'Cao': Method in Cao et al. [2009] 'random': choose 'n_clusters' observations (rows) at random from data for the initial centroids. If a list of ndarrays is passed, it should be of length 2, with shapes (n_clusters, n_features) for numerical and categorical data respectively. These are the initial centroids. gamma : float, default: None Weighing factor that determines relative importance of numerical vs. categorical attributes (see discussion in Huang [1997]). By default, automatically calculated from data. verbose : integer, optional Verbosity mode. random_state : int, RandomState instance or None, optional, default: None If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. n_jobs : int, default: 1 The number of jobs to use for the computation. This works by computing each of the n_init runs in parallel. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Attributes ---------- cluster_centroids_ : array, [n_clusters, n_features] Categories of cluster centroids labels_ : Labels of each point cost_ : float Clustering cost, defined as the sum distance of all points to their respective cluster centroids. n_iter_ : int The number of iterations the algorithm ran for. gamma : float The (potentially calculated) weighing factor. Notes ----- See: Huang, Z.: Extensions to the k-modes algorithm for clustering large data sets with categorical values, Data Mining and Knowledge Discovery 2(3), 1998. """ def __init__(self, n_clusters=8, max_iter=100, num_dissim=euclidean_dissim, cat_dissim=matching_dissim, init='Huang', n_init=10, gamma=None, verbose=0, random_state=None, n_jobs=1): super(KPrototypes, self).__init__(n_clusters, max_iter, cat_dissim, init, verbose=verbose, random_state=random_state, n_jobs=n_jobs) self.num_dissim = num_dissim self.gamma = gamma self.n_init = n_init if isinstance(self.init, list) and self.n_init > 1: if self.verbose: print("Initialization method is deterministic. " "Setting n_init to 1.") self.n_init = 1 def fit(self, X, y=None, categorical=None): """Compute k-prototypes clustering. Parameters ---------- X : array-like, shape=[n_samples, n_features] categorical : Index of columns that contain categorical data """ if categorical is not None: assert isinstance(categorical, (int, list, tuple)), "The 'categorical' \ argument needs to be an integer with the index of the categorical \ column in your data, or a list or tuple of several of them, \ but it is a {}.".format(type(categorical)) X = pandas_to_numpy(X) random_state = check_random_state(self.random_state) # If self.gamma is None, gamma will be automatically determined from # the data. The function below returns its value. self._enc_cluster_centroids, self._enc_map, self.labels_, self.cost_,\ self.n_iter_, self.gamma = k_prototypes(X, categorical, self.n_clusters, self.max_iter, self.num_dissim, self.cat_dissim, self.gamma, self.init, self.n_init, self.verbose, random_state, self.n_jobs) return self def predict(self, X, categorical=None): """Predict the closest cluster each sample in X belongs to. Parameters ---------- X : array-like, shape = [n_samples, n_features] New data to predict. categorical : Indices of columns that contain categorical data Returns ------- labels : array, shape [n_samples,] Index of the cluster each sample belongs to. """ assert hasattr(self, '_enc_cluster_centroids'), "Model not yet fitted." if categorical is not None: assert isinstance(categorical, (int, list, tuple)), "The 'categorical' \ argument needs to be an integer with the index of the categorical \ column in your data, or a list or tuple of several of them, \ but it is a {}.".format(type(categorical)) X = pandas_to_numpy(X) Xnum, Xcat = _split_num_cat(X, categorical) Xnum, Xcat = check_array(Xnum), check_array(Xcat, dtype=None) Xcat, _ = encode_features(Xcat, enc_map=self._enc_map) return _labels_cost(Xnum, Xcat, self._enc_cluster_centroids, self.num_dissim, self.cat_dissim, self.gamma)[0] @property def cluster_centroids_(self): if hasattr(self, '_enc_cluster_centroids'): return [ self._enc_cluster_centroids[0], decode_centroids(self._enc_cluster_centroids[1], self._enc_map) ] else: raise AttributeError("'{}' object has no attribute 'cluster_centroids_' " "because the model is not yet fitted.")
nicodv/kmodes
kmodes/kprototypes.py
_k_prototypes_iter
python
def _k_prototypes_iter(Xnum, Xcat, centroids, cl_attr_sum, cl_memb_sum, cl_attr_freq, membship, num_dissim, cat_dissim, gamma, random_state): moves = 0 for ipoint in range(Xnum.shape[0]): clust = np.argmin( num_dissim(centroids[0], Xnum[ipoint]) + gamma * cat_dissim(centroids[1], Xcat[ipoint], X=Xcat, membship=membship) ) if membship[clust, ipoint]: # Point is already in its right place. continue # Move point, and update old/new cluster frequencies and centroids. moves += 1 old_clust = np.argwhere(membship[:, ipoint])[0][0] # Note that membship gets updated by kmodes.move_point_cat. # move_point_num only updates things specific to the k-means part. cl_attr_sum, cl_memb_sum = move_point_num( Xnum[ipoint], clust, old_clust, cl_attr_sum, cl_memb_sum ) cl_attr_freq, membship, centroids[1] = kmodes.move_point_cat( Xcat[ipoint], ipoint, clust, old_clust, cl_attr_freq, membship, centroids[1] ) # Update old and new centroids for numerical attributes using # the means and sums of all values for iattr in range(len(Xnum[ipoint])): for curc in (clust, old_clust): if cl_memb_sum[curc]: centroids[0][curc, iattr] = cl_attr_sum[curc, iattr] / cl_memb_sum[curc] else: centroids[0][curc, iattr] = 0. # In case of an empty cluster, reinitialize with a random point # from largest cluster. if not cl_memb_sum[old_clust]: from_clust = membship.sum(axis=1).argmax() choices = [ii for ii, ch in enumerate(membship[from_clust, :]) if ch] rindx = random_state.choice(choices) cl_attr_sum, cl_memb_sum = move_point_num( Xnum[rindx], old_clust, from_clust, cl_attr_sum, cl_memb_sum ) cl_attr_freq, membship, centroids[1] = kmodes.move_point_cat( Xcat[rindx], rindx, old_clust, from_clust, cl_attr_freq, membship, centroids[1] ) return centroids, moves
Single iteration of the k-prototypes algorithm
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/kprototypes.py#L76-L127
null
""" K-prototypes clustering for mixed categorical and numerical data """ # pylint: disable=super-on-old-class,unused-argument,attribute-defined-outside-init from collections import defaultdict import numpy as np from joblib import Parallel, delayed from scipy import sparse from sklearn.utils import check_random_state from sklearn.utils.validation import check_array from . import kmodes from .util import get_max_value_key, encode_features, get_unique_rows, \ decode_centroids, pandas_to_numpy from .util.dissim import matching_dissim, euclidean_dissim # Number of tries we give the initialization methods to find non-empty # clusters before we switch to random initialization. MAX_INIT_TRIES = 20 # Number of tries we give the initialization before we raise an # initialization error. RAISE_INIT_TRIES = 100 def move_point_num(point, to_clust, from_clust, cl_attr_sum, cl_memb_sum): """Move point between clusters, numerical attributes.""" # Update sum of attributes in cluster. for iattr, curattr in enumerate(point): cl_attr_sum[to_clust][iattr] += curattr cl_attr_sum[from_clust][iattr] -= curattr # Update sums of memberships in cluster cl_memb_sum[to_clust] += 1 cl_memb_sum[from_clust] -= 1 return cl_attr_sum, cl_memb_sum def _split_num_cat(X, categorical): """Extract numerical and categorical columns. Convert to numpy arrays, if needed. :param X: Feature matrix :param categorical: Indices of categorical columns """ Xnum = np.asanyarray(X[:, [ii for ii in range(X.shape[1]) if ii not in categorical]]).astype(np.float64) Xcat = np.asanyarray(X[:, categorical]) return Xnum, Xcat def _labels_cost(Xnum, Xcat, centroids, num_dissim, cat_dissim, gamma, membship=None): """Calculate labels and cost function given a matrix of points and a list of centroids for the k-prototypes algorithm. """ n_points = Xnum.shape[0] Xnum = check_array(Xnum) cost = 0. labels = np.empty(n_points, dtype=np.uint16) for ipoint in range(n_points): # Numerical cost = sum of Euclidean distances num_costs = num_dissim(centroids[0], Xnum[ipoint]) cat_costs = cat_dissim(centroids[1], Xcat[ipoint], X=Xcat, membship=membship) # Gamma relates the categorical cost to the numerical cost. tot_costs = num_costs + gamma * cat_costs clust = np.argmin(tot_costs) labels[ipoint] = clust cost += tot_costs[clust] return labels, cost def k_prototypes_single(Xnum, Xcat, nnumattrs, ncatattrs, n_clusters, n_points, max_iter, num_dissim, cat_dissim, gamma, init, init_no, verbose, random_state): # For numerical part of initialization, we don't have a guarantee # that there is not an empty cluster, so we need to retry until # there is none. random_state = check_random_state(random_state) init_tries = 0 while True: init_tries += 1 # _____ INIT _____ if verbose: print("Init: initializing centroids") if isinstance(init, str) and init.lower() == 'huang': centroids = kmodes.init_huang(Xcat, n_clusters, cat_dissim, random_state) elif isinstance(init, str) and init.lower() == 'cao': centroids = kmodes.init_cao(Xcat, n_clusters, cat_dissim) elif isinstance(init, str) and init.lower() == 'random': seeds = random_state.choice(range(n_points), n_clusters) centroids = Xcat[seeds] elif isinstance(init, list): # Make sure inits are 2D arrays. init = [np.atleast_2d(cur_init).T if len(cur_init.shape) == 1 else cur_init for cur_init in init] assert init[0].shape[0] == n_clusters, \ "Wrong number of initial numerical centroids in init " \ "({}, should be {}).".format(init[0].shape[0], n_clusters) assert init[0].shape[1] == nnumattrs, \ "Wrong number of numerical attributes in init ({}, should be {})." \ .format(init[0].shape[1], nnumattrs) assert init[1].shape[0] == n_clusters, \ "Wrong number of initial categorical centroids in init ({}, " \ "should be {}).".format(init[1].shape[0], n_clusters) assert init[1].shape[1] == ncatattrs, \ "Wrong number of categorical attributes in init ({}, should be {})." \ .format(init[1].shape[1], ncatattrs) centroids = [np.asarray(init[0], dtype=np.float64), np.asarray(init[1], dtype=np.uint16)] else: raise NotImplementedError("Initialization method not supported.") if not isinstance(init, list): # Numerical is initialized by drawing from normal distribution, # categorical following the k-modes methods. meanx = np.mean(Xnum, axis=0) stdx = np.std(Xnum, axis=0) centroids = [ meanx + random_state.randn(n_clusters, nnumattrs) * stdx, centroids ] if verbose: print("Init: initializing clusters") membship = np.zeros((n_clusters, n_points), dtype=np.uint8) # Keep track of the sum of attribute values per cluster so that we # can do k-means on the numerical attributes. cl_attr_sum = np.zeros((n_clusters, nnumattrs), dtype=np.float64) # Same for the membership sum per cluster cl_memb_sum = np.zeros(n_clusters, dtype=int) # cl_attr_freq is a list of lists with dictionaries that contain # the frequencies of values per cluster and attribute. cl_attr_freq = [[defaultdict(int) for _ in range(ncatattrs)] for _ in range(n_clusters)] for ipoint in range(n_points): # Initial assignment to clusters clust = np.argmin( num_dissim(centroids[0], Xnum[ipoint]) + gamma * cat_dissim(centroids[1], Xcat[ipoint], X=Xcat, membship=membship) ) membship[clust, ipoint] = 1 cl_memb_sum[clust] += 1 # Count attribute values per cluster. for iattr, curattr in enumerate(Xnum[ipoint]): cl_attr_sum[clust, iattr] += curattr for iattr, curattr in enumerate(Xcat[ipoint]): cl_attr_freq[clust][iattr][curattr] += 1 # If no empty clusters, then consider initialization finalized. if membship.sum(axis=1).min() > 0: break if init_tries == MAX_INIT_TRIES: # Could not get rid of empty clusters. Randomly # initialize instead. init = 'random' elif init_tries == RAISE_INIT_TRIES: raise ValueError( "Clustering algorithm could not initialize. " "Consider assigning the initial clusters manually." ) # Perform an initial centroid update. for ik in range(n_clusters): for iattr in range(nnumattrs): centroids[0][ik, iattr] = cl_attr_sum[ik, iattr] / cl_memb_sum[ik] for iattr in range(ncatattrs): centroids[1][ik, iattr] = get_max_value_key(cl_attr_freq[ik][iattr]) # _____ ITERATION _____ if verbose: print("Starting iterations...") itr = 0 labels = None converged = False cost = np.Inf while itr <= max_iter and not converged: itr += 1 centroids, moves = _k_prototypes_iter(Xnum, Xcat, centroids, cl_attr_sum, cl_memb_sum, cl_attr_freq, membship, num_dissim, cat_dissim, gamma, random_state) # All points seen in this iteration labels, ncost = _labels_cost(Xnum, Xcat, centroids, num_dissim, cat_dissim, gamma, membship) converged = (moves == 0) or (ncost >= cost) cost = ncost if verbose: print("Run: {}, iteration: {}/{}, moves: {}, ncost: {}" .format(init_no + 1, itr, max_iter, moves, ncost)) return centroids, labels, cost, itr def k_prototypes(X, categorical, n_clusters, max_iter, num_dissim, cat_dissim, gamma, init, n_init, verbose, random_state, n_jobs): """k-prototypes algorithm""" random_state = check_random_state(random_state) if sparse.issparse(X): raise TypeError("k-prototypes does not support sparse data.") if categorical is None or not categorical: raise NotImplementedError( "No categorical data selected, effectively doing k-means. " "Present a list of categorical columns, or use scikit-learn's " "KMeans instead." ) if isinstance(categorical, int): categorical = [categorical] assert len(categorical) != X.shape[1], \ "All columns are categorical, use k-modes instead of k-prototypes." assert max(categorical) < X.shape[1], \ "Categorical index larger than number of columns." ncatattrs = len(categorical) nnumattrs = X.shape[1] - ncatattrs n_points = X.shape[0] assert n_clusters <= n_points, "Cannot have more clusters ({}) " \ "than data points ({}).".format(n_clusters, n_points) Xnum, Xcat = _split_num_cat(X, categorical) Xnum, Xcat = check_array(Xnum), check_array(Xcat, dtype=None) # Convert the categorical values in Xcat to integers for speed. # Based on the unique values in Xcat, we can make a mapping to achieve this. Xcat, enc_map = encode_features(Xcat) # Are there more n_clusters than unique rows? Then set the unique # rows as initial values and skip iteration. unique = get_unique_rows(X) n_unique = unique.shape[0] if n_unique <= n_clusters: max_iter = 0 n_init = 1 n_clusters = n_unique init = list(_split_num_cat(unique, categorical)) init[1], _ = encode_features(init[1], enc_map) # Estimate a good value for gamma, which determines the weighing of # categorical values in clusters (see Huang [1997]). if gamma is None: gamma = 0.5 * Xnum.std() results = [] seeds = random_state.randint(np.iinfo(np.int32).max, size=n_init) if n_jobs == 1: for init_no in range(n_init): results.append(k_prototypes_single(Xnum, Xcat, nnumattrs, ncatattrs, n_clusters, n_points, max_iter, num_dissim, cat_dissim, gamma, init, init_no, verbose, seeds[init_no])) else: results = Parallel(n_jobs=n_jobs, verbose=0)( delayed(k_prototypes_single)(Xnum, Xcat, nnumattrs, ncatattrs, n_clusters, n_points, max_iter, num_dissim, cat_dissim, gamma, init, init_no, verbose, seed) for init_no, seed in enumerate(seeds)) all_centroids, all_labels, all_costs, all_n_iters = zip(*results) best = np.argmin(all_costs) if n_init > 1 and verbose: print("Best run was number {}".format(best + 1)) # Note: return gamma in case it was automatically determined. return all_centroids[best], enc_map, all_labels[best], \ all_costs[best], all_n_iters[best], gamma class KPrototypes(kmodes.KModes): """k-protoypes clustering algorithm for mixed numerical/categorical data. Parameters ----------- n_clusters : int, optional, default: 8 The number of clusters to form as well as the number of centroids to generate. max_iter : int, default: 300 Maximum number of iterations of the k-modes algorithm for a single run. num_dissim : func, default: euclidian_dissim Dissimilarity function used by the algorithm for numerical variables. Defaults to the Euclidian dissimilarity function. cat_dissim : func, default: matching_dissim Dissimilarity function used by the kmodes algorithm for categorical variables. Defaults to the matching dissimilarity function. n_init : int, default: 10 Number of time the k-modes algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of cost. init : {'Huang', 'Cao', 'random' or a list of ndarrays}, default: 'Cao' Method for initialization: 'Huang': Method in Huang [1997, 1998] 'Cao': Method in Cao et al. [2009] 'random': choose 'n_clusters' observations (rows) at random from data for the initial centroids. If a list of ndarrays is passed, it should be of length 2, with shapes (n_clusters, n_features) for numerical and categorical data respectively. These are the initial centroids. gamma : float, default: None Weighing factor that determines relative importance of numerical vs. categorical attributes (see discussion in Huang [1997]). By default, automatically calculated from data. verbose : integer, optional Verbosity mode. random_state : int, RandomState instance or None, optional, default: None If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. n_jobs : int, default: 1 The number of jobs to use for the computation. This works by computing each of the n_init runs in parallel. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Attributes ---------- cluster_centroids_ : array, [n_clusters, n_features] Categories of cluster centroids labels_ : Labels of each point cost_ : float Clustering cost, defined as the sum distance of all points to their respective cluster centroids. n_iter_ : int The number of iterations the algorithm ran for. gamma : float The (potentially calculated) weighing factor. Notes ----- See: Huang, Z.: Extensions to the k-modes algorithm for clustering large data sets with categorical values, Data Mining and Knowledge Discovery 2(3), 1998. """ def __init__(self, n_clusters=8, max_iter=100, num_dissim=euclidean_dissim, cat_dissim=matching_dissim, init='Huang', n_init=10, gamma=None, verbose=0, random_state=None, n_jobs=1): super(KPrototypes, self).__init__(n_clusters, max_iter, cat_dissim, init, verbose=verbose, random_state=random_state, n_jobs=n_jobs) self.num_dissim = num_dissim self.gamma = gamma self.n_init = n_init if isinstance(self.init, list) and self.n_init > 1: if self.verbose: print("Initialization method is deterministic. " "Setting n_init to 1.") self.n_init = 1 def fit(self, X, y=None, categorical=None): """Compute k-prototypes clustering. Parameters ---------- X : array-like, shape=[n_samples, n_features] categorical : Index of columns that contain categorical data """ if categorical is not None: assert isinstance(categorical, (int, list, tuple)), "The 'categorical' \ argument needs to be an integer with the index of the categorical \ column in your data, or a list or tuple of several of them, \ but it is a {}.".format(type(categorical)) X = pandas_to_numpy(X) random_state = check_random_state(self.random_state) # If self.gamma is None, gamma will be automatically determined from # the data. The function below returns its value. self._enc_cluster_centroids, self._enc_map, self.labels_, self.cost_,\ self.n_iter_, self.gamma = k_prototypes(X, categorical, self.n_clusters, self.max_iter, self.num_dissim, self.cat_dissim, self.gamma, self.init, self.n_init, self.verbose, random_state, self.n_jobs) return self def predict(self, X, categorical=None): """Predict the closest cluster each sample in X belongs to. Parameters ---------- X : array-like, shape = [n_samples, n_features] New data to predict. categorical : Indices of columns that contain categorical data Returns ------- labels : array, shape [n_samples,] Index of the cluster each sample belongs to. """ assert hasattr(self, '_enc_cluster_centroids'), "Model not yet fitted." if categorical is not None: assert isinstance(categorical, (int, list, tuple)), "The 'categorical' \ argument needs to be an integer with the index of the categorical \ column in your data, or a list or tuple of several of them, \ but it is a {}.".format(type(categorical)) X = pandas_to_numpy(X) Xnum, Xcat = _split_num_cat(X, categorical) Xnum, Xcat = check_array(Xnum), check_array(Xcat, dtype=None) Xcat, _ = encode_features(Xcat, enc_map=self._enc_map) return _labels_cost(Xnum, Xcat, self._enc_cluster_centroids, self.num_dissim, self.cat_dissim, self.gamma)[0] @property def cluster_centroids_(self): if hasattr(self, '_enc_cluster_centroids'): return [ self._enc_cluster_centroids[0], decode_centroids(self._enc_cluster_centroids[1], self._enc_map) ] else: raise AttributeError("'{}' object has no attribute 'cluster_centroids_' " "because the model is not yet fitted.")
nicodv/kmodes
kmodes/kprototypes.py
k_prototypes
python
def k_prototypes(X, categorical, n_clusters, max_iter, num_dissim, cat_dissim, gamma, init, n_init, verbose, random_state, n_jobs): random_state = check_random_state(random_state) if sparse.issparse(X): raise TypeError("k-prototypes does not support sparse data.") if categorical is None or not categorical: raise NotImplementedError( "No categorical data selected, effectively doing k-means. " "Present a list of categorical columns, or use scikit-learn's " "KMeans instead." ) if isinstance(categorical, int): categorical = [categorical] assert len(categorical) != X.shape[1], \ "All columns are categorical, use k-modes instead of k-prototypes." assert max(categorical) < X.shape[1], \ "Categorical index larger than number of columns." ncatattrs = len(categorical) nnumattrs = X.shape[1] - ncatattrs n_points = X.shape[0] assert n_clusters <= n_points, "Cannot have more clusters ({}) " \ "than data points ({}).".format(n_clusters, n_points) Xnum, Xcat = _split_num_cat(X, categorical) Xnum, Xcat = check_array(Xnum), check_array(Xcat, dtype=None) # Convert the categorical values in Xcat to integers for speed. # Based on the unique values in Xcat, we can make a mapping to achieve this. Xcat, enc_map = encode_features(Xcat) # Are there more n_clusters than unique rows? Then set the unique # rows as initial values and skip iteration. unique = get_unique_rows(X) n_unique = unique.shape[0] if n_unique <= n_clusters: max_iter = 0 n_init = 1 n_clusters = n_unique init = list(_split_num_cat(unique, categorical)) init[1], _ = encode_features(init[1], enc_map) # Estimate a good value for gamma, which determines the weighing of # categorical values in clusters (see Huang [1997]). if gamma is None: gamma = 0.5 * Xnum.std() results = [] seeds = random_state.randint(np.iinfo(np.int32).max, size=n_init) if n_jobs == 1: for init_no in range(n_init): results.append(k_prototypes_single(Xnum, Xcat, nnumattrs, ncatattrs, n_clusters, n_points, max_iter, num_dissim, cat_dissim, gamma, init, init_no, verbose, seeds[init_no])) else: results = Parallel(n_jobs=n_jobs, verbose=0)( delayed(k_prototypes_single)(Xnum, Xcat, nnumattrs, ncatattrs, n_clusters, n_points, max_iter, num_dissim, cat_dissim, gamma, init, init_no, verbose, seed) for init_no, seed in enumerate(seeds)) all_centroids, all_labels, all_costs, all_n_iters = zip(*results) best = np.argmin(all_costs) if n_init > 1 and verbose: print("Best run was number {}".format(best + 1)) # Note: return gamma in case it was automatically determined. return all_centroids[best], enc_map, all_labels[best], \ all_costs[best], all_n_iters[best], gamma
k-prototypes algorithm
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/kprototypes.py#L255-L327
null
""" K-prototypes clustering for mixed categorical and numerical data """ # pylint: disable=super-on-old-class,unused-argument,attribute-defined-outside-init from collections import defaultdict import numpy as np from joblib import Parallel, delayed from scipy import sparse from sklearn.utils import check_random_state from sklearn.utils.validation import check_array from . import kmodes from .util import get_max_value_key, encode_features, get_unique_rows, \ decode_centroids, pandas_to_numpy from .util.dissim import matching_dissim, euclidean_dissim # Number of tries we give the initialization methods to find non-empty # clusters before we switch to random initialization. MAX_INIT_TRIES = 20 # Number of tries we give the initialization before we raise an # initialization error. RAISE_INIT_TRIES = 100 def move_point_num(point, to_clust, from_clust, cl_attr_sum, cl_memb_sum): """Move point between clusters, numerical attributes.""" # Update sum of attributes in cluster. for iattr, curattr in enumerate(point): cl_attr_sum[to_clust][iattr] += curattr cl_attr_sum[from_clust][iattr] -= curattr # Update sums of memberships in cluster cl_memb_sum[to_clust] += 1 cl_memb_sum[from_clust] -= 1 return cl_attr_sum, cl_memb_sum def _split_num_cat(X, categorical): """Extract numerical and categorical columns. Convert to numpy arrays, if needed. :param X: Feature matrix :param categorical: Indices of categorical columns """ Xnum = np.asanyarray(X[:, [ii for ii in range(X.shape[1]) if ii not in categorical]]).astype(np.float64) Xcat = np.asanyarray(X[:, categorical]) return Xnum, Xcat def _labels_cost(Xnum, Xcat, centroids, num_dissim, cat_dissim, gamma, membship=None): """Calculate labels and cost function given a matrix of points and a list of centroids for the k-prototypes algorithm. """ n_points = Xnum.shape[0] Xnum = check_array(Xnum) cost = 0. labels = np.empty(n_points, dtype=np.uint16) for ipoint in range(n_points): # Numerical cost = sum of Euclidean distances num_costs = num_dissim(centroids[0], Xnum[ipoint]) cat_costs = cat_dissim(centroids[1], Xcat[ipoint], X=Xcat, membship=membship) # Gamma relates the categorical cost to the numerical cost. tot_costs = num_costs + gamma * cat_costs clust = np.argmin(tot_costs) labels[ipoint] = clust cost += tot_costs[clust] return labels, cost def _k_prototypes_iter(Xnum, Xcat, centroids, cl_attr_sum, cl_memb_sum, cl_attr_freq, membship, num_dissim, cat_dissim, gamma, random_state): """Single iteration of the k-prototypes algorithm""" moves = 0 for ipoint in range(Xnum.shape[0]): clust = np.argmin( num_dissim(centroids[0], Xnum[ipoint]) + gamma * cat_dissim(centroids[1], Xcat[ipoint], X=Xcat, membship=membship) ) if membship[clust, ipoint]: # Point is already in its right place. continue # Move point, and update old/new cluster frequencies and centroids. moves += 1 old_clust = np.argwhere(membship[:, ipoint])[0][0] # Note that membship gets updated by kmodes.move_point_cat. # move_point_num only updates things specific to the k-means part. cl_attr_sum, cl_memb_sum = move_point_num( Xnum[ipoint], clust, old_clust, cl_attr_sum, cl_memb_sum ) cl_attr_freq, membship, centroids[1] = kmodes.move_point_cat( Xcat[ipoint], ipoint, clust, old_clust, cl_attr_freq, membship, centroids[1] ) # Update old and new centroids for numerical attributes using # the means and sums of all values for iattr in range(len(Xnum[ipoint])): for curc in (clust, old_clust): if cl_memb_sum[curc]: centroids[0][curc, iattr] = cl_attr_sum[curc, iattr] / cl_memb_sum[curc] else: centroids[0][curc, iattr] = 0. # In case of an empty cluster, reinitialize with a random point # from largest cluster. if not cl_memb_sum[old_clust]: from_clust = membship.sum(axis=1).argmax() choices = [ii for ii, ch in enumerate(membship[from_clust, :]) if ch] rindx = random_state.choice(choices) cl_attr_sum, cl_memb_sum = move_point_num( Xnum[rindx], old_clust, from_clust, cl_attr_sum, cl_memb_sum ) cl_attr_freq, membship, centroids[1] = kmodes.move_point_cat( Xcat[rindx], rindx, old_clust, from_clust, cl_attr_freq, membship, centroids[1] ) return centroids, moves def k_prototypes_single(Xnum, Xcat, nnumattrs, ncatattrs, n_clusters, n_points, max_iter, num_dissim, cat_dissim, gamma, init, init_no, verbose, random_state): # For numerical part of initialization, we don't have a guarantee # that there is not an empty cluster, so we need to retry until # there is none. random_state = check_random_state(random_state) init_tries = 0 while True: init_tries += 1 # _____ INIT _____ if verbose: print("Init: initializing centroids") if isinstance(init, str) and init.lower() == 'huang': centroids = kmodes.init_huang(Xcat, n_clusters, cat_dissim, random_state) elif isinstance(init, str) and init.lower() == 'cao': centroids = kmodes.init_cao(Xcat, n_clusters, cat_dissim) elif isinstance(init, str) and init.lower() == 'random': seeds = random_state.choice(range(n_points), n_clusters) centroids = Xcat[seeds] elif isinstance(init, list): # Make sure inits are 2D arrays. init = [np.atleast_2d(cur_init).T if len(cur_init.shape) == 1 else cur_init for cur_init in init] assert init[0].shape[0] == n_clusters, \ "Wrong number of initial numerical centroids in init " \ "({}, should be {}).".format(init[0].shape[0], n_clusters) assert init[0].shape[1] == nnumattrs, \ "Wrong number of numerical attributes in init ({}, should be {})." \ .format(init[0].shape[1], nnumattrs) assert init[1].shape[0] == n_clusters, \ "Wrong number of initial categorical centroids in init ({}, " \ "should be {}).".format(init[1].shape[0], n_clusters) assert init[1].shape[1] == ncatattrs, \ "Wrong number of categorical attributes in init ({}, should be {})." \ .format(init[1].shape[1], ncatattrs) centroids = [np.asarray(init[0], dtype=np.float64), np.asarray(init[1], dtype=np.uint16)] else: raise NotImplementedError("Initialization method not supported.") if not isinstance(init, list): # Numerical is initialized by drawing from normal distribution, # categorical following the k-modes methods. meanx = np.mean(Xnum, axis=0) stdx = np.std(Xnum, axis=0) centroids = [ meanx + random_state.randn(n_clusters, nnumattrs) * stdx, centroids ] if verbose: print("Init: initializing clusters") membship = np.zeros((n_clusters, n_points), dtype=np.uint8) # Keep track of the sum of attribute values per cluster so that we # can do k-means on the numerical attributes. cl_attr_sum = np.zeros((n_clusters, nnumattrs), dtype=np.float64) # Same for the membership sum per cluster cl_memb_sum = np.zeros(n_clusters, dtype=int) # cl_attr_freq is a list of lists with dictionaries that contain # the frequencies of values per cluster and attribute. cl_attr_freq = [[defaultdict(int) for _ in range(ncatattrs)] for _ in range(n_clusters)] for ipoint in range(n_points): # Initial assignment to clusters clust = np.argmin( num_dissim(centroids[0], Xnum[ipoint]) + gamma * cat_dissim(centroids[1], Xcat[ipoint], X=Xcat, membship=membship) ) membship[clust, ipoint] = 1 cl_memb_sum[clust] += 1 # Count attribute values per cluster. for iattr, curattr in enumerate(Xnum[ipoint]): cl_attr_sum[clust, iattr] += curattr for iattr, curattr in enumerate(Xcat[ipoint]): cl_attr_freq[clust][iattr][curattr] += 1 # If no empty clusters, then consider initialization finalized. if membship.sum(axis=1).min() > 0: break if init_tries == MAX_INIT_TRIES: # Could not get rid of empty clusters. Randomly # initialize instead. init = 'random' elif init_tries == RAISE_INIT_TRIES: raise ValueError( "Clustering algorithm could not initialize. " "Consider assigning the initial clusters manually." ) # Perform an initial centroid update. for ik in range(n_clusters): for iattr in range(nnumattrs): centroids[0][ik, iattr] = cl_attr_sum[ik, iattr] / cl_memb_sum[ik] for iattr in range(ncatattrs): centroids[1][ik, iattr] = get_max_value_key(cl_attr_freq[ik][iattr]) # _____ ITERATION _____ if verbose: print("Starting iterations...") itr = 0 labels = None converged = False cost = np.Inf while itr <= max_iter and not converged: itr += 1 centroids, moves = _k_prototypes_iter(Xnum, Xcat, centroids, cl_attr_sum, cl_memb_sum, cl_attr_freq, membship, num_dissim, cat_dissim, gamma, random_state) # All points seen in this iteration labels, ncost = _labels_cost(Xnum, Xcat, centroids, num_dissim, cat_dissim, gamma, membship) converged = (moves == 0) or (ncost >= cost) cost = ncost if verbose: print("Run: {}, iteration: {}/{}, moves: {}, ncost: {}" .format(init_no + 1, itr, max_iter, moves, ncost)) return centroids, labels, cost, itr class KPrototypes(kmodes.KModes): """k-protoypes clustering algorithm for mixed numerical/categorical data. Parameters ----------- n_clusters : int, optional, default: 8 The number of clusters to form as well as the number of centroids to generate. max_iter : int, default: 300 Maximum number of iterations of the k-modes algorithm for a single run. num_dissim : func, default: euclidian_dissim Dissimilarity function used by the algorithm for numerical variables. Defaults to the Euclidian dissimilarity function. cat_dissim : func, default: matching_dissim Dissimilarity function used by the kmodes algorithm for categorical variables. Defaults to the matching dissimilarity function. n_init : int, default: 10 Number of time the k-modes algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of cost. init : {'Huang', 'Cao', 'random' or a list of ndarrays}, default: 'Cao' Method for initialization: 'Huang': Method in Huang [1997, 1998] 'Cao': Method in Cao et al. [2009] 'random': choose 'n_clusters' observations (rows) at random from data for the initial centroids. If a list of ndarrays is passed, it should be of length 2, with shapes (n_clusters, n_features) for numerical and categorical data respectively. These are the initial centroids. gamma : float, default: None Weighing factor that determines relative importance of numerical vs. categorical attributes (see discussion in Huang [1997]). By default, automatically calculated from data. verbose : integer, optional Verbosity mode. random_state : int, RandomState instance or None, optional, default: None If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. n_jobs : int, default: 1 The number of jobs to use for the computation. This works by computing each of the n_init runs in parallel. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Attributes ---------- cluster_centroids_ : array, [n_clusters, n_features] Categories of cluster centroids labels_ : Labels of each point cost_ : float Clustering cost, defined as the sum distance of all points to their respective cluster centroids. n_iter_ : int The number of iterations the algorithm ran for. gamma : float The (potentially calculated) weighing factor. Notes ----- See: Huang, Z.: Extensions to the k-modes algorithm for clustering large data sets with categorical values, Data Mining and Knowledge Discovery 2(3), 1998. """ def __init__(self, n_clusters=8, max_iter=100, num_dissim=euclidean_dissim, cat_dissim=matching_dissim, init='Huang', n_init=10, gamma=None, verbose=0, random_state=None, n_jobs=1): super(KPrototypes, self).__init__(n_clusters, max_iter, cat_dissim, init, verbose=verbose, random_state=random_state, n_jobs=n_jobs) self.num_dissim = num_dissim self.gamma = gamma self.n_init = n_init if isinstance(self.init, list) and self.n_init > 1: if self.verbose: print("Initialization method is deterministic. " "Setting n_init to 1.") self.n_init = 1 def fit(self, X, y=None, categorical=None): """Compute k-prototypes clustering. Parameters ---------- X : array-like, shape=[n_samples, n_features] categorical : Index of columns that contain categorical data """ if categorical is not None: assert isinstance(categorical, (int, list, tuple)), "The 'categorical' \ argument needs to be an integer with the index of the categorical \ column in your data, or a list or tuple of several of them, \ but it is a {}.".format(type(categorical)) X = pandas_to_numpy(X) random_state = check_random_state(self.random_state) # If self.gamma is None, gamma will be automatically determined from # the data. The function below returns its value. self._enc_cluster_centroids, self._enc_map, self.labels_, self.cost_,\ self.n_iter_, self.gamma = k_prototypes(X, categorical, self.n_clusters, self.max_iter, self.num_dissim, self.cat_dissim, self.gamma, self.init, self.n_init, self.verbose, random_state, self.n_jobs) return self def predict(self, X, categorical=None): """Predict the closest cluster each sample in X belongs to. Parameters ---------- X : array-like, shape = [n_samples, n_features] New data to predict. categorical : Indices of columns that contain categorical data Returns ------- labels : array, shape [n_samples,] Index of the cluster each sample belongs to. """ assert hasattr(self, '_enc_cluster_centroids'), "Model not yet fitted." if categorical is not None: assert isinstance(categorical, (int, list, tuple)), "The 'categorical' \ argument needs to be an integer with the index of the categorical \ column in your data, or a list or tuple of several of them, \ but it is a {}.".format(type(categorical)) X = pandas_to_numpy(X) Xnum, Xcat = _split_num_cat(X, categorical) Xnum, Xcat = check_array(Xnum), check_array(Xcat, dtype=None) Xcat, _ = encode_features(Xcat, enc_map=self._enc_map) return _labels_cost(Xnum, Xcat, self._enc_cluster_centroids, self.num_dissim, self.cat_dissim, self.gamma)[0] @property def cluster_centroids_(self): if hasattr(self, '_enc_cluster_centroids'): return [ self._enc_cluster_centroids[0], decode_centroids(self._enc_cluster_centroids[1], self._enc_map) ] else: raise AttributeError("'{}' object has no attribute 'cluster_centroids_' " "because the model is not yet fitted.")
nicodv/kmodes
kmodes/kprototypes.py
KPrototypes.fit
python
def fit(self, X, y=None, categorical=None): if categorical is not None: assert isinstance(categorical, (int, list, tuple)), "The 'categorical' \ argument needs to be an integer with the index of the categorical \ column in your data, or a list or tuple of several of them, \ but it is a {}.".format(type(categorical)) X = pandas_to_numpy(X) random_state = check_random_state(self.random_state) # If self.gamma is None, gamma will be automatically determined from # the data. The function below returns its value. self._enc_cluster_centroids, self._enc_map, self.labels_, self.cost_,\ self.n_iter_, self.gamma = k_prototypes(X, categorical, self.n_clusters, self.max_iter, self.num_dissim, self.cat_dissim, self.gamma, self.init, self.n_init, self.verbose, random_state, self.n_jobs) return self
Compute k-prototypes clustering. Parameters ---------- X : array-like, shape=[n_samples, n_features] categorical : Index of columns that contain categorical data
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/kprototypes.py#L431-L463
[ "def pandas_to_numpy(x):\n return x.values if 'pandas' in str(x.__class__) else x\n", "def k_prototypes(X, categorical, n_clusters, max_iter, num_dissim, cat_dissim,\n gamma, init, n_init, verbose, random_state, n_jobs):\n \"\"\"k-prototypes algorithm\"\"\"\n random_state = check_random_state(random_state)\n if sparse.issparse(X):\n raise TypeError(\"k-prototypes does not support sparse data.\")\n\n if categorical is None or not categorical:\n raise NotImplementedError(\n \"No categorical data selected, effectively doing k-means. \"\n \"Present a list of categorical columns, or use scikit-learn's \"\n \"KMeans instead.\"\n )\n if isinstance(categorical, int):\n categorical = [categorical]\n assert len(categorical) != X.shape[1], \\\n \"All columns are categorical, use k-modes instead of k-prototypes.\"\n assert max(categorical) < X.shape[1], \\\n \"Categorical index larger than number of columns.\"\n\n ncatattrs = len(categorical)\n nnumattrs = X.shape[1] - ncatattrs\n n_points = X.shape[0]\n assert n_clusters <= n_points, \"Cannot have more clusters ({}) \" \\\n \"than data points ({}).\".format(n_clusters, n_points)\n\n Xnum, Xcat = _split_num_cat(X, categorical)\n Xnum, Xcat = check_array(Xnum), check_array(Xcat, dtype=None)\n\n # Convert the categorical values in Xcat to integers for speed.\n # Based on the unique values in Xcat, we can make a mapping to achieve this.\n Xcat, enc_map = encode_features(Xcat)\n\n # Are there more n_clusters than unique rows? Then set the unique\n # rows as initial values and skip iteration.\n unique = get_unique_rows(X)\n n_unique = unique.shape[0]\n if n_unique <= n_clusters:\n max_iter = 0\n n_init = 1\n n_clusters = n_unique\n init = list(_split_num_cat(unique, categorical))\n init[1], _ = encode_features(init[1], enc_map)\n\n # Estimate a good value for gamma, which determines the weighing of\n # categorical values in clusters (see Huang [1997]).\n if gamma is None:\n gamma = 0.5 * Xnum.std()\n\n results = []\n seeds = random_state.randint(np.iinfo(np.int32).max, size=n_init)\n if n_jobs == 1:\n for init_no in range(n_init):\n results.append(k_prototypes_single(Xnum, Xcat, nnumattrs, ncatattrs,\n n_clusters, n_points, max_iter,\n num_dissim, cat_dissim, gamma,\n init, init_no, verbose, seeds[init_no]))\n else:\n results = Parallel(n_jobs=n_jobs, verbose=0)(\n delayed(k_prototypes_single)(Xnum, Xcat, nnumattrs, ncatattrs,\n n_clusters, n_points, max_iter,\n num_dissim, cat_dissim, gamma,\n init, init_no, verbose, seed)\n for init_no, seed in enumerate(seeds))\n all_centroids, all_labels, all_costs, all_n_iters = zip(*results)\n\n best = np.argmin(all_costs)\n if n_init > 1 and verbose:\n print(\"Best run was number {}\".format(best + 1))\n\n # Note: return gamma in case it was automatically determined.\n return all_centroids[best], enc_map, all_labels[best], \\\n all_costs[best], all_n_iters[best], gamma\n" ]
class KPrototypes(kmodes.KModes): """k-protoypes clustering algorithm for mixed numerical/categorical data. Parameters ----------- n_clusters : int, optional, default: 8 The number of clusters to form as well as the number of centroids to generate. max_iter : int, default: 300 Maximum number of iterations of the k-modes algorithm for a single run. num_dissim : func, default: euclidian_dissim Dissimilarity function used by the algorithm for numerical variables. Defaults to the Euclidian dissimilarity function. cat_dissim : func, default: matching_dissim Dissimilarity function used by the kmodes algorithm for categorical variables. Defaults to the matching dissimilarity function. n_init : int, default: 10 Number of time the k-modes algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of cost. init : {'Huang', 'Cao', 'random' or a list of ndarrays}, default: 'Cao' Method for initialization: 'Huang': Method in Huang [1997, 1998] 'Cao': Method in Cao et al. [2009] 'random': choose 'n_clusters' observations (rows) at random from data for the initial centroids. If a list of ndarrays is passed, it should be of length 2, with shapes (n_clusters, n_features) for numerical and categorical data respectively. These are the initial centroids. gamma : float, default: None Weighing factor that determines relative importance of numerical vs. categorical attributes (see discussion in Huang [1997]). By default, automatically calculated from data. verbose : integer, optional Verbosity mode. random_state : int, RandomState instance or None, optional, default: None If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. n_jobs : int, default: 1 The number of jobs to use for the computation. This works by computing each of the n_init runs in parallel. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Attributes ---------- cluster_centroids_ : array, [n_clusters, n_features] Categories of cluster centroids labels_ : Labels of each point cost_ : float Clustering cost, defined as the sum distance of all points to their respective cluster centroids. n_iter_ : int The number of iterations the algorithm ran for. gamma : float The (potentially calculated) weighing factor. Notes ----- See: Huang, Z.: Extensions to the k-modes algorithm for clustering large data sets with categorical values, Data Mining and Knowledge Discovery 2(3), 1998. """ def __init__(self, n_clusters=8, max_iter=100, num_dissim=euclidean_dissim, cat_dissim=matching_dissim, init='Huang', n_init=10, gamma=None, verbose=0, random_state=None, n_jobs=1): super(KPrototypes, self).__init__(n_clusters, max_iter, cat_dissim, init, verbose=verbose, random_state=random_state, n_jobs=n_jobs) self.num_dissim = num_dissim self.gamma = gamma self.n_init = n_init if isinstance(self.init, list) and self.n_init > 1: if self.verbose: print("Initialization method is deterministic. " "Setting n_init to 1.") self.n_init = 1 def predict(self, X, categorical=None): """Predict the closest cluster each sample in X belongs to. Parameters ---------- X : array-like, shape = [n_samples, n_features] New data to predict. categorical : Indices of columns that contain categorical data Returns ------- labels : array, shape [n_samples,] Index of the cluster each sample belongs to. """ assert hasattr(self, '_enc_cluster_centroids'), "Model not yet fitted." if categorical is not None: assert isinstance(categorical, (int, list, tuple)), "The 'categorical' \ argument needs to be an integer with the index of the categorical \ column in your data, or a list or tuple of several of them, \ but it is a {}.".format(type(categorical)) X = pandas_to_numpy(X) Xnum, Xcat = _split_num_cat(X, categorical) Xnum, Xcat = check_array(Xnum), check_array(Xcat, dtype=None) Xcat, _ = encode_features(Xcat, enc_map=self._enc_map) return _labels_cost(Xnum, Xcat, self._enc_cluster_centroids, self.num_dissim, self.cat_dissim, self.gamma)[0] @property def cluster_centroids_(self): if hasattr(self, '_enc_cluster_centroids'): return [ self._enc_cluster_centroids[0], decode_centroids(self._enc_cluster_centroids[1], self._enc_map) ] else: raise AttributeError("'{}' object has no attribute 'cluster_centroids_' " "because the model is not yet fitted.")
nicodv/kmodes
kmodes/kprototypes.py
KPrototypes.predict
python
def predict(self, X, categorical=None): assert hasattr(self, '_enc_cluster_centroids'), "Model not yet fitted." if categorical is not None: assert isinstance(categorical, (int, list, tuple)), "The 'categorical' \ argument needs to be an integer with the index of the categorical \ column in your data, or a list or tuple of several of them, \ but it is a {}.".format(type(categorical)) X = pandas_to_numpy(X) Xnum, Xcat = _split_num_cat(X, categorical) Xnum, Xcat = check_array(Xnum), check_array(Xcat, dtype=None) Xcat, _ = encode_features(Xcat, enc_map=self._enc_map) return _labels_cost(Xnum, Xcat, self._enc_cluster_centroids, self.num_dissim, self.cat_dissim, self.gamma)[0]
Predict the closest cluster each sample in X belongs to. Parameters ---------- X : array-like, shape = [n_samples, n_features] New data to predict. categorical : Indices of columns that contain categorical data Returns ------- labels : array, shape [n_samples,] Index of the cluster each sample belongs to.
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/kprototypes.py#L465-L492
[ "def pandas_to_numpy(x):\n return x.values if 'pandas' in str(x.__class__) else x\n", "def encode_features(X, enc_map=None):\n \"\"\"Converts categorical values in each column of X to integers in the range\n [0, n_unique_values_in_column - 1], if X is not already of integer type.\n\n If mapping is not provided, it is calculated based on the values in X.\n\n Unknown values during prediction get a value of -1. np.NaNs are ignored\n during encoding, and get treated as unknowns during prediction.\n \"\"\"\n if enc_map is None:\n fit = True\n # We will calculate enc_map, so initialize the list of column mappings.\n enc_map = []\n else:\n fit = False\n\n Xenc = np.zeros(X.shape).astype('int')\n for ii in range(X.shape[1]):\n if fit:\n col_enc = {val: jj for jj, val in enumerate(np.unique(X[:, ii]))\n if not (isinstance(val, float) and np.isnan(val))}\n enc_map.append(col_enc)\n # Unknown categories (including np.NaNs) all get a value of -1.\n Xenc[:, ii] = np.array([enc_map[ii].get(x, -1) for x in X[:, ii]])\n\n return Xenc, enc_map\n", "def _labels_cost(Xnum, Xcat, centroids, num_dissim, cat_dissim, gamma, membship=None):\n \"\"\"Calculate labels and cost function given a matrix of points and\n a list of centroids for the k-prototypes algorithm.\n \"\"\"\n\n n_points = Xnum.shape[0]\n Xnum = check_array(Xnum)\n\n cost = 0.\n labels = np.empty(n_points, dtype=np.uint16)\n for ipoint in range(n_points):\n # Numerical cost = sum of Euclidean distances\n num_costs = num_dissim(centroids[0], Xnum[ipoint])\n cat_costs = cat_dissim(centroids[1], Xcat[ipoint], X=Xcat, membship=membship)\n # Gamma relates the categorical cost to the numerical cost.\n tot_costs = num_costs + gamma * cat_costs\n clust = np.argmin(tot_costs)\n labels[ipoint] = clust\n cost += tot_costs[clust]\n\n return labels, cost\n", "def _split_num_cat(X, categorical):\n \"\"\"Extract numerical and categorical columns.\n Convert to numpy arrays, if needed.\n\n :param X: Feature matrix\n :param categorical: Indices of categorical columns\n \"\"\"\n Xnum = np.asanyarray(X[:, [ii for ii in range(X.shape[1])\n if ii not in categorical]]).astype(np.float64)\n Xcat = np.asanyarray(X[:, categorical])\n return Xnum, Xcat\n" ]
class KPrototypes(kmodes.KModes): """k-protoypes clustering algorithm for mixed numerical/categorical data. Parameters ----------- n_clusters : int, optional, default: 8 The number of clusters to form as well as the number of centroids to generate. max_iter : int, default: 300 Maximum number of iterations of the k-modes algorithm for a single run. num_dissim : func, default: euclidian_dissim Dissimilarity function used by the algorithm for numerical variables. Defaults to the Euclidian dissimilarity function. cat_dissim : func, default: matching_dissim Dissimilarity function used by the kmodes algorithm for categorical variables. Defaults to the matching dissimilarity function. n_init : int, default: 10 Number of time the k-modes algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of cost. init : {'Huang', 'Cao', 'random' or a list of ndarrays}, default: 'Cao' Method for initialization: 'Huang': Method in Huang [1997, 1998] 'Cao': Method in Cao et al. [2009] 'random': choose 'n_clusters' observations (rows) at random from data for the initial centroids. If a list of ndarrays is passed, it should be of length 2, with shapes (n_clusters, n_features) for numerical and categorical data respectively. These are the initial centroids. gamma : float, default: None Weighing factor that determines relative importance of numerical vs. categorical attributes (see discussion in Huang [1997]). By default, automatically calculated from data. verbose : integer, optional Verbosity mode. random_state : int, RandomState instance or None, optional, default: None If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. n_jobs : int, default: 1 The number of jobs to use for the computation. This works by computing each of the n_init runs in parallel. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Attributes ---------- cluster_centroids_ : array, [n_clusters, n_features] Categories of cluster centroids labels_ : Labels of each point cost_ : float Clustering cost, defined as the sum distance of all points to their respective cluster centroids. n_iter_ : int The number of iterations the algorithm ran for. gamma : float The (potentially calculated) weighing factor. Notes ----- See: Huang, Z.: Extensions to the k-modes algorithm for clustering large data sets with categorical values, Data Mining and Knowledge Discovery 2(3), 1998. """ def __init__(self, n_clusters=8, max_iter=100, num_dissim=euclidean_dissim, cat_dissim=matching_dissim, init='Huang', n_init=10, gamma=None, verbose=0, random_state=None, n_jobs=1): super(KPrototypes, self).__init__(n_clusters, max_iter, cat_dissim, init, verbose=verbose, random_state=random_state, n_jobs=n_jobs) self.num_dissim = num_dissim self.gamma = gamma self.n_init = n_init if isinstance(self.init, list) and self.n_init > 1: if self.verbose: print("Initialization method is deterministic. " "Setting n_init to 1.") self.n_init = 1 def fit(self, X, y=None, categorical=None): """Compute k-prototypes clustering. Parameters ---------- X : array-like, shape=[n_samples, n_features] categorical : Index of columns that contain categorical data """ if categorical is not None: assert isinstance(categorical, (int, list, tuple)), "The 'categorical' \ argument needs to be an integer with the index of the categorical \ column in your data, or a list or tuple of several of them, \ but it is a {}.".format(type(categorical)) X = pandas_to_numpy(X) random_state = check_random_state(self.random_state) # If self.gamma is None, gamma will be automatically determined from # the data. The function below returns its value. self._enc_cluster_centroids, self._enc_map, self.labels_, self.cost_,\ self.n_iter_, self.gamma = k_prototypes(X, categorical, self.n_clusters, self.max_iter, self.num_dissim, self.cat_dissim, self.gamma, self.init, self.n_init, self.verbose, random_state, self.n_jobs) return self @property def cluster_centroids_(self): if hasattr(self, '_enc_cluster_centroids'): return [ self._enc_cluster_centroids[0], decode_centroids(self._enc_cluster_centroids[1], self._enc_map) ] else: raise AttributeError("'{}' object has no attribute 'cluster_centroids_' " "because the model is not yet fitted.")
nicodv/kmodes
kmodes/util/dissim.py
euclidean_dissim
python
def euclidean_dissim(a, b, **_): if np.isnan(a).any() or np.isnan(b).any(): raise ValueError("Missing values detected in numerical columns.") return np.sum((a - b) ** 2, axis=1)
Euclidean distance dissimilarity function
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/util/dissim.py#L13-L17
null
""" Dissimilarity measures for clustering """ import numpy as np def matching_dissim(a, b, **_): """Simple matching dissimilarity function""" return np.sum(a != b, axis=1) def ng_dissim(a, b, X=None, membship=None): """Ng et al.'s dissimilarity measure, as presented in Michael K. Ng, Mark Junjie Li, Joshua Zhexue Huang, and Zengyou He, "On the Impact of Dissimilarity Measure in k-Modes Clustering Algorithm", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 29, No. 3, January, 2007 This function can potentially speed up training convergence. Note that membship must be a rectangular array such that the len(membship) = len(a) and len(membship[i]) = X.shape[1] In case of missing membship, this function reverts back to matching dissimilarity (e.g., when predicting). """ # Without membership, revert to matching dissimilarity if membship is None: return matching_dissim(a, b) def calc_cjr(b, X, memj, idr): """Num objects w/ category value x_{i,r} for rth attr in jth cluster""" xcids = np.where(memj == 1) return float((np.take(X, xcids, axis=0)[0][:, idr] == b[idr]).sum(0)) def calc_dissim(b, X, memj, idr): # Size of jth cluster cj = float(np.sum(memj)) return (1.0 - (calc_cjr(b, X, memj, idr) / cj)) if cj != 0.0 else 0.0 if len(membship) != a.shape[0] and len(membship[0]) != X.shape[1]: raise ValueError("'membship' must be a rectangular array where " "the number of rows in 'membship' equals the " "number of rows in 'a' and the number of " "columns in 'membship' equals the number of rows in 'X'.") return np.array([np.array([calc_dissim(b, X, membship[idj], idr) if b[idr] == t else 1.0 for idr, t in enumerate(val_a)]).sum(0) for idj, val_a in enumerate(a)])
nicodv/kmodes
kmodes/util/dissim.py
ng_dissim
python
def ng_dissim(a, b, X=None, membship=None): # Without membership, revert to matching dissimilarity if membship is None: return matching_dissim(a, b) def calc_cjr(b, X, memj, idr): """Num objects w/ category value x_{i,r} for rth attr in jth cluster""" xcids = np.where(memj == 1) return float((np.take(X, xcids, axis=0)[0][:, idr] == b[idr]).sum(0)) def calc_dissim(b, X, memj, idr): # Size of jth cluster cj = float(np.sum(memj)) return (1.0 - (calc_cjr(b, X, memj, idr) / cj)) if cj != 0.0 else 0.0 if len(membship) != a.shape[0] and len(membship[0]) != X.shape[1]: raise ValueError("'membship' must be a rectangular array where " "the number of rows in 'membship' equals the " "number of rows in 'a' and the number of " "columns in 'membship' equals the number of rows in 'X'.") return np.array([np.array([calc_dissim(b, X, membship[idj], idr) if b[idr] == t else 1.0 for idr, t in enumerate(val_a)]).sum(0) for idj, val_a in enumerate(a)])
Ng et al.'s dissimilarity measure, as presented in Michael K. Ng, Mark Junjie Li, Joshua Zhexue Huang, and Zengyou He, "On the Impact of Dissimilarity Measure in k-Modes Clustering Algorithm", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 29, No. 3, January, 2007 This function can potentially speed up training convergence. Note that membship must be a rectangular array such that the len(membship) = len(a) and len(membship[i]) = X.shape[1] In case of missing membship, this function reverts back to matching dissimilarity (e.g., when predicting).
train
https://github.com/nicodv/kmodes/blob/cdb19fe5448aba1bf501626694bb52e68eafab45/kmodes/util/dissim.py#L20-L58
[ "def matching_dissim(a, b, **_):\n \"\"\"Simple matching dissimilarity function\"\"\"\n return np.sum(a != b, axis=1)\n" ]
""" Dissimilarity measures for clustering """ import numpy as np def matching_dissim(a, b, **_): """Simple matching dissimilarity function""" return np.sum(a != b, axis=1) def euclidean_dissim(a, b, **_): """Euclidean distance dissimilarity function""" if np.isnan(a).any() or np.isnan(b).any(): raise ValueError("Missing values detected in numerical columns.") return np.sum((a - b) ** 2, axis=1)
aerkalov/ebooklib
ebooklib/epub.py
write_epub
python
def write_epub(name, book, options=None): epub = EpubWriter(name, book, options) epub.process() try: epub.write() except IOError: pass
Creates epub file with the content defined in EpubBook. >>> ebooklib.write_epub('book.epub', book) :Args: - name: file name for the output file - book: instance of EpubBook - options: extra opions as dictionary (optional)
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L1705-L1723
[ "def process(self):\n # should cache this html parsing so we don't do it for every plugin\n for plg in self.options.get('plugins', []):\n if hasattr(plg, 'before_write'):\n plg.before_write(self.book)\n\n for item in self.book.get_items():\n if isinstance(item, EpubHtml):\n for plg in self.options.get('plugins', []):\n if hasattr(plg, 'html_before_write'):\n plg.html_before_write(self.book, item)\n", "def write(self):\n # check for the option allowZip64\n self.out = zipfile.ZipFile(self.file_name, 'w', zipfile.ZIP_DEFLATED)\n self.out.writestr('mimetype', 'application/epub+zip', compress_type=zipfile.ZIP_STORED)\n\n self._write_container()\n self._write_opf()\n self._write_items()\n\n self.out.close()\n" ]
# This file is part of EbookLib. # Copyright (c) 2013 Aleksandar Erkalovic <aerkalov@gmail.com> # # EbookLib is free software: you can redistribute it and/or modify # it under the terms of the GNU Affero General Public License as published by # the Free Software Foundation, either version 3 of the License, or # (at your option) any later version. # # EbookLib is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU Affero General Public License for more details. # # You should have received a copy of the GNU Affero General Public License # along with EbookLib. If not, see <http://www.gnu.org/licenses/>. import zipfile import six import logging import uuid import posixpath as zip_path import os.path from collections import OrderedDict try: from urllib.parse import unquote except ImportError: from urllib import unquote from lxml import etree import ebooklib from ebooklib.utils import parse_string, parse_html_string, guess_type, get_pages_for_items # Version of EPUB library VERSION = (0, 17, 1) NAMESPACES = {'XML': 'http://www.w3.org/XML/1998/namespace', 'EPUB': 'http://www.idpf.org/2007/ops', 'DAISY': 'http://www.daisy.org/z3986/2005/ncx/', 'OPF': 'http://www.idpf.org/2007/opf', 'CONTAINERNS': 'urn:oasis:names:tc:opendocument:xmlns:container', 'DC': 'http://purl.org/dc/elements/1.1/', 'XHTML': 'http://www.w3.org/1999/xhtml'} # XML Templates CONTAINER_PATH = 'META-INF/container.xml' CONTAINER_XML = '''<?xml version='1.0' encoding='utf-8'?> <container xmlns="urn:oasis:names:tc:opendocument:xmlns:container" version="1.0"> <rootfiles> <rootfile media-type="application/oebps-package+xml" full-path="%(folder_name)s/content.opf"/> </rootfiles> </container> ''' NCX_XML = six.b('''<!DOCTYPE ncx PUBLIC "-//NISO//DTD ncx 2005-1//EN" "http://www.daisy.org/z3986/2005/ncx-2005-1.dtd"> <ncx xmlns="http://www.daisy.org/z3986/2005/ncx/" version="2005-1" />''') NAV_XML = six.b('''<?xml version="1.0" encoding="utf-8"?><!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops"/>''') CHAPTER_XML = six.b('''<?xml version="1.0" encoding="UTF-8"?><!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" epub:prefix="z3998: http://www.daisy.org/z3998/2012/vocab/structure/#"></html>''') COVER_XML = six.b('''<?xml version="1.0" encoding="UTF-8"?> <!DOCTYPE html> <html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" lang="en" xml:lang="en"> <head> <style> body { margin: 0em; padding: 0em; } img { max-width: 100%; max-height: 100%; } </style> </head> <body> <img src="" alt="" /> </body> </html>''') IMAGE_MEDIA_TYPES = ['image/jpeg', 'image/jpg', 'image/png', 'image/svg+xml'] # TOC and navigation elements class Section(object): def __init__(self, title, href=''): self.title = title self.href = href class Link(object): def __init__(self, href, title, uid=None): self.href = href self.title = title self.uid = uid # Exceptions class EpubException(Exception): def __init__(self, code, msg): self.code = code self.msg = msg def __str__(self): return repr(self.msg) # Items class EpubItem(object): """ Base class for the items in a book. """ def __init__(self, uid=None, file_name='', media_type='', content=six.b(''), manifest=True): """ :Args: - uid: Unique identifier for this item (optional) - file_name: File name for this item (optional) - media_type: Media type for this item (optional) - content: Content for this item (optional) - manifest: Manifest for this item (optional) """ self.id = uid self.file_name = file_name self.media_type = media_type self.content = content self.is_linear = True self.manifest = manifest self.book = None def get_id(self): """ Returns unique identifier for this item. :Returns: Returns uid number as string. """ return self.id def get_name(self): """ Returns name for this item. By default it is always file name but it does not have to be. :Returns: Returns file name for this item. """ return self.file_name def get_type(self): """ Guess type according to the file extension. Might not be the best way how to do it, but it works for now. Items can be of type: - ITEM_UNKNOWN = 0 - ITEM_IMAGE = 1 - ITEM_STYLE = 2 - ITEM_SCRIPT = 3 - ITEM_NAVIGATION = 4 - ITEM_VECTOR = 5 - ITEM_FONT = 6 - ITEM_VIDEO = 7 - ITEM_AUDIO = 8 - ITEM_DOCUMENT = 9 - ITEM_COVER = 10 We map type according to the extensions which are defined in ebooklib.EXTENSIONS. :Returns: Returns type of the item as number. """ _, ext = zip_path.splitext(self.get_name()) ext = ext.lower() for uid, ext_list in six.iteritems(ebooklib.EXTENSIONS): if ext in ext_list: return uid return ebooklib.ITEM_UNKNOWN def get_content(self, default=six.b('')): """ Returns content of the item. Content should be of type 'str' (Python 2) or 'bytes' (Python 3) :Args: - default: Default value for the content if it is not already defined. :Returns: Returns content of the item. """ return self.content or default def set_content(self, content): """ Sets content value for this item. :Args: - content: Content value """ self.content = content def __str__(self): return '<EpubItem:%s>' % self.id class EpubNcx(EpubItem): "Represents Navigation Control File (NCX) in the EPUB." def __init__(self, uid='ncx', file_name='toc.ncx'): super(EpubNcx, self).__init__(uid=uid, file_name=file_name, media_type='application/x-dtbncx+xml') def __str__(self): return '<EpubNcx:%s>' % self.id class EpubCover(EpubItem): """ Represents Cover image in the EPUB file. """ def __init__(self, uid='cover-img', file_name=''): super(EpubCover, self).__init__(uid=uid, file_name=file_name) def get_type(self): return ebooklib.ITEM_COVER def __str__(self): return '<EpubCover:%s:%s>' % (self.id, self.file_name) class EpubHtml(EpubItem): """ Represents HTML document in the EPUB file. """ _template_name = 'chapter' def __init__(self, uid=None, file_name='', media_type='', content=None, title='', lang=None, direction=None, media_overlay=None, media_duration=None): super(EpubHtml, self).__init__(uid, file_name, media_type, content) self.title = title self.lang = lang self.direction = direction self.media_overlay = media_overlay self.media_duration = media_duration self.links = [] self.properties = [] self.pages = [] def is_chapter(self): """ Returns if this document is chapter or not. :Returns: Returns book value. """ return True def get_type(self): """ Always returns ebooklib.ITEM_DOCUMENT as type of this document. :Returns: Always returns ebooklib.ITEM_DOCUMENT """ return ebooklib.ITEM_DOCUMENT def set_language(self, lang): """ Sets language for this book item. By default it will use language of the book but it can be overwritten with this call. """ self.lang = lang def get_language(self): """ Get language code for this book item. Language of the book item can be different from the language settings defined globaly for book. :Returns: As string returns language code. """ return self.lang def add_link(self, **kwgs): """ Add additional link to the document. Links will be embeded only inside of this document. >>> add_link(href='styles.css', rel='stylesheet', type='text/css') """ self.links.append(kwgs) if kwgs.get('type') == 'text/javascript': if 'scripted' not in self.properties: self.properties.append('scripted') def get_links(self): """ Returns list of additional links defined for this document. :Returns: As tuple return list of links. """ return (link for link in self.links) def get_links_of_type(self, link_type): """ Returns list of additional links of specific type. :Returns: As tuple returns list of links. """ return (link for link in self.links if link.get('type', '') == link_type) def add_item(self, item): """ Add other item to this document. It will create additional links according to the item type. :Args: - item: item we want to add defined as instance of EpubItem """ if item.get_type() == ebooklib.ITEM_STYLE: self.add_link(href=item.get_name(), rel='stylesheet', type='text/css') if item.get_type() == ebooklib.ITEM_SCRIPT: self.add_link(src=item.get_name(), type='text/javascript') def get_body_content(self): """ Returns content of BODY element for this HTML document. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Returns: Returns content of this document. """ try: html_tree = parse_html_string(self.content) except: return '' html_root = html_tree.getroottree() if len(html_root.find('body')) != 0: body = html_tree.find('body') tree_str = etree.tostring(body, pretty_print=True, encoding='utf-8', xml_declaration=False) # this is so stupid if tree_str.startswith(six.b('<body>')): n = tree_str.rindex(six.b('</body>')) return tree_str[6:n] return tree_str return '' def get_content(self, default=None): """ Returns content for this document as HTML string. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Args: - default: Default value for the content if it is not defined. :Returns: Returns content of this document. """ tree = parse_string(self.book.get_template(self._template_name)) tree_root = tree.getroot() tree_root.set('lang', self.lang or self.book.language) tree_root.attrib['{%s}lang' % NAMESPACES['XML']] = self.lang or self.book.language # add to the head also # <meta charset="utf-8" /> try: html_tree = parse_html_string(self.content) except: return '' html_root = html_tree.getroottree() # create and populate head _head = etree.SubElement(tree_root, 'head') if self.title != '': _title = etree.SubElement(_head, 'title') _title.text = self.title for lnk in self.links: if lnk.get('type') == 'text/javascript': _lnk = etree.SubElement(_head, 'script', lnk) # force <script></script> _lnk.text = '' else: _lnk = etree.SubElement(_head, 'link', lnk) # this should not be like this # head = html_root.find('head') # if head is not None: # for i in head.getchildren(): # if i.tag == 'title' and self.title != '': # continue # _head.append(i) # create and populate body _body = etree.SubElement(tree_root, 'body') if self.direction: _body.set('dir', self.direction) tree_root.set('dir', self.direction) body = html_tree.find('body') if body is not None: for i in body.getchildren(): _body.append(i) tree_str = etree.tostring(tree, pretty_print=True, encoding='utf-8', xml_declaration=True) return tree_str def __str__(self): return '<EpubHtml:%s:%s>' % (self.id, self.file_name) class EpubCoverHtml(EpubHtml): """ Represents Cover page in the EPUB file. """ def __init__(self, uid='cover', file_name='cover.xhtml', image_name='', title='Cover'): super(EpubCoverHtml, self).__init__(uid=uid, file_name=file_name, title=title) self.image_name = image_name self.is_linear = False def is_chapter(self): """ Returns if this document is chapter or not. :Returns: Returns book value. """ return False def get_content(self): """ Returns content for cover page as HTML string. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Returns: Returns content of this document. """ self.content = self.book.get_template('cover') tree = parse_string(super(EpubCoverHtml, self).get_content()) tree_root = tree.getroot() images = tree_root.xpath('//xhtml:img', namespaces={'xhtml': NAMESPACES['XHTML']}) images[0].set('src', self.image_name) images[0].set('alt', self.title) tree_str = etree.tostring(tree, pretty_print=True, encoding='utf-8', xml_declaration=True) return tree_str def __str__(self): return '<EpubCoverHtml:%s:%s>' % (self.id, self.file_name) class EpubNav(EpubHtml): """ Represents Navigation Document in the EPUB file. """ def __init__(self, uid='nav', file_name='nav.xhtml', media_type='application/xhtml+xml'): super(EpubNav, self).__init__(uid=uid, file_name=file_name, media_type=media_type) def is_chapter(self): """ Returns if this document is chapter or not. :Returns: Returns book value. """ return False def __str__(self): return '<EpubNav:%s:%s>' % (self.id, self.file_name) class EpubImage(EpubItem): """ Represents Image in the EPUB file. """ def __init__(self): super(EpubImage, self).__init__() def get_type(self): return ebooklib.ITEM_IMAGE def __str__(self): return '<EpubImage:%s:%s>' % (self.id, self.file_name) class EpubSMIL(EpubItem): def __init__(self, uid=None, file_name='', content=None): super(EpubSMIL, self).__init__(uid=uid, file_name=file_name, media_type='application/smil+xml', content=content) def get_type(self): return ebooklib.ITEM_SMIL def __str__(self): return '<EpubSMIL:%s:%s>' % (self.id, self.file_name) # EpubBook class EpubBook(object): def __init__(self): self.EPUB_VERSION = None self.reset() # we should have options here def reset(self): "Initialises all needed variables to default values" self.metadata = {} self.items = [] self.spine = [] self.guide = [] self.pages = [] self.toc = [] self.bindings = [] self.IDENTIFIER_ID = 'id' self.FOLDER_NAME = 'EPUB' self._id_html = 0 self._id_image = 0 self._id_static = 0 self.title = '' self.language = 'en' self.direction = None self.templates = { 'ncx': NCX_XML, 'nav': NAV_XML, 'chapter': CHAPTER_XML, 'cover': COVER_XML } self.add_metadata('OPF', 'generator', '', { 'name': 'generator', 'content': 'Ebook-lib %s' % '.'.join([str(s) for s in VERSION]) }) # default to using a randomly-unique identifier if one is not specified manually self.set_identifier(str(uuid.uuid4())) # custom prefixes and namespaces to be set to the content.opf doc self.prefixes = [] self.namespaces = {} def set_identifier(self, uid): """ Sets unique id for this epub :Args: - uid: Value of unique identifier for this book """ self.uid = uid self.set_unique_metadata('DC', 'identifier', self.uid, {'id': self.IDENTIFIER_ID}) def set_title(self, title): """ Set title. You can set multiple titles. :Args: - title: Title value """ self.title = title self.add_metadata('DC', 'title', self.title) def set_language(self, lang): """ Set language for this epub. You can set multiple languages. Specific items in the book can have different language settings. :Args: - lang: Language code """ self.language = lang self.add_metadata('DC', 'language', lang) def set_direction(self, direction): """ :Args: - direction: Options are "ltr", "rtl" and "default" """ self.direction = direction def set_cover(self, file_name, content, create_page=True): """ Set cover and create cover document if needed. :Args: - file_name: file name of the cover page - content: Content for the cover image - create_page: Should cover page be defined. Defined as bool value (optional). Default value is True. """ # as it is now, it can only be called once c0 = EpubCover(file_name=file_name) c0.content = content self.add_item(c0) if create_page: c1 = EpubCoverHtml(image_name=file_name) self.add_item(c1) self.add_metadata(None, 'meta', '', OrderedDict([('name', 'cover'), ('content', 'cover-img')])) def add_author(self, author, file_as=None, role=None, uid='creator'): "Add author for this document" self.add_metadata('DC', 'creator', author, {'id': uid}) if file_as: self.add_metadata(None, 'meta', file_as, {'refines': '#' + uid, 'property': 'file-as', 'scheme': 'marc:relators'}) if role: self.add_metadata(None, 'meta', role, {'refines': '#' + uid, 'property': 'role', 'scheme': 'marc:relators'}) def add_metadata(self, namespace, name, value, others=None): "Add metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace not in self.metadata: self.metadata[namespace] = {} if name not in self.metadata[namespace]: self.metadata[namespace][name] = [] self.metadata[namespace][name].append((value, others)) def get_metadata(self, namespace, name): "Retrieve metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] return self.metadata[namespace].get(name, []) def set_unique_metadata(self, namespace, name, value, others=None): "Add metadata if metadata with this identifier does not already exist, otherwise update existing metadata." if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace in self.metadata and name in self.metadata[namespace]: self.metadata[namespace][name] = [(value, others)] else: self.add_metadata(namespace, name, value, others) def add_item(self, item): """ Add additional item to the book. If not defined, media type and chapter id will be defined for the item. :Args: - item: Item instance """ if item.media_type == '': (has_guessed, media_type) = guess_type(item.get_name().lower()) if has_guessed: if media_type is not None: item.media_type = media_type else: item.media_type = has_guessed else: item.media_type = 'application/octet-stream' if not item.get_id(): # make chapter_, image_ and static_ configurable if isinstance(item, EpubHtml): item.id = 'chapter_%d' % self._id_html self._id_html += 1 # If there's a page list, append it to the book's page list self.pages += item.pages elif isinstance(item, EpubImage): item.id = 'image_%d' % self._id_image self._id_image += 1 else: item.id = 'static_%d' % self._id_image self._id_image += 1 item.book = self self.items.append(item) return item def get_item_with_id(self, uid): """ Returns item for defined UID. >>> book.get_item_with_id('image_001') :Args: - uid: UID for the item :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.id == uid: return item return None def get_item_with_href(self, href): """ Returns item for defined HREF. >>> book.get_item_with_href('EPUB/document.xhtml') :Args: - href: HREF for the item we are searching for :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.get_name() == href: return item return None def get_items(self): """ Returns all items attached to this book. :Returns: Returns all items as tuple. """ return (item for item in self.items) def get_items_of_type(self, item_type): """ Returns all items of specified type. >>> book.get_items_of_type(epub.ITEM_IMAGE) :Args: - item_type: Type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.get_type() == item_type) def get_items_of_media_type(self, media_type): """ Returns all items of specified media type. :Args: - media_type: Media type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.media_type == media_type) def set_template(self, name, value): """ Defines templates which are used to generate certain types of pages. When defining new value for the template we have to use content of type 'str' (Python 2) or 'bytes' (Python 3). At the moment we use these templates: - ncx - nav - chapter - cover :Args: - name: Name for the template - value: Content for the template """ self.templates[name] = value def get_template(self, name): """ Returns value for the template. :Args: - name: template name :Returns: Value of the template. """ return self.templates.get(name) def add_prefix(self, name, uri): """ Appends custom prefix to be added to the content.opf document >>> epub_book.add_prefix('bkterms', 'http://booktype.org/') :Args: - name: namespave name - uri: URI for the namespace """ self.prefixes.append('%s: %s' % (name, uri)) class EpubWriter(object): DEFAULT_OPTIONS = { 'epub2_guide': True, 'epub3_landmark': True, 'epub3_pages': True, 'landmark_title': 'Guide', 'pages_title': 'Pages', 'spine_direction': True, 'package_direction': False, 'play_order': { 'enabled': False, 'start_from': 1 } } def __init__(self, name, book, options=None): self.file_name = name self.book = book self.options = dict(self.DEFAULT_OPTIONS) if options: self.options.update(options) self._init_play_order() def _init_play_order(self): self._play_order = { 'enabled': False, 'start_from': 1 } try: self._play_order['enabled'] = self.options['play_order']['enabled'] self._play_order['start_from'] = self.options['play_order']['start_from'] except KeyError: pass def process(self): # should cache this html parsing so we don't do it for every plugin for plg in self.options.get('plugins', []): if hasattr(plg, 'before_write'): plg.before_write(self.book) for item in self.book.get_items(): if isinstance(item, EpubHtml): for plg in self.options.get('plugins', []): if hasattr(plg, 'html_before_write'): plg.html_before_write(self.book, item) def _write_container(self): container_xml = CONTAINER_XML % {'folder_name': self.book.FOLDER_NAME} self.out.writestr(CONTAINER_PATH, container_xml) def _write_opf_metadata(self, root): # This is really not needed # problem is uppercase/lowercase # for ns_name, values in six.iteritems(self.book.metadata): # if ns_name: # for n_id, ns_url in six.iteritems(NAMESPACES): # if ns_name == ns_url: # nsmap[n_id.lower()] = NAMESPACES[n_id] nsmap = {'dc': NAMESPACES['DC'], 'opf': NAMESPACES['OPF']} nsmap.update(self.book.namespaces) metadata = etree.SubElement(root, 'metadata', nsmap=nsmap) el = etree.SubElement(metadata, 'meta', {'property': 'dcterms:modified'}) if 'mtime' in self.options: mtime = self.options['mtime'] else: import datetime mtime = datetime.datetime.now() el.text = mtime.strftime('%Y-%m-%dT%H:%M:%SZ') for ns_name, values in six.iteritems(self.book.metadata): if ns_name == NAMESPACES['OPF']: for values in values.values(): for v in values: if 'property' in v[1] and v[1]['property'] == 'dcterms:modified': continue try: el = etree.SubElement(metadata, 'meta', v[1]) if v[0]: el.text = v[0] except ValueError: logging.error('Could not create metadata.') else: for name, values in six.iteritems(values): for v in values: try: if ns_name: el = etree.SubElement(metadata, '{%s}%s' % (ns_name, name), v[1]) else: el = etree.SubElement(metadata, '%s' % name, v[1]) el.text = v[0] except ValueError: logging.error('Could not create metadata "{}".'.format(name)) def _write_opf_manifest(self, root): manifest = etree.SubElement(root, 'manifest') _ncx_id = None # mathml, scripted, svg, remote-resources, and switch # nav # cover-image for item in self.book.get_items(): if not item.manifest: continue if isinstance(item, EpubNav): etree.SubElement(manifest, 'item', {'href': item.get_name(), 'id': item.id, 'media-type': item.media_type, 'properties': 'nav'}) elif isinstance(item, EpubNcx): _ncx_id = item.id etree.SubElement(manifest, 'item', {'href': item.file_name, 'id': item.id, 'media-type': item.media_type}) elif isinstance(item, EpubCover): etree.SubElement(manifest, 'item', {'href': item.file_name, 'id': item.id, 'media-type': item.media_type, 'properties': 'cover-image'}) else: opts = {'href': item.file_name, 'id': item.id, 'media-type': item.media_type} if hasattr(item, 'properties') and len(item.properties) > 0: opts['properties'] = ' '.join(item.properties) if hasattr(item, 'media_overlay') and item.media_overlay is not None: opts['media-overlay'] = item.media_overlay if hasattr(item, 'media_duration') and item.media_duration is not None: opts['duration'] = item.media_duration etree.SubElement(manifest, 'item', opts) return _ncx_id def _write_opf_spine(self, root, ncx_id): spine_attributes = {'toc': ncx_id or 'ncx'} if self.book.direction and self.options['spine_direction']: spine_attributes['page-progression-direction'] = self.book.direction spine = etree.SubElement(root, 'spine', spine_attributes) for _item in self.book.spine: # this is for now # later we should be able to fetch things from tuple is_linear = True if isinstance(_item, tuple): item = _item[0] if len(_item) > 1: if _item[1] == 'no': is_linear = False else: item = _item if isinstance(item, EpubHtml): opts = {'idref': item.get_id()} if not item.is_linear or not is_linear: opts['linear'] = 'no' elif isinstance(item, EpubItem): opts = {'idref': item.get_id()} if not item.is_linear or not is_linear: opts['linear'] = 'no' else: opts = {'idref': item} try: itm = self.book.get_item_with_id(item) if not itm.is_linear or not is_linear: opts['linear'] = 'no' except: pass etree.SubElement(spine, 'itemref', opts) def _write_opf_guide(self, root): # - http://www.idpf.org/epub/20/spec/OPF_2.0.1_draft.htm#Section2.6 if len(self.book.guide) > 0 and self.options.get('epub2_guide'): guide = etree.SubElement(root, 'guide', {}) for item in self.book.guide: if 'item' in item: chap = item.get('item') if chap: _href = chap.file_name _title = chap.title else: _href = item.get('href', '') _title = item.get('title', '') if _title is None: _title = '' ref = etree.SubElement(guide, 'reference', {'type': item.get('type', ''), 'title': _title, 'href': _href}) def _write_opf_bindings(self, root): if len(self.book.bindings) > 0: bindings = etree.SubElement(root, 'bindings', {}) for item in self.book.bindings: etree.SubElement(bindings, 'mediaType', item) def _write_opf_file(self, root): tree_str = etree.tostring(root, pretty_print=True, encoding='utf-8', xml_declaration=True) self.out.writestr('%s/content.opf' % self.book.FOLDER_NAME, tree_str) def _write_opf(self): package_attributes = {'xmlns': NAMESPACES['OPF'], 'unique-identifier': self.book.IDENTIFIER_ID, 'version': '3.0'} if self.book.direction and self.options['package_direction']: package_attributes['dir'] = self.book.direction root = etree.Element('package', package_attributes) prefixes = ['rendition: http://www.idpf.org/vocab/rendition/#'] + self.book.prefixes root.attrib['prefix'] = ' '.join(prefixes) # METADATA self._write_opf_metadata(root) # MANIFEST _ncx_id = self._write_opf_manifest(root) # SPINE self._write_opf_spine(root, _ncx_id) # GUIDE self._write_opf_guide(root) # BINDINGS self._write_opf_bindings(root) # WRITE FILE self._write_opf_file(root) def _get_nav(self, item): # just a basic navigation for now nav_xml = parse_string(self.book.get_template('nav')) root = nav_xml.getroot() root.set('lang', self.book.language) root.attrib['{%s}lang' % NAMESPACES['XML']] = self.book.language nav_dir_name = os.path.dirname(item.file_name) head = etree.SubElement(root, 'head') title = etree.SubElement(head, 'title') title.text = self.book.title # for now this just handles css files and ignores others for _link in item.links: _lnk = etree.SubElement(head, 'link', { 'href': _link.get('href', ''), 'rel': 'stylesheet', 'type': 'text/css' }) body = etree.SubElement(root, 'body') nav = etree.SubElement(body, 'nav', { '{%s}type' % NAMESPACES['EPUB']: 'toc', 'id': 'id', 'role': 'doc-toc', }) content_title = etree.SubElement(nav, 'h2') content_title.text = self.book.title def _create_section(itm, items): ol = etree.SubElement(itm, 'ol') for item in items: if isinstance(item, tuple) or isinstance(item, list): li = etree.SubElement(ol, 'li') if isinstance(item[0], EpubHtml): a = etree.SubElement(li, 'a', {'href': os.path.relpath(item[0].file_name, nav_dir_name)}) elif isinstance(item[0], Section) and item[0].href != '': a = etree.SubElement(li, 'a', {'href': os.path.relpath(item[0].href, nav_dir_name)}) elif isinstance(item[0], Link): a = etree.SubElement(li, 'a', {'href': os.path.relpath(item[0].href, nav_dir_name)}) else: a = etree.SubElement(li, 'span') a.text = item[0].title _create_section(li, item[1]) elif isinstance(item, Link): li = etree.SubElement(ol, 'li') a = etree.SubElement(li, 'a', {'href': os.path.relpath(item.href, nav_dir_name)}) a.text = item.title elif isinstance(item, EpubHtml): li = etree.SubElement(ol, 'li') a = etree.SubElement(li, 'a', {'href': os.path.relpath(item.file_name, nav_dir_name)}) a.text = item.title _create_section(nav, self.book.toc) # LANDMARKS / GUIDE # - http://www.idpf.org/epub/30/spec/epub30-contentdocs.html#sec-xhtml-nav-def-types-landmarks if len(self.book.guide) > 0 and self.options.get('epub3_landmark'): # Epub2 guide types do not map completely to epub3 landmark types. guide_to_landscape_map = { 'notes': 'rearnotes', 'text': 'bodymatter' } guide_nav = etree.SubElement(body, 'nav', {'{%s}type' % NAMESPACES['EPUB']: 'landmarks'}) guide_content_title = etree.SubElement(guide_nav, 'h2') guide_content_title.text = self.options.get('landmark_title', 'Guide') guild_ol = etree.SubElement(guide_nav, 'ol') for elem in self.book.guide: li_item = etree.SubElement(guild_ol, 'li') if 'item' in elem: chap = elem.get('item', None) if chap: _href = chap.file_name _title = chap.title else: _href = elem.get('href', '') _title = elem.get('title', '') guide_type = elem.get('type', '') a_item = etree.SubElement(li_item, 'a', { '{%s}type' % NAMESPACES['EPUB']: guide_to_landscape_map.get(guide_type, guide_type), 'href': os.path.relpath(_href, nav_dir_name) }) a_item.text = _title # PAGE-LIST if self.options.get('epub3_pages'): inserted_pages = get_pages_for_items([item for item in self.book.get_items_of_type(ebooklib.ITEM_DOCUMENT) \ if not isinstance(item, EpubNav)]) if len(inserted_pages) > 0: pagelist_nav = etree.SubElement( body, 'nav', { '{%s}type' % NAMESPACES['EPUB']: 'page-list', 'id': 'pages', 'hidden': 'hidden', } ) pagelist_content_title = etree.SubElement(pagelist_nav, 'h2') pagelist_content_title.text = self.options.get( 'pages_title', 'Pages' ) pages_ol = etree.SubElement(pagelist_nav, 'ol') for filename, pageref, label in inserted_pages: li_item = etree.SubElement(pages_ol, 'li') _href = u'{}#{}'.format(filename, pageref) _title = label a_item = etree.SubElement(li_item, 'a', { 'href': os.path.relpath(_href, nav_dir_name), }) a_item.text = _title tree_str = etree.tostring(nav_xml, pretty_print=True, encoding='utf-8', xml_declaration=True) return tree_str def _get_ncx(self): # we should be able to setup language for NCX as also ncx = parse_string(self.book.get_template('ncx')) root = ncx.getroot() head = etree.SubElement(root, 'head') # get this id uid = etree.SubElement(head, 'meta', {'content': self.book.uid, 'name': 'dtb:uid'}) uid = etree.SubElement(head, 'meta', {'content': '0', 'name': 'dtb:depth'}) uid = etree.SubElement(head, 'meta', {'content': '0', 'name': 'dtb:totalPageCount'}) uid = etree.SubElement(head, 'meta', {'content': '0', 'name': 'dtb:maxPageNumber'}) doc_title = etree.SubElement(root, 'docTitle') title = etree.SubElement(doc_title, 'text') title.text = self.book.title # doc_author = etree.SubElement(root, 'docAuthor') # author = etree.SubElement(doc_author, 'text') # author.text = 'Name of the person' # For now just make a very simple navMap nav_map = etree.SubElement(root, 'navMap') def _add_play_order(nav_point): nav_point.set('playOrder', str(self._play_order['start_from'])) self._play_order['start_from'] += 1 def _create_section(itm, items, uid): for item in items: if isinstance(item, tuple) or isinstance(item, list): section, subsection = item[0], item[1] np = etree.SubElement(itm, 'navPoint', { 'id': section.get_id() if isinstance(section, EpubHtml) else 'sep_%d' % uid }) if self._play_order['enabled']: _add_play_order(np) nl = etree.SubElement(np, 'navLabel') nt = etree.SubElement(nl, 'text') nt.text = section.title # CAN NOT HAVE EMPTY SRC HERE href = '' if isinstance(section, EpubHtml): href = section.file_name elif isinstance(section, Section) and section.href != '': href = section.href elif isinstance(section, Link): href = section.href nc = etree.SubElement(np, 'content', {'src': href}) uid = _create_section(np, subsection, uid + 1) elif isinstance(item, Link): _parent = itm _content = _parent.find('content') if _content is not None: if _content.get('src') == '': _content.set('src', item.href) np = etree.SubElement(itm, 'navPoint', {'id': item.uid}) if self._play_order['enabled']: _add_play_order(np) nl = etree.SubElement(np, 'navLabel') nt = etree.SubElement(nl, 'text') nt.text = item.title nc = etree.SubElement(np, 'content', {'src': item.href}) elif isinstance(item, EpubHtml): _parent = itm _content = _parent.find('content') if _content is not None: if _content.get('src') == '': _content.set('src', item.file_name) np = etree.SubElement(itm, 'navPoint', {'id': item.get_id()}) if self._play_order['enabled']: _add_play_order(np) nl = etree.SubElement(np, 'navLabel') nt = etree.SubElement(nl, 'text') nt.text = item.title nc = etree.SubElement(np, 'content', {'src': item.file_name}) return uid _create_section(nav_map, self.book.toc, 0) tree_str = etree.tostring(root, pretty_print=True, encoding='utf-8', xml_declaration=True) return tree_str def _write_items(self): for item in self.book.get_items(): if isinstance(item, EpubNcx): self.out.writestr('%s/%s' % (self.book.FOLDER_NAME, item.file_name), self._get_ncx()) elif isinstance(item, EpubNav): self.out.writestr('%s/%s' % (self.book.FOLDER_NAME, item.file_name), self._get_nav(item)) elif item.manifest: self.out.writestr('%s/%s' % (self.book.FOLDER_NAME, item.file_name), item.get_content()) else: self.out.writestr('%s' % item.file_name, item.get_content()) def write(self): # check for the option allowZip64 self.out = zipfile.ZipFile(self.file_name, 'w', zipfile.ZIP_DEFLATED) self.out.writestr('mimetype', 'application/epub+zip', compress_type=zipfile.ZIP_STORED) self._write_container() self._write_opf() self._write_items() self.out.close() class EpubReader(object): DEFAULT_OPTIONS = {} def __init__(self, epub_file_name, options=None): self.file_name = epub_file_name self.book = EpubBook() self.zf = None self.opf_file = '' self.opf_dir = '' self.options = dict(self.DEFAULT_OPTIONS) if options: self.options.update(options) def process(self): # should cache this html parsing so we don't do it for every plugin for plg in self.options.get('plugins', []): if hasattr(plg, 'after_read'): plg.after_read(self.book) for item in self.book.get_items(): if isinstance(item, EpubHtml): for plg in self.options.get('plugins', []): if hasattr(plg, 'html_after_read'): plg.html_after_read(self.book, item) def load(self): self._load() return self.book def read_file(self, name): # Raises KeyError name = zip_path.normpath(name) return self.zf.read(name) def _load_container(self): meta_inf = self.read_file('META-INF/container.xml') tree = parse_string(meta_inf) for root_file in tree.findall('//xmlns:rootfile[@media-type]', namespaces={'xmlns': NAMESPACES['CONTAINERNS']}): if root_file.get('media-type') == 'application/oebps-package+xml': self.opf_file = root_file.get('full-path') self.opf_dir = zip_path.dirname(self.opf_file) def _load_metadata(self): container_root = self.container.getroot() # get epub version self.book.version = container_root.get('version', None) # get unique-identifier if container_root.get('unique-identifier', None): self.book.IDENTIFIER_ID = container_root.get('unique-identifier') # get xml:lang # get metadata metadata = self.container.find('{%s}%s' % (NAMESPACES['OPF'], 'metadata')) nsmap = metadata.nsmap nstags = dict((k, '{%s}' % v) for k, v in six.iteritems(nsmap)) default_ns = nstags.get(None, '') nsdict = dict((v, {}) for v in nsmap.values()) def add_item(ns, tag, value, extra): if ns not in nsdict: nsdict[ns] = {} values = nsdict[ns].setdefault(tag, []) values.append((value, extra)) for t in metadata: if not etree.iselement(t) or t.tag is etree.Comment: continue if t.tag == default_ns + 'meta': name = t.get('name') others = dict((k, v) for k, v in t.items()) if name and ':' in name: prefix, name = name.split(':', 1) else: prefix = None add_item(t.nsmap.get(prefix, prefix), name, t.text, others) else: tag = t.tag[t.tag.rfind('}') + 1:] if (t.prefix and t.prefix.lower() == 'dc') and tag == 'identifier': _id = t.get('id', None) if _id: self.book.IDENTIFIER_ID = _id others = dict((k, v) for k, v in t.items()) add_item(t.nsmap[t.prefix], tag, t.text, others) self.book.metadata = nsdict titles = self.book.get_metadata('DC', 'title') if len(titles) > 0: self.book.title = titles[0][0] for value, others in self.book.get_metadata('DC', 'identifier'): if others.get('id') == self.book.IDENTIFIER_ID: self.book.uid = value def _load_manifest(self): for r in self.container.find('{%s}%s' % (NAMESPACES['OPF'], 'manifest')): if r is not None and r.tag != '{%s}item' % NAMESPACES['OPF']: continue media_type = r.get('media-type') _properties = r.get('properties', '') if _properties: properties = _properties.split(' ') else: properties = [] # people use wrong content types if media_type == 'image/jpg': media_type = 'image/jpeg' if media_type == 'application/x-dtbncx+xml': ei = EpubNcx(uid=r.get('id'), file_name=unquote(r.get('href'))) ei.content = self.read_file(zip_path.join(self.opf_dir, ei.file_name)) if media_type == 'application/smil+xml': ei = EpubSMIL(uid=r.get('id'), file_name=unquote(r.get('href'))) ei.content = self.read_file(zip_path.join(self.opf_dir, ei.file_name)) elif media_type == 'application/xhtml+xml': if 'nav' in properties: ei = EpubNav(uid=r.get('id'), file_name=unquote(r.get('href'))) ei.content = self.read_file(zip_path.join(self.opf_dir, r.get('href'))) elif 'cover' in properties: ei = EpubCoverHtml() ei.content = self.read_file(zip_path.join(self.opf_dir, unquote(r.get('href')))) else: ei = EpubHtml() ei.id = r.get('id') ei.file_name = unquote(r.get('href')) ei.media_type = media_type ei.media_overlay = r.get('media-overlay', None) ei.media_duration = r.get('duration', None) ei.content = self.read_file(zip_path.join(self.opf_dir, ei.get_name())) ei.properties = properties elif media_type in IMAGE_MEDIA_TYPES: if 'cover-image' in properties: ei = EpubCover(uid=r.get('id'), file_name=unquote(r.get('href'))) ei.media_type = media_type ei.content = self.read_file(zip_path.join(self.opf_dir, ei.get_name())) else: ei = EpubImage() ei.id = r.get('id') ei.file_name = unquote(r.get('href')) ei.media_type = media_type ei.content = self.read_file(zip_path.join(self.opf_dir, ei.get_name())) else: # different types ei = EpubItem() ei.id = r.get('id') ei.file_name = unquote(r.get('href')) ei.media_type = media_type ei.content = self.read_file(zip_path.join(self.opf_dir, ei.get_name())) self.book.add_item(ei) def _parse_ncx(self, data): tree = parse_string(data) tree_root = tree.getroot() nav_map = tree_root.find('{%s}navMap' % NAMESPACES['DAISY']) def _get_children(elems, n, nid): label, content = '', '' children = [] for a in elems.getchildren(): if a.tag == '{%s}navLabel' % NAMESPACES['DAISY']: label = a.getchildren()[0].text if a.tag == '{%s}content' % NAMESPACES['DAISY']: content = a.get('src', '') if a.tag == '{%s}navPoint' % NAMESPACES['DAISY']: children.append(_get_children(a, n + 1, a.get('id', ''))) if len(children) > 0: if n == 0: return children return (Section(label, href=content), children) else: return Link(content, label, nid) self.book.toc = _get_children(nav_map, 0, '') def _parse_nav(self, data, base_path, navtype='toc'): html_node = parse_html_string(data) if navtype == 'toc': # parsing the table of contents nav_node = html_node.xpath("//nav[@*='toc']")[0] else: # parsing the list of pages _page_list = html_node.xpath("//nav[@*='page-list']") if len(_page_list) == 0: return nav_node = _page_list[0] def parse_list(list_node): items = [] for item_node in list_node.findall('li'): sublist_node = item_node.find('ol') link_node = item_node.find('a') if sublist_node is not None: title = item_node[0].text children = parse_list(sublist_node) if link_node is not None: href = zip_path.normpath(zip_path.join(base_path, link_node.get('href'))) items.append((Section(title, href=href), children)) else: items.append((Section(title), children)) elif link_node is not None: title = link_node.text href = zip_path.normpath(zip_path.join(base_path, link_node.get('href'))) items.append(Link(href, title)) return items if navtype == 'toc': self.book.toc = parse_list(nav_node.find('ol')) elif nav_node is not None: # generate the pages list if there is one self.book.pages = parse_list(nav_node.find('ol')) # generate the per-file pages lists # because of the order of parsing the files, this can't be done # when building the EpubHtml objects htmlfiles = dict() for htmlfile in self.book.items: if isinstance(htmlfile, EpubHtml): htmlfiles[htmlfile.file_name] = htmlfile for page in self.book.pages: try: (filename, idref) = page.href.split('#') except ValueError: filename = page.href if filename in htmlfiles: htmlfiles[filename].pages.append(page) def _load_spine(self): spine = self.container.find('{%s}%s' % (NAMESPACES['OPF'], 'spine')) self.book.spine = [(t.get('idref'), t.get('linear', 'yes')) for t in spine] toc = spine.get('toc', '') self.book.set_direction(spine.get('page-progression-direction', None)) # should read ncx or nav file if toc: try: ncxFile = self.read_file(zip_path.join(self.opf_dir, self.book.get_item_with_id(toc).get_name())) except KeyError: raise EpubException(-1, 'Can not find ncx file.') self._parse_ncx(ncxFile) def _load_guide(self): guide = self.container.find('{%s}%s' % (NAMESPACES['OPF'], 'guide')) if guide is not None: self.book.guide = [{'href': t.get('href'), 'title': t.get('title'), 'type': t.get('type')} for t in guide] def _load_opf_file(self): try: s = self.read_file(self.opf_file) except KeyError: raise EpubException(-1, 'Can not find container file') self.container = parse_string(s) self._load_metadata() self._load_manifest() self._load_spine() self._load_guide() # read nav file if found # nav_item = next((item for item in self.book.items if isinstance(item, EpubNav)), None) if nav_item: if not self.book.toc: self._parse_nav( nav_item.content, zip_path.dirname(nav_item.file_name), navtype='toc' ) self._parse_nav( nav_item.content, zip_path.dirname(nav_item.file_name), navtype='pages' ) def _load(self): try: self.zf = zipfile.ZipFile(self.file_name, 'r', compression=zipfile.ZIP_DEFLATED, allowZip64=True) except zipfile.BadZipfile as bz: raise EpubException(0, 'Bad Zip file') except zipfile.LargeZipFile as bz: raise EpubException(1, 'Large Zip file') # 1st check metadata self._load_container() self._load_opf_file() self.zf.close() # WRITE # READ def read_epub(name, options=None): """ Creates new instance of EpubBook with the content defined in the input file. >>> book = ebooklib.read_epub('book.epub') :Args: - name: full path to the input file - options: extra options as dictionary (optional) :Returns: Instance of EpubBook. """ reader = EpubReader(name, options) book = reader.load() reader.process() return book
aerkalov/ebooklib
ebooklib/epub.py
read_epub
python
def read_epub(name, options=None): reader = EpubReader(name, options) book = reader.load() reader.process() return book
Creates new instance of EpubBook with the content defined in the input file. >>> book = ebooklib.read_epub('book.epub') :Args: - name: full path to the input file - options: extra options as dictionary (optional) :Returns: Instance of EpubBook.
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L1728-L1746
[ "def process(self):\n # should cache this html parsing so we don't do it for every plugin\n for plg in self.options.get('plugins', []):\n if hasattr(plg, 'after_read'):\n plg.after_read(self.book)\n\n for item in self.book.get_items():\n if isinstance(item, EpubHtml):\n for plg in self.options.get('plugins', []):\n if hasattr(plg, 'html_after_read'):\n plg.html_after_read(self.book, item)\n", "def load(self):\n self._load()\n\n return self.book\n" ]
# This file is part of EbookLib. # Copyright (c) 2013 Aleksandar Erkalovic <aerkalov@gmail.com> # # EbookLib is free software: you can redistribute it and/or modify # it under the terms of the GNU Affero General Public License as published by # the Free Software Foundation, either version 3 of the License, or # (at your option) any later version. # # EbookLib is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU Affero General Public License for more details. # # You should have received a copy of the GNU Affero General Public License # along with EbookLib. If not, see <http://www.gnu.org/licenses/>. import zipfile import six import logging import uuid import posixpath as zip_path import os.path from collections import OrderedDict try: from urllib.parse import unquote except ImportError: from urllib import unquote from lxml import etree import ebooklib from ebooklib.utils import parse_string, parse_html_string, guess_type, get_pages_for_items # Version of EPUB library VERSION = (0, 17, 1) NAMESPACES = {'XML': 'http://www.w3.org/XML/1998/namespace', 'EPUB': 'http://www.idpf.org/2007/ops', 'DAISY': 'http://www.daisy.org/z3986/2005/ncx/', 'OPF': 'http://www.idpf.org/2007/opf', 'CONTAINERNS': 'urn:oasis:names:tc:opendocument:xmlns:container', 'DC': 'http://purl.org/dc/elements/1.1/', 'XHTML': 'http://www.w3.org/1999/xhtml'} # XML Templates CONTAINER_PATH = 'META-INF/container.xml' CONTAINER_XML = '''<?xml version='1.0' encoding='utf-8'?> <container xmlns="urn:oasis:names:tc:opendocument:xmlns:container" version="1.0"> <rootfiles> <rootfile media-type="application/oebps-package+xml" full-path="%(folder_name)s/content.opf"/> </rootfiles> </container> ''' NCX_XML = six.b('''<!DOCTYPE ncx PUBLIC "-//NISO//DTD ncx 2005-1//EN" "http://www.daisy.org/z3986/2005/ncx-2005-1.dtd"> <ncx xmlns="http://www.daisy.org/z3986/2005/ncx/" version="2005-1" />''') NAV_XML = six.b('''<?xml version="1.0" encoding="utf-8"?><!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops"/>''') CHAPTER_XML = six.b('''<?xml version="1.0" encoding="UTF-8"?><!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" epub:prefix="z3998: http://www.daisy.org/z3998/2012/vocab/structure/#"></html>''') COVER_XML = six.b('''<?xml version="1.0" encoding="UTF-8"?> <!DOCTYPE html> <html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" lang="en" xml:lang="en"> <head> <style> body { margin: 0em; padding: 0em; } img { max-width: 100%; max-height: 100%; } </style> </head> <body> <img src="" alt="" /> </body> </html>''') IMAGE_MEDIA_TYPES = ['image/jpeg', 'image/jpg', 'image/png', 'image/svg+xml'] # TOC and navigation elements class Section(object): def __init__(self, title, href=''): self.title = title self.href = href class Link(object): def __init__(self, href, title, uid=None): self.href = href self.title = title self.uid = uid # Exceptions class EpubException(Exception): def __init__(self, code, msg): self.code = code self.msg = msg def __str__(self): return repr(self.msg) # Items class EpubItem(object): """ Base class for the items in a book. """ def __init__(self, uid=None, file_name='', media_type='', content=six.b(''), manifest=True): """ :Args: - uid: Unique identifier for this item (optional) - file_name: File name for this item (optional) - media_type: Media type for this item (optional) - content: Content for this item (optional) - manifest: Manifest for this item (optional) """ self.id = uid self.file_name = file_name self.media_type = media_type self.content = content self.is_linear = True self.manifest = manifest self.book = None def get_id(self): """ Returns unique identifier for this item. :Returns: Returns uid number as string. """ return self.id def get_name(self): """ Returns name for this item. By default it is always file name but it does not have to be. :Returns: Returns file name for this item. """ return self.file_name def get_type(self): """ Guess type according to the file extension. Might not be the best way how to do it, but it works for now. Items can be of type: - ITEM_UNKNOWN = 0 - ITEM_IMAGE = 1 - ITEM_STYLE = 2 - ITEM_SCRIPT = 3 - ITEM_NAVIGATION = 4 - ITEM_VECTOR = 5 - ITEM_FONT = 6 - ITEM_VIDEO = 7 - ITEM_AUDIO = 8 - ITEM_DOCUMENT = 9 - ITEM_COVER = 10 We map type according to the extensions which are defined in ebooklib.EXTENSIONS. :Returns: Returns type of the item as number. """ _, ext = zip_path.splitext(self.get_name()) ext = ext.lower() for uid, ext_list in six.iteritems(ebooklib.EXTENSIONS): if ext in ext_list: return uid return ebooklib.ITEM_UNKNOWN def get_content(self, default=six.b('')): """ Returns content of the item. Content should be of type 'str' (Python 2) or 'bytes' (Python 3) :Args: - default: Default value for the content if it is not already defined. :Returns: Returns content of the item. """ return self.content or default def set_content(self, content): """ Sets content value for this item. :Args: - content: Content value """ self.content = content def __str__(self): return '<EpubItem:%s>' % self.id class EpubNcx(EpubItem): "Represents Navigation Control File (NCX) in the EPUB." def __init__(self, uid='ncx', file_name='toc.ncx'): super(EpubNcx, self).__init__(uid=uid, file_name=file_name, media_type='application/x-dtbncx+xml') def __str__(self): return '<EpubNcx:%s>' % self.id class EpubCover(EpubItem): """ Represents Cover image in the EPUB file. """ def __init__(self, uid='cover-img', file_name=''): super(EpubCover, self).__init__(uid=uid, file_name=file_name) def get_type(self): return ebooklib.ITEM_COVER def __str__(self): return '<EpubCover:%s:%s>' % (self.id, self.file_name) class EpubHtml(EpubItem): """ Represents HTML document in the EPUB file. """ _template_name = 'chapter' def __init__(self, uid=None, file_name='', media_type='', content=None, title='', lang=None, direction=None, media_overlay=None, media_duration=None): super(EpubHtml, self).__init__(uid, file_name, media_type, content) self.title = title self.lang = lang self.direction = direction self.media_overlay = media_overlay self.media_duration = media_duration self.links = [] self.properties = [] self.pages = [] def is_chapter(self): """ Returns if this document is chapter or not. :Returns: Returns book value. """ return True def get_type(self): """ Always returns ebooklib.ITEM_DOCUMENT as type of this document. :Returns: Always returns ebooklib.ITEM_DOCUMENT """ return ebooklib.ITEM_DOCUMENT def set_language(self, lang): """ Sets language for this book item. By default it will use language of the book but it can be overwritten with this call. """ self.lang = lang def get_language(self): """ Get language code for this book item. Language of the book item can be different from the language settings defined globaly for book. :Returns: As string returns language code. """ return self.lang def add_link(self, **kwgs): """ Add additional link to the document. Links will be embeded only inside of this document. >>> add_link(href='styles.css', rel='stylesheet', type='text/css') """ self.links.append(kwgs) if kwgs.get('type') == 'text/javascript': if 'scripted' not in self.properties: self.properties.append('scripted') def get_links(self): """ Returns list of additional links defined for this document. :Returns: As tuple return list of links. """ return (link for link in self.links) def get_links_of_type(self, link_type): """ Returns list of additional links of specific type. :Returns: As tuple returns list of links. """ return (link for link in self.links if link.get('type', '') == link_type) def add_item(self, item): """ Add other item to this document. It will create additional links according to the item type. :Args: - item: item we want to add defined as instance of EpubItem """ if item.get_type() == ebooklib.ITEM_STYLE: self.add_link(href=item.get_name(), rel='stylesheet', type='text/css') if item.get_type() == ebooklib.ITEM_SCRIPT: self.add_link(src=item.get_name(), type='text/javascript') def get_body_content(self): """ Returns content of BODY element for this HTML document. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Returns: Returns content of this document. """ try: html_tree = parse_html_string(self.content) except: return '' html_root = html_tree.getroottree() if len(html_root.find('body')) != 0: body = html_tree.find('body') tree_str = etree.tostring(body, pretty_print=True, encoding='utf-8', xml_declaration=False) # this is so stupid if tree_str.startswith(six.b('<body>')): n = tree_str.rindex(six.b('</body>')) return tree_str[6:n] return tree_str return '' def get_content(self, default=None): """ Returns content for this document as HTML string. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Args: - default: Default value for the content if it is not defined. :Returns: Returns content of this document. """ tree = parse_string(self.book.get_template(self._template_name)) tree_root = tree.getroot() tree_root.set('lang', self.lang or self.book.language) tree_root.attrib['{%s}lang' % NAMESPACES['XML']] = self.lang or self.book.language # add to the head also # <meta charset="utf-8" /> try: html_tree = parse_html_string(self.content) except: return '' html_root = html_tree.getroottree() # create and populate head _head = etree.SubElement(tree_root, 'head') if self.title != '': _title = etree.SubElement(_head, 'title') _title.text = self.title for lnk in self.links: if lnk.get('type') == 'text/javascript': _lnk = etree.SubElement(_head, 'script', lnk) # force <script></script> _lnk.text = '' else: _lnk = etree.SubElement(_head, 'link', lnk) # this should not be like this # head = html_root.find('head') # if head is not None: # for i in head.getchildren(): # if i.tag == 'title' and self.title != '': # continue # _head.append(i) # create and populate body _body = etree.SubElement(tree_root, 'body') if self.direction: _body.set('dir', self.direction) tree_root.set('dir', self.direction) body = html_tree.find('body') if body is not None: for i in body.getchildren(): _body.append(i) tree_str = etree.tostring(tree, pretty_print=True, encoding='utf-8', xml_declaration=True) return tree_str def __str__(self): return '<EpubHtml:%s:%s>' % (self.id, self.file_name) class EpubCoverHtml(EpubHtml): """ Represents Cover page in the EPUB file. """ def __init__(self, uid='cover', file_name='cover.xhtml', image_name='', title='Cover'): super(EpubCoverHtml, self).__init__(uid=uid, file_name=file_name, title=title) self.image_name = image_name self.is_linear = False def is_chapter(self): """ Returns if this document is chapter or not. :Returns: Returns book value. """ return False def get_content(self): """ Returns content for cover page as HTML string. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Returns: Returns content of this document. """ self.content = self.book.get_template('cover') tree = parse_string(super(EpubCoverHtml, self).get_content()) tree_root = tree.getroot() images = tree_root.xpath('//xhtml:img', namespaces={'xhtml': NAMESPACES['XHTML']}) images[0].set('src', self.image_name) images[0].set('alt', self.title) tree_str = etree.tostring(tree, pretty_print=True, encoding='utf-8', xml_declaration=True) return tree_str def __str__(self): return '<EpubCoverHtml:%s:%s>' % (self.id, self.file_name) class EpubNav(EpubHtml): """ Represents Navigation Document in the EPUB file. """ def __init__(self, uid='nav', file_name='nav.xhtml', media_type='application/xhtml+xml'): super(EpubNav, self).__init__(uid=uid, file_name=file_name, media_type=media_type) def is_chapter(self): """ Returns if this document is chapter or not. :Returns: Returns book value. """ return False def __str__(self): return '<EpubNav:%s:%s>' % (self.id, self.file_name) class EpubImage(EpubItem): """ Represents Image in the EPUB file. """ def __init__(self): super(EpubImage, self).__init__() def get_type(self): return ebooklib.ITEM_IMAGE def __str__(self): return '<EpubImage:%s:%s>' % (self.id, self.file_name) class EpubSMIL(EpubItem): def __init__(self, uid=None, file_name='', content=None): super(EpubSMIL, self).__init__(uid=uid, file_name=file_name, media_type='application/smil+xml', content=content) def get_type(self): return ebooklib.ITEM_SMIL def __str__(self): return '<EpubSMIL:%s:%s>' % (self.id, self.file_name) # EpubBook class EpubBook(object): def __init__(self): self.EPUB_VERSION = None self.reset() # we should have options here def reset(self): "Initialises all needed variables to default values" self.metadata = {} self.items = [] self.spine = [] self.guide = [] self.pages = [] self.toc = [] self.bindings = [] self.IDENTIFIER_ID = 'id' self.FOLDER_NAME = 'EPUB' self._id_html = 0 self._id_image = 0 self._id_static = 0 self.title = '' self.language = 'en' self.direction = None self.templates = { 'ncx': NCX_XML, 'nav': NAV_XML, 'chapter': CHAPTER_XML, 'cover': COVER_XML } self.add_metadata('OPF', 'generator', '', { 'name': 'generator', 'content': 'Ebook-lib %s' % '.'.join([str(s) for s in VERSION]) }) # default to using a randomly-unique identifier if one is not specified manually self.set_identifier(str(uuid.uuid4())) # custom prefixes and namespaces to be set to the content.opf doc self.prefixes = [] self.namespaces = {} def set_identifier(self, uid): """ Sets unique id for this epub :Args: - uid: Value of unique identifier for this book """ self.uid = uid self.set_unique_metadata('DC', 'identifier', self.uid, {'id': self.IDENTIFIER_ID}) def set_title(self, title): """ Set title. You can set multiple titles. :Args: - title: Title value """ self.title = title self.add_metadata('DC', 'title', self.title) def set_language(self, lang): """ Set language for this epub. You can set multiple languages. Specific items in the book can have different language settings. :Args: - lang: Language code """ self.language = lang self.add_metadata('DC', 'language', lang) def set_direction(self, direction): """ :Args: - direction: Options are "ltr", "rtl" and "default" """ self.direction = direction def set_cover(self, file_name, content, create_page=True): """ Set cover and create cover document if needed. :Args: - file_name: file name of the cover page - content: Content for the cover image - create_page: Should cover page be defined. Defined as bool value (optional). Default value is True. """ # as it is now, it can only be called once c0 = EpubCover(file_name=file_name) c0.content = content self.add_item(c0) if create_page: c1 = EpubCoverHtml(image_name=file_name) self.add_item(c1) self.add_metadata(None, 'meta', '', OrderedDict([('name', 'cover'), ('content', 'cover-img')])) def add_author(self, author, file_as=None, role=None, uid='creator'): "Add author for this document" self.add_metadata('DC', 'creator', author, {'id': uid}) if file_as: self.add_metadata(None, 'meta', file_as, {'refines': '#' + uid, 'property': 'file-as', 'scheme': 'marc:relators'}) if role: self.add_metadata(None, 'meta', role, {'refines': '#' + uid, 'property': 'role', 'scheme': 'marc:relators'}) def add_metadata(self, namespace, name, value, others=None): "Add metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace not in self.metadata: self.metadata[namespace] = {} if name not in self.metadata[namespace]: self.metadata[namespace][name] = [] self.metadata[namespace][name].append((value, others)) def get_metadata(self, namespace, name): "Retrieve metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] return self.metadata[namespace].get(name, []) def set_unique_metadata(self, namespace, name, value, others=None): "Add metadata if metadata with this identifier does not already exist, otherwise update existing metadata." if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace in self.metadata and name in self.metadata[namespace]: self.metadata[namespace][name] = [(value, others)] else: self.add_metadata(namespace, name, value, others) def add_item(self, item): """ Add additional item to the book. If not defined, media type and chapter id will be defined for the item. :Args: - item: Item instance """ if item.media_type == '': (has_guessed, media_type) = guess_type(item.get_name().lower()) if has_guessed: if media_type is not None: item.media_type = media_type else: item.media_type = has_guessed else: item.media_type = 'application/octet-stream' if not item.get_id(): # make chapter_, image_ and static_ configurable if isinstance(item, EpubHtml): item.id = 'chapter_%d' % self._id_html self._id_html += 1 # If there's a page list, append it to the book's page list self.pages += item.pages elif isinstance(item, EpubImage): item.id = 'image_%d' % self._id_image self._id_image += 1 else: item.id = 'static_%d' % self._id_image self._id_image += 1 item.book = self self.items.append(item) return item def get_item_with_id(self, uid): """ Returns item for defined UID. >>> book.get_item_with_id('image_001') :Args: - uid: UID for the item :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.id == uid: return item return None def get_item_with_href(self, href): """ Returns item for defined HREF. >>> book.get_item_with_href('EPUB/document.xhtml') :Args: - href: HREF for the item we are searching for :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.get_name() == href: return item return None def get_items(self): """ Returns all items attached to this book. :Returns: Returns all items as tuple. """ return (item for item in self.items) def get_items_of_type(self, item_type): """ Returns all items of specified type. >>> book.get_items_of_type(epub.ITEM_IMAGE) :Args: - item_type: Type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.get_type() == item_type) def get_items_of_media_type(self, media_type): """ Returns all items of specified media type. :Args: - media_type: Media type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.media_type == media_type) def set_template(self, name, value): """ Defines templates which are used to generate certain types of pages. When defining new value for the template we have to use content of type 'str' (Python 2) or 'bytes' (Python 3). At the moment we use these templates: - ncx - nav - chapter - cover :Args: - name: Name for the template - value: Content for the template """ self.templates[name] = value def get_template(self, name): """ Returns value for the template. :Args: - name: template name :Returns: Value of the template. """ return self.templates.get(name) def add_prefix(self, name, uri): """ Appends custom prefix to be added to the content.opf document >>> epub_book.add_prefix('bkterms', 'http://booktype.org/') :Args: - name: namespave name - uri: URI for the namespace """ self.prefixes.append('%s: %s' % (name, uri)) class EpubWriter(object): DEFAULT_OPTIONS = { 'epub2_guide': True, 'epub3_landmark': True, 'epub3_pages': True, 'landmark_title': 'Guide', 'pages_title': 'Pages', 'spine_direction': True, 'package_direction': False, 'play_order': { 'enabled': False, 'start_from': 1 } } def __init__(self, name, book, options=None): self.file_name = name self.book = book self.options = dict(self.DEFAULT_OPTIONS) if options: self.options.update(options) self._init_play_order() def _init_play_order(self): self._play_order = { 'enabled': False, 'start_from': 1 } try: self._play_order['enabled'] = self.options['play_order']['enabled'] self._play_order['start_from'] = self.options['play_order']['start_from'] except KeyError: pass def process(self): # should cache this html parsing so we don't do it for every plugin for plg in self.options.get('plugins', []): if hasattr(plg, 'before_write'): plg.before_write(self.book) for item in self.book.get_items(): if isinstance(item, EpubHtml): for plg in self.options.get('plugins', []): if hasattr(plg, 'html_before_write'): plg.html_before_write(self.book, item) def _write_container(self): container_xml = CONTAINER_XML % {'folder_name': self.book.FOLDER_NAME} self.out.writestr(CONTAINER_PATH, container_xml) def _write_opf_metadata(self, root): # This is really not needed # problem is uppercase/lowercase # for ns_name, values in six.iteritems(self.book.metadata): # if ns_name: # for n_id, ns_url in six.iteritems(NAMESPACES): # if ns_name == ns_url: # nsmap[n_id.lower()] = NAMESPACES[n_id] nsmap = {'dc': NAMESPACES['DC'], 'opf': NAMESPACES['OPF']} nsmap.update(self.book.namespaces) metadata = etree.SubElement(root, 'metadata', nsmap=nsmap) el = etree.SubElement(metadata, 'meta', {'property': 'dcterms:modified'}) if 'mtime' in self.options: mtime = self.options['mtime'] else: import datetime mtime = datetime.datetime.now() el.text = mtime.strftime('%Y-%m-%dT%H:%M:%SZ') for ns_name, values in six.iteritems(self.book.metadata): if ns_name == NAMESPACES['OPF']: for values in values.values(): for v in values: if 'property' in v[1] and v[1]['property'] == 'dcterms:modified': continue try: el = etree.SubElement(metadata, 'meta', v[1]) if v[0]: el.text = v[0] except ValueError: logging.error('Could not create metadata.') else: for name, values in six.iteritems(values): for v in values: try: if ns_name: el = etree.SubElement(metadata, '{%s}%s' % (ns_name, name), v[1]) else: el = etree.SubElement(metadata, '%s' % name, v[1]) el.text = v[0] except ValueError: logging.error('Could not create metadata "{}".'.format(name)) def _write_opf_manifest(self, root): manifest = etree.SubElement(root, 'manifest') _ncx_id = None # mathml, scripted, svg, remote-resources, and switch # nav # cover-image for item in self.book.get_items(): if not item.manifest: continue if isinstance(item, EpubNav): etree.SubElement(manifest, 'item', {'href': item.get_name(), 'id': item.id, 'media-type': item.media_type, 'properties': 'nav'}) elif isinstance(item, EpubNcx): _ncx_id = item.id etree.SubElement(manifest, 'item', {'href': item.file_name, 'id': item.id, 'media-type': item.media_type}) elif isinstance(item, EpubCover): etree.SubElement(manifest, 'item', {'href': item.file_name, 'id': item.id, 'media-type': item.media_type, 'properties': 'cover-image'}) else: opts = {'href': item.file_name, 'id': item.id, 'media-type': item.media_type} if hasattr(item, 'properties') and len(item.properties) > 0: opts['properties'] = ' '.join(item.properties) if hasattr(item, 'media_overlay') and item.media_overlay is not None: opts['media-overlay'] = item.media_overlay if hasattr(item, 'media_duration') and item.media_duration is not None: opts['duration'] = item.media_duration etree.SubElement(manifest, 'item', opts) return _ncx_id def _write_opf_spine(self, root, ncx_id): spine_attributes = {'toc': ncx_id or 'ncx'} if self.book.direction and self.options['spine_direction']: spine_attributes['page-progression-direction'] = self.book.direction spine = etree.SubElement(root, 'spine', spine_attributes) for _item in self.book.spine: # this is for now # later we should be able to fetch things from tuple is_linear = True if isinstance(_item, tuple): item = _item[0] if len(_item) > 1: if _item[1] == 'no': is_linear = False else: item = _item if isinstance(item, EpubHtml): opts = {'idref': item.get_id()} if not item.is_linear or not is_linear: opts['linear'] = 'no' elif isinstance(item, EpubItem): opts = {'idref': item.get_id()} if not item.is_linear or not is_linear: opts['linear'] = 'no' else: opts = {'idref': item} try: itm = self.book.get_item_with_id(item) if not itm.is_linear or not is_linear: opts['linear'] = 'no' except: pass etree.SubElement(spine, 'itemref', opts) def _write_opf_guide(self, root): # - http://www.idpf.org/epub/20/spec/OPF_2.0.1_draft.htm#Section2.6 if len(self.book.guide) > 0 and self.options.get('epub2_guide'): guide = etree.SubElement(root, 'guide', {}) for item in self.book.guide: if 'item' in item: chap = item.get('item') if chap: _href = chap.file_name _title = chap.title else: _href = item.get('href', '') _title = item.get('title', '') if _title is None: _title = '' ref = etree.SubElement(guide, 'reference', {'type': item.get('type', ''), 'title': _title, 'href': _href}) def _write_opf_bindings(self, root): if len(self.book.bindings) > 0: bindings = etree.SubElement(root, 'bindings', {}) for item in self.book.bindings: etree.SubElement(bindings, 'mediaType', item) def _write_opf_file(self, root): tree_str = etree.tostring(root, pretty_print=True, encoding='utf-8', xml_declaration=True) self.out.writestr('%s/content.opf' % self.book.FOLDER_NAME, tree_str) def _write_opf(self): package_attributes = {'xmlns': NAMESPACES['OPF'], 'unique-identifier': self.book.IDENTIFIER_ID, 'version': '3.0'} if self.book.direction and self.options['package_direction']: package_attributes['dir'] = self.book.direction root = etree.Element('package', package_attributes) prefixes = ['rendition: http://www.idpf.org/vocab/rendition/#'] + self.book.prefixes root.attrib['prefix'] = ' '.join(prefixes) # METADATA self._write_opf_metadata(root) # MANIFEST _ncx_id = self._write_opf_manifest(root) # SPINE self._write_opf_spine(root, _ncx_id) # GUIDE self._write_opf_guide(root) # BINDINGS self._write_opf_bindings(root) # WRITE FILE self._write_opf_file(root) def _get_nav(self, item): # just a basic navigation for now nav_xml = parse_string(self.book.get_template('nav')) root = nav_xml.getroot() root.set('lang', self.book.language) root.attrib['{%s}lang' % NAMESPACES['XML']] = self.book.language nav_dir_name = os.path.dirname(item.file_name) head = etree.SubElement(root, 'head') title = etree.SubElement(head, 'title') title.text = self.book.title # for now this just handles css files and ignores others for _link in item.links: _lnk = etree.SubElement(head, 'link', { 'href': _link.get('href', ''), 'rel': 'stylesheet', 'type': 'text/css' }) body = etree.SubElement(root, 'body') nav = etree.SubElement(body, 'nav', { '{%s}type' % NAMESPACES['EPUB']: 'toc', 'id': 'id', 'role': 'doc-toc', }) content_title = etree.SubElement(nav, 'h2') content_title.text = self.book.title def _create_section(itm, items): ol = etree.SubElement(itm, 'ol') for item in items: if isinstance(item, tuple) or isinstance(item, list): li = etree.SubElement(ol, 'li') if isinstance(item[0], EpubHtml): a = etree.SubElement(li, 'a', {'href': os.path.relpath(item[0].file_name, nav_dir_name)}) elif isinstance(item[0], Section) and item[0].href != '': a = etree.SubElement(li, 'a', {'href': os.path.relpath(item[0].href, nav_dir_name)}) elif isinstance(item[0], Link): a = etree.SubElement(li, 'a', {'href': os.path.relpath(item[0].href, nav_dir_name)}) else: a = etree.SubElement(li, 'span') a.text = item[0].title _create_section(li, item[1]) elif isinstance(item, Link): li = etree.SubElement(ol, 'li') a = etree.SubElement(li, 'a', {'href': os.path.relpath(item.href, nav_dir_name)}) a.text = item.title elif isinstance(item, EpubHtml): li = etree.SubElement(ol, 'li') a = etree.SubElement(li, 'a', {'href': os.path.relpath(item.file_name, nav_dir_name)}) a.text = item.title _create_section(nav, self.book.toc) # LANDMARKS / GUIDE # - http://www.idpf.org/epub/30/spec/epub30-contentdocs.html#sec-xhtml-nav-def-types-landmarks if len(self.book.guide) > 0 and self.options.get('epub3_landmark'): # Epub2 guide types do not map completely to epub3 landmark types. guide_to_landscape_map = { 'notes': 'rearnotes', 'text': 'bodymatter' } guide_nav = etree.SubElement(body, 'nav', {'{%s}type' % NAMESPACES['EPUB']: 'landmarks'}) guide_content_title = etree.SubElement(guide_nav, 'h2') guide_content_title.text = self.options.get('landmark_title', 'Guide') guild_ol = etree.SubElement(guide_nav, 'ol') for elem in self.book.guide: li_item = etree.SubElement(guild_ol, 'li') if 'item' in elem: chap = elem.get('item', None) if chap: _href = chap.file_name _title = chap.title else: _href = elem.get('href', '') _title = elem.get('title', '') guide_type = elem.get('type', '') a_item = etree.SubElement(li_item, 'a', { '{%s}type' % NAMESPACES['EPUB']: guide_to_landscape_map.get(guide_type, guide_type), 'href': os.path.relpath(_href, nav_dir_name) }) a_item.text = _title # PAGE-LIST if self.options.get('epub3_pages'): inserted_pages = get_pages_for_items([item for item in self.book.get_items_of_type(ebooklib.ITEM_DOCUMENT) \ if not isinstance(item, EpubNav)]) if len(inserted_pages) > 0: pagelist_nav = etree.SubElement( body, 'nav', { '{%s}type' % NAMESPACES['EPUB']: 'page-list', 'id': 'pages', 'hidden': 'hidden', } ) pagelist_content_title = etree.SubElement(pagelist_nav, 'h2') pagelist_content_title.text = self.options.get( 'pages_title', 'Pages' ) pages_ol = etree.SubElement(pagelist_nav, 'ol') for filename, pageref, label in inserted_pages: li_item = etree.SubElement(pages_ol, 'li') _href = u'{}#{}'.format(filename, pageref) _title = label a_item = etree.SubElement(li_item, 'a', { 'href': os.path.relpath(_href, nav_dir_name), }) a_item.text = _title tree_str = etree.tostring(nav_xml, pretty_print=True, encoding='utf-8', xml_declaration=True) return tree_str def _get_ncx(self): # we should be able to setup language for NCX as also ncx = parse_string(self.book.get_template('ncx')) root = ncx.getroot() head = etree.SubElement(root, 'head') # get this id uid = etree.SubElement(head, 'meta', {'content': self.book.uid, 'name': 'dtb:uid'}) uid = etree.SubElement(head, 'meta', {'content': '0', 'name': 'dtb:depth'}) uid = etree.SubElement(head, 'meta', {'content': '0', 'name': 'dtb:totalPageCount'}) uid = etree.SubElement(head, 'meta', {'content': '0', 'name': 'dtb:maxPageNumber'}) doc_title = etree.SubElement(root, 'docTitle') title = etree.SubElement(doc_title, 'text') title.text = self.book.title # doc_author = etree.SubElement(root, 'docAuthor') # author = etree.SubElement(doc_author, 'text') # author.text = 'Name of the person' # For now just make a very simple navMap nav_map = etree.SubElement(root, 'navMap') def _add_play_order(nav_point): nav_point.set('playOrder', str(self._play_order['start_from'])) self._play_order['start_from'] += 1 def _create_section(itm, items, uid): for item in items: if isinstance(item, tuple) or isinstance(item, list): section, subsection = item[0], item[1] np = etree.SubElement(itm, 'navPoint', { 'id': section.get_id() if isinstance(section, EpubHtml) else 'sep_%d' % uid }) if self._play_order['enabled']: _add_play_order(np) nl = etree.SubElement(np, 'navLabel') nt = etree.SubElement(nl, 'text') nt.text = section.title # CAN NOT HAVE EMPTY SRC HERE href = '' if isinstance(section, EpubHtml): href = section.file_name elif isinstance(section, Section) and section.href != '': href = section.href elif isinstance(section, Link): href = section.href nc = etree.SubElement(np, 'content', {'src': href}) uid = _create_section(np, subsection, uid + 1) elif isinstance(item, Link): _parent = itm _content = _parent.find('content') if _content is not None: if _content.get('src') == '': _content.set('src', item.href) np = etree.SubElement(itm, 'navPoint', {'id': item.uid}) if self._play_order['enabled']: _add_play_order(np) nl = etree.SubElement(np, 'navLabel') nt = etree.SubElement(nl, 'text') nt.text = item.title nc = etree.SubElement(np, 'content', {'src': item.href}) elif isinstance(item, EpubHtml): _parent = itm _content = _parent.find('content') if _content is not None: if _content.get('src') == '': _content.set('src', item.file_name) np = etree.SubElement(itm, 'navPoint', {'id': item.get_id()}) if self._play_order['enabled']: _add_play_order(np) nl = etree.SubElement(np, 'navLabel') nt = etree.SubElement(nl, 'text') nt.text = item.title nc = etree.SubElement(np, 'content', {'src': item.file_name}) return uid _create_section(nav_map, self.book.toc, 0) tree_str = etree.tostring(root, pretty_print=True, encoding='utf-8', xml_declaration=True) return tree_str def _write_items(self): for item in self.book.get_items(): if isinstance(item, EpubNcx): self.out.writestr('%s/%s' % (self.book.FOLDER_NAME, item.file_name), self._get_ncx()) elif isinstance(item, EpubNav): self.out.writestr('%s/%s' % (self.book.FOLDER_NAME, item.file_name), self._get_nav(item)) elif item.manifest: self.out.writestr('%s/%s' % (self.book.FOLDER_NAME, item.file_name), item.get_content()) else: self.out.writestr('%s' % item.file_name, item.get_content()) def write(self): # check for the option allowZip64 self.out = zipfile.ZipFile(self.file_name, 'w', zipfile.ZIP_DEFLATED) self.out.writestr('mimetype', 'application/epub+zip', compress_type=zipfile.ZIP_STORED) self._write_container() self._write_opf() self._write_items() self.out.close() class EpubReader(object): DEFAULT_OPTIONS = {} def __init__(self, epub_file_name, options=None): self.file_name = epub_file_name self.book = EpubBook() self.zf = None self.opf_file = '' self.opf_dir = '' self.options = dict(self.DEFAULT_OPTIONS) if options: self.options.update(options) def process(self): # should cache this html parsing so we don't do it for every plugin for plg in self.options.get('plugins', []): if hasattr(plg, 'after_read'): plg.after_read(self.book) for item in self.book.get_items(): if isinstance(item, EpubHtml): for plg in self.options.get('plugins', []): if hasattr(plg, 'html_after_read'): plg.html_after_read(self.book, item) def load(self): self._load() return self.book def read_file(self, name): # Raises KeyError name = zip_path.normpath(name) return self.zf.read(name) def _load_container(self): meta_inf = self.read_file('META-INF/container.xml') tree = parse_string(meta_inf) for root_file in tree.findall('//xmlns:rootfile[@media-type]', namespaces={'xmlns': NAMESPACES['CONTAINERNS']}): if root_file.get('media-type') == 'application/oebps-package+xml': self.opf_file = root_file.get('full-path') self.opf_dir = zip_path.dirname(self.opf_file) def _load_metadata(self): container_root = self.container.getroot() # get epub version self.book.version = container_root.get('version', None) # get unique-identifier if container_root.get('unique-identifier', None): self.book.IDENTIFIER_ID = container_root.get('unique-identifier') # get xml:lang # get metadata metadata = self.container.find('{%s}%s' % (NAMESPACES['OPF'], 'metadata')) nsmap = metadata.nsmap nstags = dict((k, '{%s}' % v) for k, v in six.iteritems(nsmap)) default_ns = nstags.get(None, '') nsdict = dict((v, {}) for v in nsmap.values()) def add_item(ns, tag, value, extra): if ns not in nsdict: nsdict[ns] = {} values = nsdict[ns].setdefault(tag, []) values.append((value, extra)) for t in metadata: if not etree.iselement(t) or t.tag is etree.Comment: continue if t.tag == default_ns + 'meta': name = t.get('name') others = dict((k, v) for k, v in t.items()) if name and ':' in name: prefix, name = name.split(':', 1) else: prefix = None add_item(t.nsmap.get(prefix, prefix), name, t.text, others) else: tag = t.tag[t.tag.rfind('}') + 1:] if (t.prefix and t.prefix.lower() == 'dc') and tag == 'identifier': _id = t.get('id', None) if _id: self.book.IDENTIFIER_ID = _id others = dict((k, v) for k, v in t.items()) add_item(t.nsmap[t.prefix], tag, t.text, others) self.book.metadata = nsdict titles = self.book.get_metadata('DC', 'title') if len(titles) > 0: self.book.title = titles[0][0] for value, others in self.book.get_metadata('DC', 'identifier'): if others.get('id') == self.book.IDENTIFIER_ID: self.book.uid = value def _load_manifest(self): for r in self.container.find('{%s}%s' % (NAMESPACES['OPF'], 'manifest')): if r is not None and r.tag != '{%s}item' % NAMESPACES['OPF']: continue media_type = r.get('media-type') _properties = r.get('properties', '') if _properties: properties = _properties.split(' ') else: properties = [] # people use wrong content types if media_type == 'image/jpg': media_type = 'image/jpeg' if media_type == 'application/x-dtbncx+xml': ei = EpubNcx(uid=r.get('id'), file_name=unquote(r.get('href'))) ei.content = self.read_file(zip_path.join(self.opf_dir, ei.file_name)) if media_type == 'application/smil+xml': ei = EpubSMIL(uid=r.get('id'), file_name=unquote(r.get('href'))) ei.content = self.read_file(zip_path.join(self.opf_dir, ei.file_name)) elif media_type == 'application/xhtml+xml': if 'nav' in properties: ei = EpubNav(uid=r.get('id'), file_name=unquote(r.get('href'))) ei.content = self.read_file(zip_path.join(self.opf_dir, r.get('href'))) elif 'cover' in properties: ei = EpubCoverHtml() ei.content = self.read_file(zip_path.join(self.opf_dir, unquote(r.get('href')))) else: ei = EpubHtml() ei.id = r.get('id') ei.file_name = unquote(r.get('href')) ei.media_type = media_type ei.media_overlay = r.get('media-overlay', None) ei.media_duration = r.get('duration', None) ei.content = self.read_file(zip_path.join(self.opf_dir, ei.get_name())) ei.properties = properties elif media_type in IMAGE_MEDIA_TYPES: if 'cover-image' in properties: ei = EpubCover(uid=r.get('id'), file_name=unquote(r.get('href'))) ei.media_type = media_type ei.content = self.read_file(zip_path.join(self.opf_dir, ei.get_name())) else: ei = EpubImage() ei.id = r.get('id') ei.file_name = unquote(r.get('href')) ei.media_type = media_type ei.content = self.read_file(zip_path.join(self.opf_dir, ei.get_name())) else: # different types ei = EpubItem() ei.id = r.get('id') ei.file_name = unquote(r.get('href')) ei.media_type = media_type ei.content = self.read_file(zip_path.join(self.opf_dir, ei.get_name())) self.book.add_item(ei) def _parse_ncx(self, data): tree = parse_string(data) tree_root = tree.getroot() nav_map = tree_root.find('{%s}navMap' % NAMESPACES['DAISY']) def _get_children(elems, n, nid): label, content = '', '' children = [] for a in elems.getchildren(): if a.tag == '{%s}navLabel' % NAMESPACES['DAISY']: label = a.getchildren()[0].text if a.tag == '{%s}content' % NAMESPACES['DAISY']: content = a.get('src', '') if a.tag == '{%s}navPoint' % NAMESPACES['DAISY']: children.append(_get_children(a, n + 1, a.get('id', ''))) if len(children) > 0: if n == 0: return children return (Section(label, href=content), children) else: return Link(content, label, nid) self.book.toc = _get_children(nav_map, 0, '') def _parse_nav(self, data, base_path, navtype='toc'): html_node = parse_html_string(data) if navtype == 'toc': # parsing the table of contents nav_node = html_node.xpath("//nav[@*='toc']")[0] else: # parsing the list of pages _page_list = html_node.xpath("//nav[@*='page-list']") if len(_page_list) == 0: return nav_node = _page_list[0] def parse_list(list_node): items = [] for item_node in list_node.findall('li'): sublist_node = item_node.find('ol') link_node = item_node.find('a') if sublist_node is not None: title = item_node[0].text children = parse_list(sublist_node) if link_node is not None: href = zip_path.normpath(zip_path.join(base_path, link_node.get('href'))) items.append((Section(title, href=href), children)) else: items.append((Section(title), children)) elif link_node is not None: title = link_node.text href = zip_path.normpath(zip_path.join(base_path, link_node.get('href'))) items.append(Link(href, title)) return items if navtype == 'toc': self.book.toc = parse_list(nav_node.find('ol')) elif nav_node is not None: # generate the pages list if there is one self.book.pages = parse_list(nav_node.find('ol')) # generate the per-file pages lists # because of the order of parsing the files, this can't be done # when building the EpubHtml objects htmlfiles = dict() for htmlfile in self.book.items: if isinstance(htmlfile, EpubHtml): htmlfiles[htmlfile.file_name] = htmlfile for page in self.book.pages: try: (filename, idref) = page.href.split('#') except ValueError: filename = page.href if filename in htmlfiles: htmlfiles[filename].pages.append(page) def _load_spine(self): spine = self.container.find('{%s}%s' % (NAMESPACES['OPF'], 'spine')) self.book.spine = [(t.get('idref'), t.get('linear', 'yes')) for t in spine] toc = spine.get('toc', '') self.book.set_direction(spine.get('page-progression-direction', None)) # should read ncx or nav file if toc: try: ncxFile = self.read_file(zip_path.join(self.opf_dir, self.book.get_item_with_id(toc).get_name())) except KeyError: raise EpubException(-1, 'Can not find ncx file.') self._parse_ncx(ncxFile) def _load_guide(self): guide = self.container.find('{%s}%s' % (NAMESPACES['OPF'], 'guide')) if guide is not None: self.book.guide = [{'href': t.get('href'), 'title': t.get('title'), 'type': t.get('type')} for t in guide] def _load_opf_file(self): try: s = self.read_file(self.opf_file) except KeyError: raise EpubException(-1, 'Can not find container file') self.container = parse_string(s) self._load_metadata() self._load_manifest() self._load_spine() self._load_guide() # read nav file if found # nav_item = next((item for item in self.book.items if isinstance(item, EpubNav)), None) if nav_item: if not self.book.toc: self._parse_nav( nav_item.content, zip_path.dirname(nav_item.file_name), navtype='toc' ) self._parse_nav( nav_item.content, zip_path.dirname(nav_item.file_name), navtype='pages' ) def _load(self): try: self.zf = zipfile.ZipFile(self.file_name, 'r', compression=zipfile.ZIP_DEFLATED, allowZip64=True) except zipfile.BadZipfile as bz: raise EpubException(0, 'Bad Zip file') except zipfile.LargeZipFile as bz: raise EpubException(1, 'Large Zip file') # 1st check metadata self._load_container() self._load_opf_file() self.zf.close() # WRITE def write_epub(name, book, options=None): """ Creates epub file with the content defined in EpubBook. >>> ebooklib.write_epub('book.epub', book) :Args: - name: file name for the output file - book: instance of EpubBook - options: extra opions as dictionary (optional) """ epub = EpubWriter(name, book, options) epub.process() try: epub.write() except IOError: pass # READ
aerkalov/ebooklib
ebooklib/epub.py
EpubItem.get_type
python
def get_type(self): _, ext = zip_path.splitext(self.get_name()) ext = ext.lower() for uid, ext_list in six.iteritems(ebooklib.EXTENSIONS): if ext in ext_list: return uid return ebooklib.ITEM_UNKNOWN
Guess type according to the file extension. Might not be the best way how to do it, but it works for now. Items can be of type: - ITEM_UNKNOWN = 0 - ITEM_IMAGE = 1 - ITEM_STYLE = 2 - ITEM_SCRIPT = 3 - ITEM_NAVIGATION = 4 - ITEM_VECTOR = 5 - ITEM_FONT = 6 - ITEM_VIDEO = 7 - ITEM_AUDIO = 8 - ITEM_DOCUMENT = 9 - ITEM_COVER = 10 We map type according to the extensions which are defined in ebooklib.EXTENSIONS. :Returns: Returns type of the item as number.
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L158-L187
[ "def get_name(self):\n \"\"\"\n Returns name for this item. By default it is always file name but it does not have to be.\n\n :Returns:\n Returns file name for this item.\n \"\"\"\n return self.file_name\n" ]
class EpubItem(object): """ Base class for the items in a book. """ def __init__(self, uid=None, file_name='', media_type='', content=six.b(''), manifest=True): """ :Args: - uid: Unique identifier for this item (optional) - file_name: File name for this item (optional) - media_type: Media type for this item (optional) - content: Content for this item (optional) - manifest: Manifest for this item (optional) """ self.id = uid self.file_name = file_name self.media_type = media_type self.content = content self.is_linear = True self.manifest = manifest self.book = None def get_id(self): """ Returns unique identifier for this item. :Returns: Returns uid number as string. """ return self.id def get_name(self): """ Returns name for this item. By default it is always file name but it does not have to be. :Returns: Returns file name for this item. """ return self.file_name def get_content(self, default=six.b('')): """ Returns content of the item. Content should be of type 'str' (Python 2) or 'bytes' (Python 3) :Args: - default: Default value for the content if it is not already defined. :Returns: Returns content of the item. """ return self.content or default def set_content(self, content): """ Sets content value for this item. :Args: - content: Content value """ self.content = content def __str__(self): return '<EpubItem:%s>' % self.id
aerkalov/ebooklib
ebooklib/epub.py
EpubHtml.add_link
python
def add_link(self, **kwgs): self.links.append(kwgs) if kwgs.get('type') == 'text/javascript': if 'scripted' not in self.properties: self.properties.append('scripted')
Add additional link to the document. Links will be embeded only inside of this document. >>> add_link(href='styles.css', rel='stylesheet', type='text/css')
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L299-L308
null
class EpubHtml(EpubItem): """ Represents HTML document in the EPUB file. """ _template_name = 'chapter' def __init__(self, uid=None, file_name='', media_type='', content=None, title='', lang=None, direction=None, media_overlay=None, media_duration=None): super(EpubHtml, self).__init__(uid, file_name, media_type, content) self.title = title self.lang = lang self.direction = direction self.media_overlay = media_overlay self.media_duration = media_duration self.links = [] self.properties = [] self.pages = [] def is_chapter(self): """ Returns if this document is chapter or not. :Returns: Returns book value. """ return True def get_type(self): """ Always returns ebooklib.ITEM_DOCUMENT as type of this document. :Returns: Always returns ebooklib.ITEM_DOCUMENT """ return ebooklib.ITEM_DOCUMENT def set_language(self, lang): """ Sets language for this book item. By default it will use language of the book but it can be overwritten with this call. """ self.lang = lang def get_language(self): """ Get language code for this book item. Language of the book item can be different from the language settings defined globaly for book. :Returns: As string returns language code. """ return self.lang def get_links(self): """ Returns list of additional links defined for this document. :Returns: As tuple return list of links. """ return (link for link in self.links) def get_links_of_type(self, link_type): """ Returns list of additional links of specific type. :Returns: As tuple returns list of links. """ return (link for link in self.links if link.get('type', '') == link_type) def add_item(self, item): """ Add other item to this document. It will create additional links according to the item type. :Args: - item: item we want to add defined as instance of EpubItem """ if item.get_type() == ebooklib.ITEM_STYLE: self.add_link(href=item.get_name(), rel='stylesheet', type='text/css') if item.get_type() == ebooklib.ITEM_SCRIPT: self.add_link(src=item.get_name(), type='text/javascript') def get_body_content(self): """ Returns content of BODY element for this HTML document. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Returns: Returns content of this document. """ try: html_tree = parse_html_string(self.content) except: return '' html_root = html_tree.getroottree() if len(html_root.find('body')) != 0: body = html_tree.find('body') tree_str = etree.tostring(body, pretty_print=True, encoding='utf-8', xml_declaration=False) # this is so stupid if tree_str.startswith(six.b('<body>')): n = tree_str.rindex(six.b('</body>')) return tree_str[6:n] return tree_str return '' def get_content(self, default=None): """ Returns content for this document as HTML string. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Args: - default: Default value for the content if it is not defined. :Returns: Returns content of this document. """ tree = parse_string(self.book.get_template(self._template_name)) tree_root = tree.getroot() tree_root.set('lang', self.lang or self.book.language) tree_root.attrib['{%s}lang' % NAMESPACES['XML']] = self.lang or self.book.language # add to the head also # <meta charset="utf-8" /> try: html_tree = parse_html_string(self.content) except: return '' html_root = html_tree.getroottree() # create and populate head _head = etree.SubElement(tree_root, 'head') if self.title != '': _title = etree.SubElement(_head, 'title') _title.text = self.title for lnk in self.links: if lnk.get('type') == 'text/javascript': _lnk = etree.SubElement(_head, 'script', lnk) # force <script></script> _lnk.text = '' else: _lnk = etree.SubElement(_head, 'link', lnk) # this should not be like this # head = html_root.find('head') # if head is not None: # for i in head.getchildren(): # if i.tag == 'title' and self.title != '': # continue # _head.append(i) # create and populate body _body = etree.SubElement(tree_root, 'body') if self.direction: _body.set('dir', self.direction) tree_root.set('dir', self.direction) body = html_tree.find('body') if body is not None: for i in body.getchildren(): _body.append(i) tree_str = etree.tostring(tree, pretty_print=True, encoding='utf-8', xml_declaration=True) return tree_str def __str__(self): return '<EpubHtml:%s:%s>' % (self.id, self.file_name)
aerkalov/ebooklib
ebooklib/epub.py
EpubHtml.get_links_of_type
python
def get_links_of_type(self, link_type): return (link for link in self.links if link.get('type', '') == link_type)
Returns list of additional links of specific type. :Returns: As tuple returns list of links.
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L319-L326
null
class EpubHtml(EpubItem): """ Represents HTML document in the EPUB file. """ _template_name = 'chapter' def __init__(self, uid=None, file_name='', media_type='', content=None, title='', lang=None, direction=None, media_overlay=None, media_duration=None): super(EpubHtml, self).__init__(uid, file_name, media_type, content) self.title = title self.lang = lang self.direction = direction self.media_overlay = media_overlay self.media_duration = media_duration self.links = [] self.properties = [] self.pages = [] def is_chapter(self): """ Returns if this document is chapter or not. :Returns: Returns book value. """ return True def get_type(self): """ Always returns ebooklib.ITEM_DOCUMENT as type of this document. :Returns: Always returns ebooklib.ITEM_DOCUMENT """ return ebooklib.ITEM_DOCUMENT def set_language(self, lang): """ Sets language for this book item. By default it will use language of the book but it can be overwritten with this call. """ self.lang = lang def get_language(self): """ Get language code for this book item. Language of the book item can be different from the language settings defined globaly for book. :Returns: As string returns language code. """ return self.lang def add_link(self, **kwgs): """ Add additional link to the document. Links will be embeded only inside of this document. >>> add_link(href='styles.css', rel='stylesheet', type='text/css') """ self.links.append(kwgs) if kwgs.get('type') == 'text/javascript': if 'scripted' not in self.properties: self.properties.append('scripted') def get_links(self): """ Returns list of additional links defined for this document. :Returns: As tuple return list of links. """ return (link for link in self.links) def add_item(self, item): """ Add other item to this document. It will create additional links according to the item type. :Args: - item: item we want to add defined as instance of EpubItem """ if item.get_type() == ebooklib.ITEM_STYLE: self.add_link(href=item.get_name(), rel='stylesheet', type='text/css') if item.get_type() == ebooklib.ITEM_SCRIPT: self.add_link(src=item.get_name(), type='text/javascript') def get_body_content(self): """ Returns content of BODY element for this HTML document. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Returns: Returns content of this document. """ try: html_tree = parse_html_string(self.content) except: return '' html_root = html_tree.getroottree() if len(html_root.find('body')) != 0: body = html_tree.find('body') tree_str = etree.tostring(body, pretty_print=True, encoding='utf-8', xml_declaration=False) # this is so stupid if tree_str.startswith(six.b('<body>')): n = tree_str.rindex(six.b('</body>')) return tree_str[6:n] return tree_str return '' def get_content(self, default=None): """ Returns content for this document as HTML string. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Args: - default: Default value for the content if it is not defined. :Returns: Returns content of this document. """ tree = parse_string(self.book.get_template(self._template_name)) tree_root = tree.getroot() tree_root.set('lang', self.lang or self.book.language) tree_root.attrib['{%s}lang' % NAMESPACES['XML']] = self.lang or self.book.language # add to the head also # <meta charset="utf-8" /> try: html_tree = parse_html_string(self.content) except: return '' html_root = html_tree.getroottree() # create and populate head _head = etree.SubElement(tree_root, 'head') if self.title != '': _title = etree.SubElement(_head, 'title') _title.text = self.title for lnk in self.links: if lnk.get('type') == 'text/javascript': _lnk = etree.SubElement(_head, 'script', lnk) # force <script></script> _lnk.text = '' else: _lnk = etree.SubElement(_head, 'link', lnk) # this should not be like this # head = html_root.find('head') # if head is not None: # for i in head.getchildren(): # if i.tag == 'title' and self.title != '': # continue # _head.append(i) # create and populate body _body = etree.SubElement(tree_root, 'body') if self.direction: _body.set('dir', self.direction) tree_root.set('dir', self.direction) body = html_tree.find('body') if body is not None: for i in body.getchildren(): _body.append(i) tree_str = etree.tostring(tree, pretty_print=True, encoding='utf-8', xml_declaration=True) return tree_str def __str__(self): return '<EpubHtml:%s:%s>' % (self.id, self.file_name)
aerkalov/ebooklib
ebooklib/epub.py
EpubHtml.add_item
python
def add_item(self, item): if item.get_type() == ebooklib.ITEM_STYLE: self.add_link(href=item.get_name(), rel='stylesheet', type='text/css') if item.get_type() == ebooklib.ITEM_SCRIPT: self.add_link(src=item.get_name(), type='text/javascript')
Add other item to this document. It will create additional links according to the item type. :Args: - item: item we want to add defined as instance of EpubItem
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L328-L339
[ "def get_type(self):\n \"\"\"\n Guess type according to the file extension. Might not be the best way how to do it, but it works for now.\n\n Items can be of type:\n - ITEM_UNKNOWN = 0\n - ITEM_IMAGE = 1\n - ITEM_STYLE = 2\n - ITEM_SCRIPT = 3\n - ITEM_NAVIGATION = 4\n - ITEM_VECTOR = 5\n - ITEM_FONT = 6\n - ITEM_VIDEO = 7\n - ITEM_AUDIO = 8\n - ITEM_DOCUMENT = 9\n - ITEM_COVER = 10\n\n We map type according to the extensions which are defined in ebooklib.EXTENSIONS.\n\n :Returns:\n Returns type of the item as number.\n \"\"\"\n _, ext = zip_path.splitext(self.get_name())\n ext = ext.lower()\n\n for uid, ext_list in six.iteritems(ebooklib.EXTENSIONS):\n if ext in ext_list:\n return uid\n\n return ebooklib.ITEM_UNKNOWN\n", "def add_link(self, **kwgs):\n \"\"\"\n Add additional link to the document. Links will be embeded only inside of this document.\n\n >>> add_link(href='styles.css', rel='stylesheet', type='text/css')\n \"\"\"\n self.links.append(kwgs)\n if kwgs.get('type') == 'text/javascript':\n if 'scripted' not in self.properties:\n self.properties.append('scripted')\n" ]
class EpubHtml(EpubItem): """ Represents HTML document in the EPUB file. """ _template_name = 'chapter' def __init__(self, uid=None, file_name='', media_type='', content=None, title='', lang=None, direction=None, media_overlay=None, media_duration=None): super(EpubHtml, self).__init__(uid, file_name, media_type, content) self.title = title self.lang = lang self.direction = direction self.media_overlay = media_overlay self.media_duration = media_duration self.links = [] self.properties = [] self.pages = [] def is_chapter(self): """ Returns if this document is chapter or not. :Returns: Returns book value. """ return True def get_type(self): """ Always returns ebooklib.ITEM_DOCUMENT as type of this document. :Returns: Always returns ebooklib.ITEM_DOCUMENT """ return ebooklib.ITEM_DOCUMENT def set_language(self, lang): """ Sets language for this book item. By default it will use language of the book but it can be overwritten with this call. """ self.lang = lang def get_language(self): """ Get language code for this book item. Language of the book item can be different from the language settings defined globaly for book. :Returns: As string returns language code. """ return self.lang def add_link(self, **kwgs): """ Add additional link to the document. Links will be embeded only inside of this document. >>> add_link(href='styles.css', rel='stylesheet', type='text/css') """ self.links.append(kwgs) if kwgs.get('type') == 'text/javascript': if 'scripted' not in self.properties: self.properties.append('scripted') def get_links(self): """ Returns list of additional links defined for this document. :Returns: As tuple return list of links. """ return (link for link in self.links) def get_links_of_type(self, link_type): """ Returns list of additional links of specific type. :Returns: As tuple returns list of links. """ return (link for link in self.links if link.get('type', '') == link_type) def get_body_content(self): """ Returns content of BODY element for this HTML document. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Returns: Returns content of this document. """ try: html_tree = parse_html_string(self.content) except: return '' html_root = html_tree.getroottree() if len(html_root.find('body')) != 0: body = html_tree.find('body') tree_str = etree.tostring(body, pretty_print=True, encoding='utf-8', xml_declaration=False) # this is so stupid if tree_str.startswith(six.b('<body>')): n = tree_str.rindex(six.b('</body>')) return tree_str[6:n] return tree_str return '' def get_content(self, default=None): """ Returns content for this document as HTML string. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Args: - default: Default value for the content if it is not defined. :Returns: Returns content of this document. """ tree = parse_string(self.book.get_template(self._template_name)) tree_root = tree.getroot() tree_root.set('lang', self.lang or self.book.language) tree_root.attrib['{%s}lang' % NAMESPACES['XML']] = self.lang or self.book.language # add to the head also # <meta charset="utf-8" /> try: html_tree = parse_html_string(self.content) except: return '' html_root = html_tree.getroottree() # create and populate head _head = etree.SubElement(tree_root, 'head') if self.title != '': _title = etree.SubElement(_head, 'title') _title.text = self.title for lnk in self.links: if lnk.get('type') == 'text/javascript': _lnk = etree.SubElement(_head, 'script', lnk) # force <script></script> _lnk.text = '' else: _lnk = etree.SubElement(_head, 'link', lnk) # this should not be like this # head = html_root.find('head') # if head is not None: # for i in head.getchildren(): # if i.tag == 'title' and self.title != '': # continue # _head.append(i) # create and populate body _body = etree.SubElement(tree_root, 'body') if self.direction: _body.set('dir', self.direction) tree_root.set('dir', self.direction) body = html_tree.find('body') if body is not None: for i in body.getchildren(): _body.append(i) tree_str = etree.tostring(tree, pretty_print=True, encoding='utf-8', xml_declaration=True) return tree_str def __str__(self): return '<EpubHtml:%s:%s>' % (self.id, self.file_name)
aerkalov/ebooklib
ebooklib/epub.py
EpubHtml.get_body_content
python
def get_body_content(self): try: html_tree = parse_html_string(self.content) except: return '' html_root = html_tree.getroottree() if len(html_root.find('body')) != 0: body = html_tree.find('body') tree_str = etree.tostring(body, pretty_print=True, encoding='utf-8', xml_declaration=False) # this is so stupid if tree_str.startswith(six.b('<body>')): n = tree_str.rindex(six.b('</body>')) return tree_str[6:n] return tree_str return ''
Returns content of BODY element for this HTML document. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Returns: Returns content of this document.
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L341-L370
[ "def parse_html_string(s):\n from lxml import html\n\n utf8_parser = html.HTMLParser(encoding='utf-8')\n\n html_tree = html.document_fromstring(s, parser=utf8_parser)\n\n return html_tree\n" ]
class EpubHtml(EpubItem): """ Represents HTML document in the EPUB file. """ _template_name = 'chapter' def __init__(self, uid=None, file_name='', media_type='', content=None, title='', lang=None, direction=None, media_overlay=None, media_duration=None): super(EpubHtml, self).__init__(uid, file_name, media_type, content) self.title = title self.lang = lang self.direction = direction self.media_overlay = media_overlay self.media_duration = media_duration self.links = [] self.properties = [] self.pages = [] def is_chapter(self): """ Returns if this document is chapter or not. :Returns: Returns book value. """ return True def get_type(self): """ Always returns ebooklib.ITEM_DOCUMENT as type of this document. :Returns: Always returns ebooklib.ITEM_DOCUMENT """ return ebooklib.ITEM_DOCUMENT def set_language(self, lang): """ Sets language for this book item. By default it will use language of the book but it can be overwritten with this call. """ self.lang = lang def get_language(self): """ Get language code for this book item. Language of the book item can be different from the language settings defined globaly for book. :Returns: As string returns language code. """ return self.lang def add_link(self, **kwgs): """ Add additional link to the document. Links will be embeded only inside of this document. >>> add_link(href='styles.css', rel='stylesheet', type='text/css') """ self.links.append(kwgs) if kwgs.get('type') == 'text/javascript': if 'scripted' not in self.properties: self.properties.append('scripted') def get_links(self): """ Returns list of additional links defined for this document. :Returns: As tuple return list of links. """ return (link for link in self.links) def get_links_of_type(self, link_type): """ Returns list of additional links of specific type. :Returns: As tuple returns list of links. """ return (link for link in self.links if link.get('type', '') == link_type) def add_item(self, item): """ Add other item to this document. It will create additional links according to the item type. :Args: - item: item we want to add defined as instance of EpubItem """ if item.get_type() == ebooklib.ITEM_STYLE: self.add_link(href=item.get_name(), rel='stylesheet', type='text/css') if item.get_type() == ebooklib.ITEM_SCRIPT: self.add_link(src=item.get_name(), type='text/javascript') def get_content(self, default=None): """ Returns content for this document as HTML string. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Args: - default: Default value for the content if it is not defined. :Returns: Returns content of this document. """ tree = parse_string(self.book.get_template(self._template_name)) tree_root = tree.getroot() tree_root.set('lang', self.lang or self.book.language) tree_root.attrib['{%s}lang' % NAMESPACES['XML']] = self.lang or self.book.language # add to the head also # <meta charset="utf-8" /> try: html_tree = parse_html_string(self.content) except: return '' html_root = html_tree.getroottree() # create and populate head _head = etree.SubElement(tree_root, 'head') if self.title != '': _title = etree.SubElement(_head, 'title') _title.text = self.title for lnk in self.links: if lnk.get('type') == 'text/javascript': _lnk = etree.SubElement(_head, 'script', lnk) # force <script></script> _lnk.text = '' else: _lnk = etree.SubElement(_head, 'link', lnk) # this should not be like this # head = html_root.find('head') # if head is not None: # for i in head.getchildren(): # if i.tag == 'title' and self.title != '': # continue # _head.append(i) # create and populate body _body = etree.SubElement(tree_root, 'body') if self.direction: _body.set('dir', self.direction) tree_root.set('dir', self.direction) body = html_tree.find('body') if body is not None: for i in body.getchildren(): _body.append(i) tree_str = etree.tostring(tree, pretty_print=True, encoding='utf-8', xml_declaration=True) return tree_str def __str__(self): return '<EpubHtml:%s:%s>' % (self.id, self.file_name)
aerkalov/ebooklib
ebooklib/epub.py
EpubHtml.get_content
python
def get_content(self, default=None): tree = parse_string(self.book.get_template(self._template_name)) tree_root = tree.getroot() tree_root.set('lang', self.lang or self.book.language) tree_root.attrib['{%s}lang' % NAMESPACES['XML']] = self.lang or self.book.language # add to the head also # <meta charset="utf-8" /> try: html_tree = parse_html_string(self.content) except: return '' html_root = html_tree.getroottree() # create and populate head _head = etree.SubElement(tree_root, 'head') if self.title != '': _title = etree.SubElement(_head, 'title') _title.text = self.title for lnk in self.links: if lnk.get('type') == 'text/javascript': _lnk = etree.SubElement(_head, 'script', lnk) # force <script></script> _lnk.text = '' else: _lnk = etree.SubElement(_head, 'link', lnk) # this should not be like this # head = html_root.find('head') # if head is not None: # for i in head.getchildren(): # if i.tag == 'title' and self.title != '': # continue # _head.append(i) # create and populate body _body = etree.SubElement(tree_root, 'body') if self.direction: _body.set('dir', self.direction) tree_root.set('dir', self.direction) body = html_tree.find('body') if body is not None: for i in body.getchildren(): _body.append(i) tree_str = etree.tostring(tree, pretty_print=True, encoding='utf-8', xml_declaration=True) return tree_str
Returns content for this document as HTML string. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Args: - default: Default value for the content if it is not defined. :Returns: Returns content of this document.
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L372-L438
[ "def parse_html_string(s):\n from lxml import html\n\n utf8_parser = html.HTMLParser(encoding='utf-8')\n\n html_tree = html.document_fromstring(s, parser=utf8_parser)\n\n return html_tree\n", "def parse_string(s):\n try:\n tree = etree.parse(io.BytesIO(s.encode('utf-8')))\n except:\n tree = etree.parse(io.BytesIO(s))\n\n return tree\n" ]
class EpubHtml(EpubItem): """ Represents HTML document in the EPUB file. """ _template_name = 'chapter' def __init__(self, uid=None, file_name='', media_type='', content=None, title='', lang=None, direction=None, media_overlay=None, media_duration=None): super(EpubHtml, self).__init__(uid, file_name, media_type, content) self.title = title self.lang = lang self.direction = direction self.media_overlay = media_overlay self.media_duration = media_duration self.links = [] self.properties = [] self.pages = [] def is_chapter(self): """ Returns if this document is chapter or not. :Returns: Returns book value. """ return True def get_type(self): """ Always returns ebooklib.ITEM_DOCUMENT as type of this document. :Returns: Always returns ebooklib.ITEM_DOCUMENT """ return ebooklib.ITEM_DOCUMENT def set_language(self, lang): """ Sets language for this book item. By default it will use language of the book but it can be overwritten with this call. """ self.lang = lang def get_language(self): """ Get language code for this book item. Language of the book item can be different from the language settings defined globaly for book. :Returns: As string returns language code. """ return self.lang def add_link(self, **kwgs): """ Add additional link to the document. Links will be embeded only inside of this document. >>> add_link(href='styles.css', rel='stylesheet', type='text/css') """ self.links.append(kwgs) if kwgs.get('type') == 'text/javascript': if 'scripted' not in self.properties: self.properties.append('scripted') def get_links(self): """ Returns list of additional links defined for this document. :Returns: As tuple return list of links. """ return (link for link in self.links) def get_links_of_type(self, link_type): """ Returns list of additional links of specific type. :Returns: As tuple returns list of links. """ return (link for link in self.links if link.get('type', '') == link_type) def add_item(self, item): """ Add other item to this document. It will create additional links according to the item type. :Args: - item: item we want to add defined as instance of EpubItem """ if item.get_type() == ebooklib.ITEM_STYLE: self.add_link(href=item.get_name(), rel='stylesheet', type='text/css') if item.get_type() == ebooklib.ITEM_SCRIPT: self.add_link(src=item.get_name(), type='text/javascript') def get_body_content(self): """ Returns content of BODY element for this HTML document. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Returns: Returns content of this document. """ try: html_tree = parse_html_string(self.content) except: return '' html_root = html_tree.getroottree() if len(html_root.find('body')) != 0: body = html_tree.find('body') tree_str = etree.tostring(body, pretty_print=True, encoding='utf-8', xml_declaration=False) # this is so stupid if tree_str.startswith(six.b('<body>')): n = tree_str.rindex(six.b('</body>')) return tree_str[6:n] return tree_str return '' def __str__(self): return '<EpubHtml:%s:%s>' % (self.id, self.file_name)
aerkalov/ebooklib
ebooklib/epub.py
EpubCoverHtml.get_content
python
def get_content(self): self.content = self.book.get_template('cover') tree = parse_string(super(EpubCoverHtml, self).get_content()) tree_root = tree.getroot() images = tree_root.xpath('//xhtml:img', namespaces={'xhtml': NAMESPACES['XHTML']}) images[0].set('src', self.image_name) images[0].set('alt', self.title) tree_str = etree.tostring(tree, pretty_print=True, encoding='utf-8', xml_declaration=True) return tree_str
Returns content for cover page as HTML string. Content will be of type 'str' (Python 2) or 'bytes' (Python 3). :Returns: Returns content of this document.
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L466-L486
[ "def parse_string(s):\n try:\n tree = etree.parse(io.BytesIO(s.encode('utf-8')))\n except:\n tree = etree.parse(io.BytesIO(s))\n\n return tree\n", "def get_content(self, default=None):\n \"\"\"\n Returns content for this document as HTML string. Content will be of type 'str' (Python 2)\n or 'bytes' (Python 3).\n\n :Args:\n - default: Default value for the content if it is not defined.\n\n :Returns:\n Returns content of this document.\n \"\"\"\n\n tree = parse_string(self.book.get_template(self._template_name))\n tree_root = tree.getroot()\n\n tree_root.set('lang', self.lang or self.book.language)\n tree_root.attrib['{%s}lang' % NAMESPACES['XML']] = self.lang or self.book.language\n\n # add to the head also\n # <meta charset=\"utf-8\" />\n\n try:\n html_tree = parse_html_string(self.content)\n except:\n return ''\n\n html_root = html_tree.getroottree()\n\n # create and populate head\n\n _head = etree.SubElement(tree_root, 'head')\n\n if self.title != '':\n _title = etree.SubElement(_head, 'title')\n _title.text = self.title\n\n for lnk in self.links:\n if lnk.get('type') == 'text/javascript':\n _lnk = etree.SubElement(_head, 'script', lnk)\n # force <script></script>\n _lnk.text = ''\n else:\n _lnk = etree.SubElement(_head, 'link', lnk)\n\n # this should not be like this\n # head = html_root.find('head')\n # if head is not None:\n # for i in head.getchildren():\n # if i.tag == 'title' and self.title != '':\n # continue\n # _head.append(i)\n\n # create and populate body\n\n _body = etree.SubElement(tree_root, 'body')\n if self.direction:\n _body.set('dir', self.direction)\n tree_root.set('dir', self.direction)\n\n body = html_tree.find('body')\n if body is not None:\n for i in body.getchildren():\n _body.append(i)\n\n tree_str = etree.tostring(tree, pretty_print=True, encoding='utf-8', xml_declaration=True)\n\n return tree_str\n" ]
class EpubCoverHtml(EpubHtml): """ Represents Cover page in the EPUB file. """ def __init__(self, uid='cover', file_name='cover.xhtml', image_name='', title='Cover'): super(EpubCoverHtml, self).__init__(uid=uid, file_name=file_name, title=title) self.image_name = image_name self.is_linear = False def is_chapter(self): """ Returns if this document is chapter or not. :Returns: Returns book value. """ return False def __str__(self): return '<EpubCoverHtml:%s:%s>' % (self.id, self.file_name)
aerkalov/ebooklib
ebooklib/epub.py
EpubBook.reset
python
def reset(self): "Initialises all needed variables to default values" self.metadata = {} self.items = [] self.spine = [] self.guide = [] self.pages = [] self.toc = [] self.bindings = [] self.IDENTIFIER_ID = 'id' self.FOLDER_NAME = 'EPUB' self._id_html = 0 self._id_image = 0 self._id_static = 0 self.title = '' self.language = 'en' self.direction = None self.templates = { 'ncx': NCX_XML, 'nav': NAV_XML, 'chapter': CHAPTER_XML, 'cover': COVER_XML } self.add_metadata('OPF', 'generator', '', { 'name': 'generator', 'content': 'Ebook-lib %s' % '.'.join([str(s) for s in VERSION]) }) # default to using a randomly-unique identifier if one is not specified manually self.set_identifier(str(uuid.uuid4())) # custom prefixes and namespaces to be set to the content.opf doc self.prefixes = [] self.namespaces = {}
Initialises all needed variables to default values
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L554-L592
[ "def set_identifier(self, uid):\n \"\"\"\n Sets unique id for this epub\n\n :Args:\n - uid: Value of unique identifier for this book\n \"\"\"\n\n self.uid = uid\n\n self.set_unique_metadata('DC', 'identifier', self.uid, {'id': self.IDENTIFIER_ID})\n", "def add_metadata(self, namespace, name, value, others=None):\n \"Add metadata\"\n\n if namespace in NAMESPACES:\n namespace = NAMESPACES[namespace]\n\n if namespace not in self.metadata:\n self.metadata[namespace] = {}\n\n if name not in self.metadata[namespace]:\n self.metadata[namespace][name] = []\n\n self.metadata[namespace][name].append((value, others))\n" ]
class EpubBook(object): def __init__(self): self.EPUB_VERSION = None self.reset() # we should have options here def set_identifier(self, uid): """ Sets unique id for this epub :Args: - uid: Value of unique identifier for this book """ self.uid = uid self.set_unique_metadata('DC', 'identifier', self.uid, {'id': self.IDENTIFIER_ID}) def set_title(self, title): """ Set title. You can set multiple titles. :Args: - title: Title value """ self.title = title self.add_metadata('DC', 'title', self.title) def set_language(self, lang): """ Set language for this epub. You can set multiple languages. Specific items in the book can have different language settings. :Args: - lang: Language code """ self.language = lang self.add_metadata('DC', 'language', lang) def set_direction(self, direction): """ :Args: - direction: Options are "ltr", "rtl" and "default" """ self.direction = direction def set_cover(self, file_name, content, create_page=True): """ Set cover and create cover document if needed. :Args: - file_name: file name of the cover page - content: Content for the cover image - create_page: Should cover page be defined. Defined as bool value (optional). Default value is True. """ # as it is now, it can only be called once c0 = EpubCover(file_name=file_name) c0.content = content self.add_item(c0) if create_page: c1 = EpubCoverHtml(image_name=file_name) self.add_item(c1) self.add_metadata(None, 'meta', '', OrderedDict([('name', 'cover'), ('content', 'cover-img')])) def add_author(self, author, file_as=None, role=None, uid='creator'): "Add author for this document" self.add_metadata('DC', 'creator', author, {'id': uid}) if file_as: self.add_metadata(None, 'meta', file_as, {'refines': '#' + uid, 'property': 'file-as', 'scheme': 'marc:relators'}) if role: self.add_metadata(None, 'meta', role, {'refines': '#' + uid, 'property': 'role', 'scheme': 'marc:relators'}) def add_metadata(self, namespace, name, value, others=None): "Add metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace not in self.metadata: self.metadata[namespace] = {} if name not in self.metadata[namespace]: self.metadata[namespace][name] = [] self.metadata[namespace][name].append((value, others)) def get_metadata(self, namespace, name): "Retrieve metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] return self.metadata[namespace].get(name, []) def set_unique_metadata(self, namespace, name, value, others=None): "Add metadata if metadata with this identifier does not already exist, otherwise update existing metadata." if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace in self.metadata and name in self.metadata[namespace]: self.metadata[namespace][name] = [(value, others)] else: self.add_metadata(namespace, name, value, others) def add_item(self, item): """ Add additional item to the book. If not defined, media type and chapter id will be defined for the item. :Args: - item: Item instance """ if item.media_type == '': (has_guessed, media_type) = guess_type(item.get_name().lower()) if has_guessed: if media_type is not None: item.media_type = media_type else: item.media_type = has_guessed else: item.media_type = 'application/octet-stream' if not item.get_id(): # make chapter_, image_ and static_ configurable if isinstance(item, EpubHtml): item.id = 'chapter_%d' % self._id_html self._id_html += 1 # If there's a page list, append it to the book's page list self.pages += item.pages elif isinstance(item, EpubImage): item.id = 'image_%d' % self._id_image self._id_image += 1 else: item.id = 'static_%d' % self._id_image self._id_image += 1 item.book = self self.items.append(item) return item def get_item_with_id(self, uid): """ Returns item for defined UID. >>> book.get_item_with_id('image_001') :Args: - uid: UID for the item :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.id == uid: return item return None def get_item_with_href(self, href): """ Returns item for defined HREF. >>> book.get_item_with_href('EPUB/document.xhtml') :Args: - href: HREF for the item we are searching for :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.get_name() == href: return item return None def get_items(self): """ Returns all items attached to this book. :Returns: Returns all items as tuple. """ return (item for item in self.items) def get_items_of_type(self, item_type): """ Returns all items of specified type. >>> book.get_items_of_type(epub.ITEM_IMAGE) :Args: - item_type: Type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.get_type() == item_type) def get_items_of_media_type(self, media_type): """ Returns all items of specified media type. :Args: - media_type: Media type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.media_type == media_type) def set_template(self, name, value): """ Defines templates which are used to generate certain types of pages. When defining new value for the template we have to use content of type 'str' (Python 2) or 'bytes' (Python 3). At the moment we use these templates: - ncx - nav - chapter - cover :Args: - name: Name for the template - value: Content for the template """ self.templates[name] = value def get_template(self, name): """ Returns value for the template. :Args: - name: template name :Returns: Value of the template. """ return self.templates.get(name) def add_prefix(self, name, uri): """ Appends custom prefix to be added to the content.opf document >>> epub_book.add_prefix('bkterms', 'http://booktype.org/') :Args: - name: namespave name - uri: URI for the namespace """ self.prefixes.append('%s: %s' % (name, uri))
aerkalov/ebooklib
ebooklib/epub.py
EpubBook.set_identifier
python
def set_identifier(self, uid): self.uid = uid self.set_unique_metadata('DC', 'identifier', self.uid, {'id': self.IDENTIFIER_ID})
Sets unique id for this epub :Args: - uid: Value of unique identifier for this book
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L594-L604
[ "def set_unique_metadata(self, namespace, name, value, others=None):\n \"Add metadata if metadata with this identifier does not already exist, otherwise update existing metadata.\"\n\n if namespace in NAMESPACES:\n namespace = NAMESPACES[namespace]\n\n if namespace in self.metadata and name in self.metadata[namespace]:\n self.metadata[namespace][name] = [(value, others)]\n else:\n self.add_metadata(namespace, name, value, others)\n" ]
class EpubBook(object): def __init__(self): self.EPUB_VERSION = None self.reset() # we should have options here def reset(self): "Initialises all needed variables to default values" self.metadata = {} self.items = [] self.spine = [] self.guide = [] self.pages = [] self.toc = [] self.bindings = [] self.IDENTIFIER_ID = 'id' self.FOLDER_NAME = 'EPUB' self._id_html = 0 self._id_image = 0 self._id_static = 0 self.title = '' self.language = 'en' self.direction = None self.templates = { 'ncx': NCX_XML, 'nav': NAV_XML, 'chapter': CHAPTER_XML, 'cover': COVER_XML } self.add_metadata('OPF', 'generator', '', { 'name': 'generator', 'content': 'Ebook-lib %s' % '.'.join([str(s) for s in VERSION]) }) # default to using a randomly-unique identifier if one is not specified manually self.set_identifier(str(uuid.uuid4())) # custom prefixes and namespaces to be set to the content.opf doc self.prefixes = [] self.namespaces = {} def set_title(self, title): """ Set title. You can set multiple titles. :Args: - title: Title value """ self.title = title self.add_metadata('DC', 'title', self.title) def set_language(self, lang): """ Set language for this epub. You can set multiple languages. Specific items in the book can have different language settings. :Args: - lang: Language code """ self.language = lang self.add_metadata('DC', 'language', lang) def set_direction(self, direction): """ :Args: - direction: Options are "ltr", "rtl" and "default" """ self.direction = direction def set_cover(self, file_name, content, create_page=True): """ Set cover and create cover document if needed. :Args: - file_name: file name of the cover page - content: Content for the cover image - create_page: Should cover page be defined. Defined as bool value (optional). Default value is True. """ # as it is now, it can only be called once c0 = EpubCover(file_name=file_name) c0.content = content self.add_item(c0) if create_page: c1 = EpubCoverHtml(image_name=file_name) self.add_item(c1) self.add_metadata(None, 'meta', '', OrderedDict([('name', 'cover'), ('content', 'cover-img')])) def add_author(self, author, file_as=None, role=None, uid='creator'): "Add author for this document" self.add_metadata('DC', 'creator', author, {'id': uid}) if file_as: self.add_metadata(None, 'meta', file_as, {'refines': '#' + uid, 'property': 'file-as', 'scheme': 'marc:relators'}) if role: self.add_metadata(None, 'meta', role, {'refines': '#' + uid, 'property': 'role', 'scheme': 'marc:relators'}) def add_metadata(self, namespace, name, value, others=None): "Add metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace not in self.metadata: self.metadata[namespace] = {} if name not in self.metadata[namespace]: self.metadata[namespace][name] = [] self.metadata[namespace][name].append((value, others)) def get_metadata(self, namespace, name): "Retrieve metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] return self.metadata[namespace].get(name, []) def set_unique_metadata(self, namespace, name, value, others=None): "Add metadata if metadata with this identifier does not already exist, otherwise update existing metadata." if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace in self.metadata and name in self.metadata[namespace]: self.metadata[namespace][name] = [(value, others)] else: self.add_metadata(namespace, name, value, others) def add_item(self, item): """ Add additional item to the book. If not defined, media type and chapter id will be defined for the item. :Args: - item: Item instance """ if item.media_type == '': (has_guessed, media_type) = guess_type(item.get_name().lower()) if has_guessed: if media_type is not None: item.media_type = media_type else: item.media_type = has_guessed else: item.media_type = 'application/octet-stream' if not item.get_id(): # make chapter_, image_ and static_ configurable if isinstance(item, EpubHtml): item.id = 'chapter_%d' % self._id_html self._id_html += 1 # If there's a page list, append it to the book's page list self.pages += item.pages elif isinstance(item, EpubImage): item.id = 'image_%d' % self._id_image self._id_image += 1 else: item.id = 'static_%d' % self._id_image self._id_image += 1 item.book = self self.items.append(item) return item def get_item_with_id(self, uid): """ Returns item for defined UID. >>> book.get_item_with_id('image_001') :Args: - uid: UID for the item :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.id == uid: return item return None def get_item_with_href(self, href): """ Returns item for defined HREF. >>> book.get_item_with_href('EPUB/document.xhtml') :Args: - href: HREF for the item we are searching for :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.get_name() == href: return item return None def get_items(self): """ Returns all items attached to this book. :Returns: Returns all items as tuple. """ return (item for item in self.items) def get_items_of_type(self, item_type): """ Returns all items of specified type. >>> book.get_items_of_type(epub.ITEM_IMAGE) :Args: - item_type: Type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.get_type() == item_type) def get_items_of_media_type(self, media_type): """ Returns all items of specified media type. :Args: - media_type: Media type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.media_type == media_type) def set_template(self, name, value): """ Defines templates which are used to generate certain types of pages. When defining new value for the template we have to use content of type 'str' (Python 2) or 'bytes' (Python 3). At the moment we use these templates: - ncx - nav - chapter - cover :Args: - name: Name for the template - value: Content for the template """ self.templates[name] = value def get_template(self, name): """ Returns value for the template. :Args: - name: template name :Returns: Value of the template. """ return self.templates.get(name) def add_prefix(self, name, uri): """ Appends custom prefix to be added to the content.opf document >>> epub_book.add_prefix('bkterms', 'http://booktype.org/') :Args: - name: namespave name - uri: URI for the namespace """ self.prefixes.append('%s: %s' % (name, uri))
aerkalov/ebooklib
ebooklib/epub.py
EpubBook.set_title
python
def set_title(self, title): self.title = title self.add_metadata('DC', 'title', self.title)
Set title. You can set multiple titles. :Args: - title: Title value
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L606-L616
[ "def add_metadata(self, namespace, name, value, others=None):\n \"Add metadata\"\n\n if namespace in NAMESPACES:\n namespace = NAMESPACES[namespace]\n\n if namespace not in self.metadata:\n self.metadata[namespace] = {}\n\n if name not in self.metadata[namespace]:\n self.metadata[namespace][name] = []\n\n self.metadata[namespace][name].append((value, others))\n" ]
class EpubBook(object): def __init__(self): self.EPUB_VERSION = None self.reset() # we should have options here def reset(self): "Initialises all needed variables to default values" self.metadata = {} self.items = [] self.spine = [] self.guide = [] self.pages = [] self.toc = [] self.bindings = [] self.IDENTIFIER_ID = 'id' self.FOLDER_NAME = 'EPUB' self._id_html = 0 self._id_image = 0 self._id_static = 0 self.title = '' self.language = 'en' self.direction = None self.templates = { 'ncx': NCX_XML, 'nav': NAV_XML, 'chapter': CHAPTER_XML, 'cover': COVER_XML } self.add_metadata('OPF', 'generator', '', { 'name': 'generator', 'content': 'Ebook-lib %s' % '.'.join([str(s) for s in VERSION]) }) # default to using a randomly-unique identifier if one is not specified manually self.set_identifier(str(uuid.uuid4())) # custom prefixes and namespaces to be set to the content.opf doc self.prefixes = [] self.namespaces = {} def set_identifier(self, uid): """ Sets unique id for this epub :Args: - uid: Value of unique identifier for this book """ self.uid = uid self.set_unique_metadata('DC', 'identifier', self.uid, {'id': self.IDENTIFIER_ID}) def set_language(self, lang): """ Set language for this epub. You can set multiple languages. Specific items in the book can have different language settings. :Args: - lang: Language code """ self.language = lang self.add_metadata('DC', 'language', lang) def set_direction(self, direction): """ :Args: - direction: Options are "ltr", "rtl" and "default" """ self.direction = direction def set_cover(self, file_name, content, create_page=True): """ Set cover and create cover document if needed. :Args: - file_name: file name of the cover page - content: Content for the cover image - create_page: Should cover page be defined. Defined as bool value (optional). Default value is True. """ # as it is now, it can only be called once c0 = EpubCover(file_name=file_name) c0.content = content self.add_item(c0) if create_page: c1 = EpubCoverHtml(image_name=file_name) self.add_item(c1) self.add_metadata(None, 'meta', '', OrderedDict([('name', 'cover'), ('content', 'cover-img')])) def add_author(self, author, file_as=None, role=None, uid='creator'): "Add author for this document" self.add_metadata('DC', 'creator', author, {'id': uid}) if file_as: self.add_metadata(None, 'meta', file_as, {'refines': '#' + uid, 'property': 'file-as', 'scheme': 'marc:relators'}) if role: self.add_metadata(None, 'meta', role, {'refines': '#' + uid, 'property': 'role', 'scheme': 'marc:relators'}) def add_metadata(self, namespace, name, value, others=None): "Add metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace not in self.metadata: self.metadata[namespace] = {} if name not in self.metadata[namespace]: self.metadata[namespace][name] = [] self.metadata[namespace][name].append((value, others)) def get_metadata(self, namespace, name): "Retrieve metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] return self.metadata[namespace].get(name, []) def set_unique_metadata(self, namespace, name, value, others=None): "Add metadata if metadata with this identifier does not already exist, otherwise update existing metadata." if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace in self.metadata and name in self.metadata[namespace]: self.metadata[namespace][name] = [(value, others)] else: self.add_metadata(namespace, name, value, others) def add_item(self, item): """ Add additional item to the book. If not defined, media type and chapter id will be defined for the item. :Args: - item: Item instance """ if item.media_type == '': (has_guessed, media_type) = guess_type(item.get_name().lower()) if has_guessed: if media_type is not None: item.media_type = media_type else: item.media_type = has_guessed else: item.media_type = 'application/octet-stream' if not item.get_id(): # make chapter_, image_ and static_ configurable if isinstance(item, EpubHtml): item.id = 'chapter_%d' % self._id_html self._id_html += 1 # If there's a page list, append it to the book's page list self.pages += item.pages elif isinstance(item, EpubImage): item.id = 'image_%d' % self._id_image self._id_image += 1 else: item.id = 'static_%d' % self._id_image self._id_image += 1 item.book = self self.items.append(item) return item def get_item_with_id(self, uid): """ Returns item for defined UID. >>> book.get_item_with_id('image_001') :Args: - uid: UID for the item :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.id == uid: return item return None def get_item_with_href(self, href): """ Returns item for defined HREF. >>> book.get_item_with_href('EPUB/document.xhtml') :Args: - href: HREF for the item we are searching for :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.get_name() == href: return item return None def get_items(self): """ Returns all items attached to this book. :Returns: Returns all items as tuple. """ return (item for item in self.items) def get_items_of_type(self, item_type): """ Returns all items of specified type. >>> book.get_items_of_type(epub.ITEM_IMAGE) :Args: - item_type: Type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.get_type() == item_type) def get_items_of_media_type(self, media_type): """ Returns all items of specified media type. :Args: - media_type: Media type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.media_type == media_type) def set_template(self, name, value): """ Defines templates which are used to generate certain types of pages. When defining new value for the template we have to use content of type 'str' (Python 2) or 'bytes' (Python 3). At the moment we use these templates: - ncx - nav - chapter - cover :Args: - name: Name for the template - value: Content for the template """ self.templates[name] = value def get_template(self, name): """ Returns value for the template. :Args: - name: template name :Returns: Value of the template. """ return self.templates.get(name) def add_prefix(self, name, uri): """ Appends custom prefix to be added to the content.opf document >>> epub_book.add_prefix('bkterms', 'http://booktype.org/') :Args: - name: namespave name - uri: URI for the namespace """ self.prefixes.append('%s: %s' % (name, uri))
aerkalov/ebooklib
ebooklib/epub.py
EpubBook.set_cover
python
def set_cover(self, file_name, content, create_page=True): # as it is now, it can only be called once c0 = EpubCover(file_name=file_name) c0.content = content self.add_item(c0) if create_page: c1 = EpubCoverHtml(image_name=file_name) self.add_item(c1) self.add_metadata(None, 'meta', '', OrderedDict([('name', 'cover'), ('content', 'cover-img')]))
Set cover and create cover document if needed. :Args: - file_name: file name of the cover page - content: Content for the cover image - create_page: Should cover page be defined. Defined as bool value (optional). Default value is True.
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L639-L658
[ "def add_metadata(self, namespace, name, value, others=None):\n \"Add metadata\"\n\n if namespace in NAMESPACES:\n namespace = NAMESPACES[namespace]\n\n if namespace not in self.metadata:\n self.metadata[namespace] = {}\n\n if name not in self.metadata[namespace]:\n self.metadata[namespace][name] = []\n\n self.metadata[namespace][name].append((value, others))\n", "def add_item(self, item):\n \"\"\"\n Add additional item to the book. If not defined, media type and chapter id will be defined\n for the item.\n\n :Args:\n - item: Item instance\n \"\"\"\n if item.media_type == '':\n (has_guessed, media_type) = guess_type(item.get_name().lower())\n\n if has_guessed:\n if media_type is not None:\n item.media_type = media_type\n else:\n item.media_type = has_guessed\n else:\n item.media_type = 'application/octet-stream'\n\n if not item.get_id():\n # make chapter_, image_ and static_ configurable\n if isinstance(item, EpubHtml):\n item.id = 'chapter_%d' % self._id_html\n self._id_html += 1\n # If there's a page list, append it to the book's page list\n self.pages += item.pages\n elif isinstance(item, EpubImage):\n item.id = 'image_%d' % self._id_image\n self._id_image += 1\n else:\n item.id = 'static_%d' % self._id_image\n self._id_image += 1\n\n item.book = self\n self.items.append(item)\n\n return item\n" ]
class EpubBook(object): def __init__(self): self.EPUB_VERSION = None self.reset() # we should have options here def reset(self): "Initialises all needed variables to default values" self.metadata = {} self.items = [] self.spine = [] self.guide = [] self.pages = [] self.toc = [] self.bindings = [] self.IDENTIFIER_ID = 'id' self.FOLDER_NAME = 'EPUB' self._id_html = 0 self._id_image = 0 self._id_static = 0 self.title = '' self.language = 'en' self.direction = None self.templates = { 'ncx': NCX_XML, 'nav': NAV_XML, 'chapter': CHAPTER_XML, 'cover': COVER_XML } self.add_metadata('OPF', 'generator', '', { 'name': 'generator', 'content': 'Ebook-lib %s' % '.'.join([str(s) for s in VERSION]) }) # default to using a randomly-unique identifier if one is not specified manually self.set_identifier(str(uuid.uuid4())) # custom prefixes and namespaces to be set to the content.opf doc self.prefixes = [] self.namespaces = {} def set_identifier(self, uid): """ Sets unique id for this epub :Args: - uid: Value of unique identifier for this book """ self.uid = uid self.set_unique_metadata('DC', 'identifier', self.uid, {'id': self.IDENTIFIER_ID}) def set_title(self, title): """ Set title. You can set multiple titles. :Args: - title: Title value """ self.title = title self.add_metadata('DC', 'title', self.title) def set_language(self, lang): """ Set language for this epub. You can set multiple languages. Specific items in the book can have different language settings. :Args: - lang: Language code """ self.language = lang self.add_metadata('DC', 'language', lang) def set_direction(self, direction): """ :Args: - direction: Options are "ltr", "rtl" and "default" """ self.direction = direction def add_author(self, author, file_as=None, role=None, uid='creator'): "Add author for this document" self.add_metadata('DC', 'creator', author, {'id': uid}) if file_as: self.add_metadata(None, 'meta', file_as, {'refines': '#' + uid, 'property': 'file-as', 'scheme': 'marc:relators'}) if role: self.add_metadata(None, 'meta', role, {'refines': '#' + uid, 'property': 'role', 'scheme': 'marc:relators'}) def add_metadata(self, namespace, name, value, others=None): "Add metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace not in self.metadata: self.metadata[namespace] = {} if name not in self.metadata[namespace]: self.metadata[namespace][name] = [] self.metadata[namespace][name].append((value, others)) def get_metadata(self, namespace, name): "Retrieve metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] return self.metadata[namespace].get(name, []) def set_unique_metadata(self, namespace, name, value, others=None): "Add metadata if metadata with this identifier does not already exist, otherwise update existing metadata." if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace in self.metadata and name in self.metadata[namespace]: self.metadata[namespace][name] = [(value, others)] else: self.add_metadata(namespace, name, value, others) def add_item(self, item): """ Add additional item to the book. If not defined, media type and chapter id will be defined for the item. :Args: - item: Item instance """ if item.media_type == '': (has_guessed, media_type) = guess_type(item.get_name().lower()) if has_guessed: if media_type is not None: item.media_type = media_type else: item.media_type = has_guessed else: item.media_type = 'application/octet-stream' if not item.get_id(): # make chapter_, image_ and static_ configurable if isinstance(item, EpubHtml): item.id = 'chapter_%d' % self._id_html self._id_html += 1 # If there's a page list, append it to the book's page list self.pages += item.pages elif isinstance(item, EpubImage): item.id = 'image_%d' % self._id_image self._id_image += 1 else: item.id = 'static_%d' % self._id_image self._id_image += 1 item.book = self self.items.append(item) return item def get_item_with_id(self, uid): """ Returns item for defined UID. >>> book.get_item_with_id('image_001') :Args: - uid: UID for the item :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.id == uid: return item return None def get_item_with_href(self, href): """ Returns item for defined HREF. >>> book.get_item_with_href('EPUB/document.xhtml') :Args: - href: HREF for the item we are searching for :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.get_name() == href: return item return None def get_items(self): """ Returns all items attached to this book. :Returns: Returns all items as tuple. """ return (item for item in self.items) def get_items_of_type(self, item_type): """ Returns all items of specified type. >>> book.get_items_of_type(epub.ITEM_IMAGE) :Args: - item_type: Type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.get_type() == item_type) def get_items_of_media_type(self, media_type): """ Returns all items of specified media type. :Args: - media_type: Media type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.media_type == media_type) def set_template(self, name, value): """ Defines templates which are used to generate certain types of pages. When defining new value for the template we have to use content of type 'str' (Python 2) or 'bytes' (Python 3). At the moment we use these templates: - ncx - nav - chapter - cover :Args: - name: Name for the template - value: Content for the template """ self.templates[name] = value def get_template(self, name): """ Returns value for the template. :Args: - name: template name :Returns: Value of the template. """ return self.templates.get(name) def add_prefix(self, name, uri): """ Appends custom prefix to be added to the content.opf document >>> epub_book.add_prefix('bkterms', 'http://booktype.org/') :Args: - name: namespave name - uri: URI for the namespace """ self.prefixes.append('%s: %s' % (name, uri))
aerkalov/ebooklib
ebooklib/epub.py
EpubBook.add_author
python
def add_author(self, author, file_as=None, role=None, uid='creator'): "Add author for this document" self.add_metadata('DC', 'creator', author, {'id': uid}) if file_as: self.add_metadata(None, 'meta', file_as, {'refines': '#' + uid, 'property': 'file-as', 'scheme': 'marc:relators'}) if role: self.add_metadata(None, 'meta', role, {'refines': '#' + uid, 'property': 'role', 'scheme': 'marc:relators'})
Add author for this document
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L660-L672
[ "def add_metadata(self, namespace, name, value, others=None):\n \"Add metadata\"\n\n if namespace in NAMESPACES:\n namespace = NAMESPACES[namespace]\n\n if namespace not in self.metadata:\n self.metadata[namespace] = {}\n\n if name not in self.metadata[namespace]:\n self.metadata[namespace][name] = []\n\n self.metadata[namespace][name].append((value, others))\n" ]
class EpubBook(object): def __init__(self): self.EPUB_VERSION = None self.reset() # we should have options here def reset(self): "Initialises all needed variables to default values" self.metadata = {} self.items = [] self.spine = [] self.guide = [] self.pages = [] self.toc = [] self.bindings = [] self.IDENTIFIER_ID = 'id' self.FOLDER_NAME = 'EPUB' self._id_html = 0 self._id_image = 0 self._id_static = 0 self.title = '' self.language = 'en' self.direction = None self.templates = { 'ncx': NCX_XML, 'nav': NAV_XML, 'chapter': CHAPTER_XML, 'cover': COVER_XML } self.add_metadata('OPF', 'generator', '', { 'name': 'generator', 'content': 'Ebook-lib %s' % '.'.join([str(s) for s in VERSION]) }) # default to using a randomly-unique identifier if one is not specified manually self.set_identifier(str(uuid.uuid4())) # custom prefixes and namespaces to be set to the content.opf doc self.prefixes = [] self.namespaces = {} def set_identifier(self, uid): """ Sets unique id for this epub :Args: - uid: Value of unique identifier for this book """ self.uid = uid self.set_unique_metadata('DC', 'identifier', self.uid, {'id': self.IDENTIFIER_ID}) def set_title(self, title): """ Set title. You can set multiple titles. :Args: - title: Title value """ self.title = title self.add_metadata('DC', 'title', self.title) def set_language(self, lang): """ Set language for this epub. You can set multiple languages. Specific items in the book can have different language settings. :Args: - lang: Language code """ self.language = lang self.add_metadata('DC', 'language', lang) def set_direction(self, direction): """ :Args: - direction: Options are "ltr", "rtl" and "default" """ self.direction = direction def set_cover(self, file_name, content, create_page=True): """ Set cover and create cover document if needed. :Args: - file_name: file name of the cover page - content: Content for the cover image - create_page: Should cover page be defined. Defined as bool value (optional). Default value is True. """ # as it is now, it can only be called once c0 = EpubCover(file_name=file_name) c0.content = content self.add_item(c0) if create_page: c1 = EpubCoverHtml(image_name=file_name) self.add_item(c1) self.add_metadata(None, 'meta', '', OrderedDict([('name', 'cover'), ('content', 'cover-img')])) def add_metadata(self, namespace, name, value, others=None): "Add metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace not in self.metadata: self.metadata[namespace] = {} if name not in self.metadata[namespace]: self.metadata[namespace][name] = [] self.metadata[namespace][name].append((value, others)) def get_metadata(self, namespace, name): "Retrieve metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] return self.metadata[namespace].get(name, []) def set_unique_metadata(self, namespace, name, value, others=None): "Add metadata if metadata with this identifier does not already exist, otherwise update existing metadata." if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace in self.metadata and name in self.metadata[namespace]: self.metadata[namespace][name] = [(value, others)] else: self.add_metadata(namespace, name, value, others) def add_item(self, item): """ Add additional item to the book. If not defined, media type and chapter id will be defined for the item. :Args: - item: Item instance """ if item.media_type == '': (has_guessed, media_type) = guess_type(item.get_name().lower()) if has_guessed: if media_type is not None: item.media_type = media_type else: item.media_type = has_guessed else: item.media_type = 'application/octet-stream' if not item.get_id(): # make chapter_, image_ and static_ configurable if isinstance(item, EpubHtml): item.id = 'chapter_%d' % self._id_html self._id_html += 1 # If there's a page list, append it to the book's page list self.pages += item.pages elif isinstance(item, EpubImage): item.id = 'image_%d' % self._id_image self._id_image += 1 else: item.id = 'static_%d' % self._id_image self._id_image += 1 item.book = self self.items.append(item) return item def get_item_with_id(self, uid): """ Returns item for defined UID. >>> book.get_item_with_id('image_001') :Args: - uid: UID for the item :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.id == uid: return item return None def get_item_with_href(self, href): """ Returns item for defined HREF. >>> book.get_item_with_href('EPUB/document.xhtml') :Args: - href: HREF for the item we are searching for :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.get_name() == href: return item return None def get_items(self): """ Returns all items attached to this book. :Returns: Returns all items as tuple. """ return (item for item in self.items) def get_items_of_type(self, item_type): """ Returns all items of specified type. >>> book.get_items_of_type(epub.ITEM_IMAGE) :Args: - item_type: Type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.get_type() == item_type) def get_items_of_media_type(self, media_type): """ Returns all items of specified media type. :Args: - media_type: Media type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.media_type == media_type) def set_template(self, name, value): """ Defines templates which are used to generate certain types of pages. When defining new value for the template we have to use content of type 'str' (Python 2) or 'bytes' (Python 3). At the moment we use these templates: - ncx - nav - chapter - cover :Args: - name: Name for the template - value: Content for the template """ self.templates[name] = value def get_template(self, name): """ Returns value for the template. :Args: - name: template name :Returns: Value of the template. """ return self.templates.get(name) def add_prefix(self, name, uri): """ Appends custom prefix to be added to the content.opf document >>> epub_book.add_prefix('bkterms', 'http://booktype.org/') :Args: - name: namespave name - uri: URI for the namespace """ self.prefixes.append('%s: %s' % (name, uri))
aerkalov/ebooklib
ebooklib/epub.py
EpubBook.add_metadata
python
def add_metadata(self, namespace, name, value, others=None): "Add metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace not in self.metadata: self.metadata[namespace] = {} if name not in self.metadata[namespace]: self.metadata[namespace][name] = [] self.metadata[namespace][name].append((value, others))
Add metadata
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L674-L686
null
class EpubBook(object): def __init__(self): self.EPUB_VERSION = None self.reset() # we should have options here def reset(self): "Initialises all needed variables to default values" self.metadata = {} self.items = [] self.spine = [] self.guide = [] self.pages = [] self.toc = [] self.bindings = [] self.IDENTIFIER_ID = 'id' self.FOLDER_NAME = 'EPUB' self._id_html = 0 self._id_image = 0 self._id_static = 0 self.title = '' self.language = 'en' self.direction = None self.templates = { 'ncx': NCX_XML, 'nav': NAV_XML, 'chapter': CHAPTER_XML, 'cover': COVER_XML } self.add_metadata('OPF', 'generator', '', { 'name': 'generator', 'content': 'Ebook-lib %s' % '.'.join([str(s) for s in VERSION]) }) # default to using a randomly-unique identifier if one is not specified manually self.set_identifier(str(uuid.uuid4())) # custom prefixes and namespaces to be set to the content.opf doc self.prefixes = [] self.namespaces = {} def set_identifier(self, uid): """ Sets unique id for this epub :Args: - uid: Value of unique identifier for this book """ self.uid = uid self.set_unique_metadata('DC', 'identifier', self.uid, {'id': self.IDENTIFIER_ID}) def set_title(self, title): """ Set title. You can set multiple titles. :Args: - title: Title value """ self.title = title self.add_metadata('DC', 'title', self.title) def set_language(self, lang): """ Set language for this epub. You can set multiple languages. Specific items in the book can have different language settings. :Args: - lang: Language code """ self.language = lang self.add_metadata('DC', 'language', lang) def set_direction(self, direction): """ :Args: - direction: Options are "ltr", "rtl" and "default" """ self.direction = direction def set_cover(self, file_name, content, create_page=True): """ Set cover and create cover document if needed. :Args: - file_name: file name of the cover page - content: Content for the cover image - create_page: Should cover page be defined. Defined as bool value (optional). Default value is True. """ # as it is now, it can only be called once c0 = EpubCover(file_name=file_name) c0.content = content self.add_item(c0) if create_page: c1 = EpubCoverHtml(image_name=file_name) self.add_item(c1) self.add_metadata(None, 'meta', '', OrderedDict([('name', 'cover'), ('content', 'cover-img')])) def add_author(self, author, file_as=None, role=None, uid='creator'): "Add author for this document" self.add_metadata('DC', 'creator', author, {'id': uid}) if file_as: self.add_metadata(None, 'meta', file_as, {'refines': '#' + uid, 'property': 'file-as', 'scheme': 'marc:relators'}) if role: self.add_metadata(None, 'meta', role, {'refines': '#' + uid, 'property': 'role', 'scheme': 'marc:relators'}) def get_metadata(self, namespace, name): "Retrieve metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] return self.metadata[namespace].get(name, []) def set_unique_metadata(self, namespace, name, value, others=None): "Add metadata if metadata with this identifier does not already exist, otherwise update existing metadata." if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace in self.metadata and name in self.metadata[namespace]: self.metadata[namespace][name] = [(value, others)] else: self.add_metadata(namespace, name, value, others) def add_item(self, item): """ Add additional item to the book. If not defined, media type and chapter id will be defined for the item. :Args: - item: Item instance """ if item.media_type == '': (has_guessed, media_type) = guess_type(item.get_name().lower()) if has_guessed: if media_type is not None: item.media_type = media_type else: item.media_type = has_guessed else: item.media_type = 'application/octet-stream' if not item.get_id(): # make chapter_, image_ and static_ configurable if isinstance(item, EpubHtml): item.id = 'chapter_%d' % self._id_html self._id_html += 1 # If there's a page list, append it to the book's page list self.pages += item.pages elif isinstance(item, EpubImage): item.id = 'image_%d' % self._id_image self._id_image += 1 else: item.id = 'static_%d' % self._id_image self._id_image += 1 item.book = self self.items.append(item) return item def get_item_with_id(self, uid): """ Returns item for defined UID. >>> book.get_item_with_id('image_001') :Args: - uid: UID for the item :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.id == uid: return item return None def get_item_with_href(self, href): """ Returns item for defined HREF. >>> book.get_item_with_href('EPUB/document.xhtml') :Args: - href: HREF for the item we are searching for :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.get_name() == href: return item return None def get_items(self): """ Returns all items attached to this book. :Returns: Returns all items as tuple. """ return (item for item in self.items) def get_items_of_type(self, item_type): """ Returns all items of specified type. >>> book.get_items_of_type(epub.ITEM_IMAGE) :Args: - item_type: Type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.get_type() == item_type) def get_items_of_media_type(self, media_type): """ Returns all items of specified media type. :Args: - media_type: Media type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.media_type == media_type) def set_template(self, name, value): """ Defines templates which are used to generate certain types of pages. When defining new value for the template we have to use content of type 'str' (Python 2) or 'bytes' (Python 3). At the moment we use these templates: - ncx - nav - chapter - cover :Args: - name: Name for the template - value: Content for the template """ self.templates[name] = value def get_template(self, name): """ Returns value for the template. :Args: - name: template name :Returns: Value of the template. """ return self.templates.get(name) def add_prefix(self, name, uri): """ Appends custom prefix to be added to the content.opf document >>> epub_book.add_prefix('bkterms', 'http://booktype.org/') :Args: - name: namespave name - uri: URI for the namespace """ self.prefixes.append('%s: %s' % (name, uri))
aerkalov/ebooklib
ebooklib/epub.py
EpubBook.get_metadata
python
def get_metadata(self, namespace, name): "Retrieve metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] return self.metadata[namespace].get(name, [])
Retrieve metadata
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L688-L694
null
class EpubBook(object): def __init__(self): self.EPUB_VERSION = None self.reset() # we should have options here def reset(self): "Initialises all needed variables to default values" self.metadata = {} self.items = [] self.spine = [] self.guide = [] self.pages = [] self.toc = [] self.bindings = [] self.IDENTIFIER_ID = 'id' self.FOLDER_NAME = 'EPUB' self._id_html = 0 self._id_image = 0 self._id_static = 0 self.title = '' self.language = 'en' self.direction = None self.templates = { 'ncx': NCX_XML, 'nav': NAV_XML, 'chapter': CHAPTER_XML, 'cover': COVER_XML } self.add_metadata('OPF', 'generator', '', { 'name': 'generator', 'content': 'Ebook-lib %s' % '.'.join([str(s) for s in VERSION]) }) # default to using a randomly-unique identifier if one is not specified manually self.set_identifier(str(uuid.uuid4())) # custom prefixes and namespaces to be set to the content.opf doc self.prefixes = [] self.namespaces = {} def set_identifier(self, uid): """ Sets unique id for this epub :Args: - uid: Value of unique identifier for this book """ self.uid = uid self.set_unique_metadata('DC', 'identifier', self.uid, {'id': self.IDENTIFIER_ID}) def set_title(self, title): """ Set title. You can set multiple titles. :Args: - title: Title value """ self.title = title self.add_metadata('DC', 'title', self.title) def set_language(self, lang): """ Set language for this epub. You can set multiple languages. Specific items in the book can have different language settings. :Args: - lang: Language code """ self.language = lang self.add_metadata('DC', 'language', lang) def set_direction(self, direction): """ :Args: - direction: Options are "ltr", "rtl" and "default" """ self.direction = direction def set_cover(self, file_name, content, create_page=True): """ Set cover and create cover document if needed. :Args: - file_name: file name of the cover page - content: Content for the cover image - create_page: Should cover page be defined. Defined as bool value (optional). Default value is True. """ # as it is now, it can only be called once c0 = EpubCover(file_name=file_name) c0.content = content self.add_item(c0) if create_page: c1 = EpubCoverHtml(image_name=file_name) self.add_item(c1) self.add_metadata(None, 'meta', '', OrderedDict([('name', 'cover'), ('content', 'cover-img')])) def add_author(self, author, file_as=None, role=None, uid='creator'): "Add author for this document" self.add_metadata('DC', 'creator', author, {'id': uid}) if file_as: self.add_metadata(None, 'meta', file_as, {'refines': '#' + uid, 'property': 'file-as', 'scheme': 'marc:relators'}) if role: self.add_metadata(None, 'meta', role, {'refines': '#' + uid, 'property': 'role', 'scheme': 'marc:relators'}) def add_metadata(self, namespace, name, value, others=None): "Add metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace not in self.metadata: self.metadata[namespace] = {} if name not in self.metadata[namespace]: self.metadata[namespace][name] = [] self.metadata[namespace][name].append((value, others)) def set_unique_metadata(self, namespace, name, value, others=None): "Add metadata if metadata with this identifier does not already exist, otherwise update existing metadata." if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace in self.metadata and name in self.metadata[namespace]: self.metadata[namespace][name] = [(value, others)] else: self.add_metadata(namespace, name, value, others) def add_item(self, item): """ Add additional item to the book. If not defined, media type and chapter id will be defined for the item. :Args: - item: Item instance """ if item.media_type == '': (has_guessed, media_type) = guess_type(item.get_name().lower()) if has_guessed: if media_type is not None: item.media_type = media_type else: item.media_type = has_guessed else: item.media_type = 'application/octet-stream' if not item.get_id(): # make chapter_, image_ and static_ configurable if isinstance(item, EpubHtml): item.id = 'chapter_%d' % self._id_html self._id_html += 1 # If there's a page list, append it to the book's page list self.pages += item.pages elif isinstance(item, EpubImage): item.id = 'image_%d' % self._id_image self._id_image += 1 else: item.id = 'static_%d' % self._id_image self._id_image += 1 item.book = self self.items.append(item) return item def get_item_with_id(self, uid): """ Returns item for defined UID. >>> book.get_item_with_id('image_001') :Args: - uid: UID for the item :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.id == uid: return item return None def get_item_with_href(self, href): """ Returns item for defined HREF. >>> book.get_item_with_href('EPUB/document.xhtml') :Args: - href: HREF for the item we are searching for :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.get_name() == href: return item return None def get_items(self): """ Returns all items attached to this book. :Returns: Returns all items as tuple. """ return (item for item in self.items) def get_items_of_type(self, item_type): """ Returns all items of specified type. >>> book.get_items_of_type(epub.ITEM_IMAGE) :Args: - item_type: Type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.get_type() == item_type) def get_items_of_media_type(self, media_type): """ Returns all items of specified media type. :Args: - media_type: Media type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.media_type == media_type) def set_template(self, name, value): """ Defines templates which are used to generate certain types of pages. When defining new value for the template we have to use content of type 'str' (Python 2) or 'bytes' (Python 3). At the moment we use these templates: - ncx - nav - chapter - cover :Args: - name: Name for the template - value: Content for the template """ self.templates[name] = value def get_template(self, name): """ Returns value for the template. :Args: - name: template name :Returns: Value of the template. """ return self.templates.get(name) def add_prefix(self, name, uri): """ Appends custom prefix to be added to the content.opf document >>> epub_book.add_prefix('bkterms', 'http://booktype.org/') :Args: - name: namespave name - uri: URI for the namespace """ self.prefixes.append('%s: %s' % (name, uri))
aerkalov/ebooklib
ebooklib/epub.py
EpubBook.add_item
python
def add_item(self, item): if item.media_type == '': (has_guessed, media_type) = guess_type(item.get_name().lower()) if has_guessed: if media_type is not None: item.media_type = media_type else: item.media_type = has_guessed else: item.media_type = 'application/octet-stream' if not item.get_id(): # make chapter_, image_ and static_ configurable if isinstance(item, EpubHtml): item.id = 'chapter_%d' % self._id_html self._id_html += 1 # If there's a page list, append it to the book's page list self.pages += item.pages elif isinstance(item, EpubImage): item.id = 'image_%d' % self._id_image self._id_image += 1 else: item.id = 'static_%d' % self._id_image self._id_image += 1 item.book = self self.items.append(item) return item
Add additional item to the book. If not defined, media type and chapter id will be defined for the item. :Args: - item: Item instance
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L707-L743
[ "def guess_type(extenstion):\n global mimetype_initialised\n\n if not mimetype_initialised:\n mimetypes.init()\n mimetypes.add_type('application/xhtml+xml', '.xhtml')\n mimetype_initialised = True\n\n return mimetypes.guess_type(extenstion)\n", "def get_id(self):\n \"\"\"\n Returns unique identifier for this item.\n\n :Returns:\n Returns uid number as string.\n \"\"\"\n return self.id\n", "def get_name(self):\n \"\"\"\n Returns name for this item. By default it is always file name but it does not have to be.\n\n :Returns:\n Returns file name for this item.\n \"\"\"\n return self.file_name\n" ]
class EpubBook(object): def __init__(self): self.EPUB_VERSION = None self.reset() # we should have options here def reset(self): "Initialises all needed variables to default values" self.metadata = {} self.items = [] self.spine = [] self.guide = [] self.pages = [] self.toc = [] self.bindings = [] self.IDENTIFIER_ID = 'id' self.FOLDER_NAME = 'EPUB' self._id_html = 0 self._id_image = 0 self._id_static = 0 self.title = '' self.language = 'en' self.direction = None self.templates = { 'ncx': NCX_XML, 'nav': NAV_XML, 'chapter': CHAPTER_XML, 'cover': COVER_XML } self.add_metadata('OPF', 'generator', '', { 'name': 'generator', 'content': 'Ebook-lib %s' % '.'.join([str(s) for s in VERSION]) }) # default to using a randomly-unique identifier if one is not specified manually self.set_identifier(str(uuid.uuid4())) # custom prefixes and namespaces to be set to the content.opf doc self.prefixes = [] self.namespaces = {} def set_identifier(self, uid): """ Sets unique id for this epub :Args: - uid: Value of unique identifier for this book """ self.uid = uid self.set_unique_metadata('DC', 'identifier', self.uid, {'id': self.IDENTIFIER_ID}) def set_title(self, title): """ Set title. You can set multiple titles. :Args: - title: Title value """ self.title = title self.add_metadata('DC', 'title', self.title) def set_language(self, lang): """ Set language for this epub. You can set multiple languages. Specific items in the book can have different language settings. :Args: - lang: Language code """ self.language = lang self.add_metadata('DC', 'language', lang) def set_direction(self, direction): """ :Args: - direction: Options are "ltr", "rtl" and "default" """ self.direction = direction def set_cover(self, file_name, content, create_page=True): """ Set cover and create cover document if needed. :Args: - file_name: file name of the cover page - content: Content for the cover image - create_page: Should cover page be defined. Defined as bool value (optional). Default value is True. """ # as it is now, it can only be called once c0 = EpubCover(file_name=file_name) c0.content = content self.add_item(c0) if create_page: c1 = EpubCoverHtml(image_name=file_name) self.add_item(c1) self.add_metadata(None, 'meta', '', OrderedDict([('name', 'cover'), ('content', 'cover-img')])) def add_author(self, author, file_as=None, role=None, uid='creator'): "Add author for this document" self.add_metadata('DC', 'creator', author, {'id': uid}) if file_as: self.add_metadata(None, 'meta', file_as, {'refines': '#' + uid, 'property': 'file-as', 'scheme': 'marc:relators'}) if role: self.add_metadata(None, 'meta', role, {'refines': '#' + uid, 'property': 'role', 'scheme': 'marc:relators'}) def add_metadata(self, namespace, name, value, others=None): "Add metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace not in self.metadata: self.metadata[namespace] = {} if name not in self.metadata[namespace]: self.metadata[namespace][name] = [] self.metadata[namespace][name].append((value, others)) def get_metadata(self, namespace, name): "Retrieve metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] return self.metadata[namespace].get(name, []) def set_unique_metadata(self, namespace, name, value, others=None): "Add metadata if metadata with this identifier does not already exist, otherwise update existing metadata." if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace in self.metadata and name in self.metadata[namespace]: self.metadata[namespace][name] = [(value, others)] else: self.add_metadata(namespace, name, value, others) def get_item_with_id(self, uid): """ Returns item for defined UID. >>> book.get_item_with_id('image_001') :Args: - uid: UID for the item :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.id == uid: return item return None def get_item_with_href(self, href): """ Returns item for defined HREF. >>> book.get_item_with_href('EPUB/document.xhtml') :Args: - href: HREF for the item we are searching for :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.get_name() == href: return item return None def get_items(self): """ Returns all items attached to this book. :Returns: Returns all items as tuple. """ return (item for item in self.items) def get_items_of_type(self, item_type): """ Returns all items of specified type. >>> book.get_items_of_type(epub.ITEM_IMAGE) :Args: - item_type: Type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.get_type() == item_type) def get_items_of_media_type(self, media_type): """ Returns all items of specified media type. :Args: - media_type: Media type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.media_type == media_type) def set_template(self, name, value): """ Defines templates which are used to generate certain types of pages. When defining new value for the template we have to use content of type 'str' (Python 2) or 'bytes' (Python 3). At the moment we use these templates: - ncx - nav - chapter - cover :Args: - name: Name for the template - value: Content for the template """ self.templates[name] = value def get_template(self, name): """ Returns value for the template. :Args: - name: template name :Returns: Value of the template. """ return self.templates.get(name) def add_prefix(self, name, uri): """ Appends custom prefix to be added to the content.opf document >>> epub_book.add_prefix('bkterms', 'http://booktype.org/') :Args: - name: namespave name - uri: URI for the namespace """ self.prefixes.append('%s: %s' % (name, uri))
aerkalov/ebooklib
ebooklib/epub.py
EpubBook.get_item_with_id
python
def get_item_with_id(self, uid): for item in self.get_items(): if item.id == uid: return item return None
Returns item for defined UID. >>> book.get_item_with_id('image_001') :Args: - uid: UID for the item :Returns: Returns item object. Returns None if nothing was found.
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L745-L761
[ "def get_items(self):\n \"\"\"\n Returns all items attached to this book.\n\n :Returns:\n Returns all items as tuple.\n \"\"\"\n return (item for item in self.items)\n" ]
class EpubBook(object): def __init__(self): self.EPUB_VERSION = None self.reset() # we should have options here def reset(self): "Initialises all needed variables to default values" self.metadata = {} self.items = [] self.spine = [] self.guide = [] self.pages = [] self.toc = [] self.bindings = [] self.IDENTIFIER_ID = 'id' self.FOLDER_NAME = 'EPUB' self._id_html = 0 self._id_image = 0 self._id_static = 0 self.title = '' self.language = 'en' self.direction = None self.templates = { 'ncx': NCX_XML, 'nav': NAV_XML, 'chapter': CHAPTER_XML, 'cover': COVER_XML } self.add_metadata('OPF', 'generator', '', { 'name': 'generator', 'content': 'Ebook-lib %s' % '.'.join([str(s) for s in VERSION]) }) # default to using a randomly-unique identifier if one is not specified manually self.set_identifier(str(uuid.uuid4())) # custom prefixes and namespaces to be set to the content.opf doc self.prefixes = [] self.namespaces = {} def set_identifier(self, uid): """ Sets unique id for this epub :Args: - uid: Value of unique identifier for this book """ self.uid = uid self.set_unique_metadata('DC', 'identifier', self.uid, {'id': self.IDENTIFIER_ID}) def set_title(self, title): """ Set title. You can set multiple titles. :Args: - title: Title value """ self.title = title self.add_metadata('DC', 'title', self.title) def set_language(self, lang): """ Set language for this epub. You can set multiple languages. Specific items in the book can have different language settings. :Args: - lang: Language code """ self.language = lang self.add_metadata('DC', 'language', lang) def set_direction(self, direction): """ :Args: - direction: Options are "ltr", "rtl" and "default" """ self.direction = direction def set_cover(self, file_name, content, create_page=True): """ Set cover and create cover document if needed. :Args: - file_name: file name of the cover page - content: Content for the cover image - create_page: Should cover page be defined. Defined as bool value (optional). Default value is True. """ # as it is now, it can only be called once c0 = EpubCover(file_name=file_name) c0.content = content self.add_item(c0) if create_page: c1 = EpubCoverHtml(image_name=file_name) self.add_item(c1) self.add_metadata(None, 'meta', '', OrderedDict([('name', 'cover'), ('content', 'cover-img')])) def add_author(self, author, file_as=None, role=None, uid='creator'): "Add author for this document" self.add_metadata('DC', 'creator', author, {'id': uid}) if file_as: self.add_metadata(None, 'meta', file_as, {'refines': '#' + uid, 'property': 'file-as', 'scheme': 'marc:relators'}) if role: self.add_metadata(None, 'meta', role, {'refines': '#' + uid, 'property': 'role', 'scheme': 'marc:relators'}) def add_metadata(self, namespace, name, value, others=None): "Add metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace not in self.metadata: self.metadata[namespace] = {} if name not in self.metadata[namespace]: self.metadata[namespace][name] = [] self.metadata[namespace][name].append((value, others)) def get_metadata(self, namespace, name): "Retrieve metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] return self.metadata[namespace].get(name, []) def set_unique_metadata(self, namespace, name, value, others=None): "Add metadata if metadata with this identifier does not already exist, otherwise update existing metadata." if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace in self.metadata and name in self.metadata[namespace]: self.metadata[namespace][name] = [(value, others)] else: self.add_metadata(namespace, name, value, others) def add_item(self, item): """ Add additional item to the book. If not defined, media type and chapter id will be defined for the item. :Args: - item: Item instance """ if item.media_type == '': (has_guessed, media_type) = guess_type(item.get_name().lower()) if has_guessed: if media_type is not None: item.media_type = media_type else: item.media_type = has_guessed else: item.media_type = 'application/octet-stream' if not item.get_id(): # make chapter_, image_ and static_ configurable if isinstance(item, EpubHtml): item.id = 'chapter_%d' % self._id_html self._id_html += 1 # If there's a page list, append it to the book's page list self.pages += item.pages elif isinstance(item, EpubImage): item.id = 'image_%d' % self._id_image self._id_image += 1 else: item.id = 'static_%d' % self._id_image self._id_image += 1 item.book = self self.items.append(item) return item def get_item_with_href(self, href): """ Returns item for defined HREF. >>> book.get_item_with_href('EPUB/document.xhtml') :Args: - href: HREF for the item we are searching for :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.get_name() == href: return item return None def get_items(self): """ Returns all items attached to this book. :Returns: Returns all items as tuple. """ return (item for item in self.items) def get_items_of_type(self, item_type): """ Returns all items of specified type. >>> book.get_items_of_type(epub.ITEM_IMAGE) :Args: - item_type: Type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.get_type() == item_type) def get_items_of_media_type(self, media_type): """ Returns all items of specified media type. :Args: - media_type: Media type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.media_type == media_type) def set_template(self, name, value): """ Defines templates which are used to generate certain types of pages. When defining new value for the template we have to use content of type 'str' (Python 2) or 'bytes' (Python 3). At the moment we use these templates: - ncx - nav - chapter - cover :Args: - name: Name for the template - value: Content for the template """ self.templates[name] = value def get_template(self, name): """ Returns value for the template. :Args: - name: template name :Returns: Value of the template. """ return self.templates.get(name) def add_prefix(self, name, uri): """ Appends custom prefix to be added to the content.opf document >>> epub_book.add_prefix('bkterms', 'http://booktype.org/') :Args: - name: namespave name - uri: URI for the namespace """ self.prefixes.append('%s: %s' % (name, uri))
aerkalov/ebooklib
ebooklib/epub.py
EpubBook.get_item_with_href
python
def get_item_with_href(self, href): for item in self.get_items(): if item.get_name() == href: return item return None
Returns item for defined HREF. >>> book.get_item_with_href('EPUB/document.xhtml') :Args: - href: HREF for the item we are searching for :Returns: Returns item object. Returns None if nothing was found.
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L763-L779
[ "def get_items(self):\n \"\"\"\n Returns all items attached to this book.\n\n :Returns:\n Returns all items as tuple.\n \"\"\"\n return (item for item in self.items)\n" ]
class EpubBook(object): def __init__(self): self.EPUB_VERSION = None self.reset() # we should have options here def reset(self): "Initialises all needed variables to default values" self.metadata = {} self.items = [] self.spine = [] self.guide = [] self.pages = [] self.toc = [] self.bindings = [] self.IDENTIFIER_ID = 'id' self.FOLDER_NAME = 'EPUB' self._id_html = 0 self._id_image = 0 self._id_static = 0 self.title = '' self.language = 'en' self.direction = None self.templates = { 'ncx': NCX_XML, 'nav': NAV_XML, 'chapter': CHAPTER_XML, 'cover': COVER_XML } self.add_metadata('OPF', 'generator', '', { 'name': 'generator', 'content': 'Ebook-lib %s' % '.'.join([str(s) for s in VERSION]) }) # default to using a randomly-unique identifier if one is not specified manually self.set_identifier(str(uuid.uuid4())) # custom prefixes and namespaces to be set to the content.opf doc self.prefixes = [] self.namespaces = {} def set_identifier(self, uid): """ Sets unique id for this epub :Args: - uid: Value of unique identifier for this book """ self.uid = uid self.set_unique_metadata('DC', 'identifier', self.uid, {'id': self.IDENTIFIER_ID}) def set_title(self, title): """ Set title. You can set multiple titles. :Args: - title: Title value """ self.title = title self.add_metadata('DC', 'title', self.title) def set_language(self, lang): """ Set language for this epub. You can set multiple languages. Specific items in the book can have different language settings. :Args: - lang: Language code """ self.language = lang self.add_metadata('DC', 'language', lang) def set_direction(self, direction): """ :Args: - direction: Options are "ltr", "rtl" and "default" """ self.direction = direction def set_cover(self, file_name, content, create_page=True): """ Set cover and create cover document if needed. :Args: - file_name: file name of the cover page - content: Content for the cover image - create_page: Should cover page be defined. Defined as bool value (optional). Default value is True. """ # as it is now, it can only be called once c0 = EpubCover(file_name=file_name) c0.content = content self.add_item(c0) if create_page: c1 = EpubCoverHtml(image_name=file_name) self.add_item(c1) self.add_metadata(None, 'meta', '', OrderedDict([('name', 'cover'), ('content', 'cover-img')])) def add_author(self, author, file_as=None, role=None, uid='creator'): "Add author for this document" self.add_metadata('DC', 'creator', author, {'id': uid}) if file_as: self.add_metadata(None, 'meta', file_as, {'refines': '#' + uid, 'property': 'file-as', 'scheme': 'marc:relators'}) if role: self.add_metadata(None, 'meta', role, {'refines': '#' + uid, 'property': 'role', 'scheme': 'marc:relators'}) def add_metadata(self, namespace, name, value, others=None): "Add metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace not in self.metadata: self.metadata[namespace] = {} if name not in self.metadata[namespace]: self.metadata[namespace][name] = [] self.metadata[namespace][name].append((value, others)) def get_metadata(self, namespace, name): "Retrieve metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] return self.metadata[namespace].get(name, []) def set_unique_metadata(self, namespace, name, value, others=None): "Add metadata if metadata with this identifier does not already exist, otherwise update existing metadata." if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace in self.metadata and name in self.metadata[namespace]: self.metadata[namespace][name] = [(value, others)] else: self.add_metadata(namespace, name, value, others) def add_item(self, item): """ Add additional item to the book. If not defined, media type and chapter id will be defined for the item. :Args: - item: Item instance """ if item.media_type == '': (has_guessed, media_type) = guess_type(item.get_name().lower()) if has_guessed: if media_type is not None: item.media_type = media_type else: item.media_type = has_guessed else: item.media_type = 'application/octet-stream' if not item.get_id(): # make chapter_, image_ and static_ configurable if isinstance(item, EpubHtml): item.id = 'chapter_%d' % self._id_html self._id_html += 1 # If there's a page list, append it to the book's page list self.pages += item.pages elif isinstance(item, EpubImage): item.id = 'image_%d' % self._id_image self._id_image += 1 else: item.id = 'static_%d' % self._id_image self._id_image += 1 item.book = self self.items.append(item) return item def get_item_with_id(self, uid): """ Returns item for defined UID. >>> book.get_item_with_id('image_001') :Args: - uid: UID for the item :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.id == uid: return item return None def get_items(self): """ Returns all items attached to this book. :Returns: Returns all items as tuple. """ return (item for item in self.items) def get_items_of_type(self, item_type): """ Returns all items of specified type. >>> book.get_items_of_type(epub.ITEM_IMAGE) :Args: - item_type: Type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.get_type() == item_type) def get_items_of_media_type(self, media_type): """ Returns all items of specified media type. :Args: - media_type: Media type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.media_type == media_type) def set_template(self, name, value): """ Defines templates which are used to generate certain types of pages. When defining new value for the template we have to use content of type 'str' (Python 2) or 'bytes' (Python 3). At the moment we use these templates: - ncx - nav - chapter - cover :Args: - name: Name for the template - value: Content for the template """ self.templates[name] = value def get_template(self, name): """ Returns value for the template. :Args: - name: template name :Returns: Value of the template. """ return self.templates.get(name) def add_prefix(self, name, uri): """ Appends custom prefix to be added to the content.opf document >>> epub_book.add_prefix('bkterms', 'http://booktype.org/') :Args: - name: namespave name - uri: URI for the namespace """ self.prefixes.append('%s: %s' % (name, uri))
aerkalov/ebooklib
ebooklib/epub.py
EpubBook.get_items_of_type
python
def get_items_of_type(self, item_type): return (item for item in self.items if item.get_type() == item_type)
Returns all items of specified type. >>> book.get_items_of_type(epub.ITEM_IMAGE) :Args: - item_type: Type for items we are searching for :Returns: Returns found items as tuple.
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L790-L802
null
class EpubBook(object): def __init__(self): self.EPUB_VERSION = None self.reset() # we should have options here def reset(self): "Initialises all needed variables to default values" self.metadata = {} self.items = [] self.spine = [] self.guide = [] self.pages = [] self.toc = [] self.bindings = [] self.IDENTIFIER_ID = 'id' self.FOLDER_NAME = 'EPUB' self._id_html = 0 self._id_image = 0 self._id_static = 0 self.title = '' self.language = 'en' self.direction = None self.templates = { 'ncx': NCX_XML, 'nav': NAV_XML, 'chapter': CHAPTER_XML, 'cover': COVER_XML } self.add_metadata('OPF', 'generator', '', { 'name': 'generator', 'content': 'Ebook-lib %s' % '.'.join([str(s) for s in VERSION]) }) # default to using a randomly-unique identifier if one is not specified manually self.set_identifier(str(uuid.uuid4())) # custom prefixes and namespaces to be set to the content.opf doc self.prefixes = [] self.namespaces = {} def set_identifier(self, uid): """ Sets unique id for this epub :Args: - uid: Value of unique identifier for this book """ self.uid = uid self.set_unique_metadata('DC', 'identifier', self.uid, {'id': self.IDENTIFIER_ID}) def set_title(self, title): """ Set title. You can set multiple titles. :Args: - title: Title value """ self.title = title self.add_metadata('DC', 'title', self.title) def set_language(self, lang): """ Set language for this epub. You can set multiple languages. Specific items in the book can have different language settings. :Args: - lang: Language code """ self.language = lang self.add_metadata('DC', 'language', lang) def set_direction(self, direction): """ :Args: - direction: Options are "ltr", "rtl" and "default" """ self.direction = direction def set_cover(self, file_name, content, create_page=True): """ Set cover and create cover document if needed. :Args: - file_name: file name of the cover page - content: Content for the cover image - create_page: Should cover page be defined. Defined as bool value (optional). Default value is True. """ # as it is now, it can only be called once c0 = EpubCover(file_name=file_name) c0.content = content self.add_item(c0) if create_page: c1 = EpubCoverHtml(image_name=file_name) self.add_item(c1) self.add_metadata(None, 'meta', '', OrderedDict([('name', 'cover'), ('content', 'cover-img')])) def add_author(self, author, file_as=None, role=None, uid='creator'): "Add author for this document" self.add_metadata('DC', 'creator', author, {'id': uid}) if file_as: self.add_metadata(None, 'meta', file_as, {'refines': '#' + uid, 'property': 'file-as', 'scheme': 'marc:relators'}) if role: self.add_metadata(None, 'meta', role, {'refines': '#' + uid, 'property': 'role', 'scheme': 'marc:relators'}) def add_metadata(self, namespace, name, value, others=None): "Add metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace not in self.metadata: self.metadata[namespace] = {} if name not in self.metadata[namespace]: self.metadata[namespace][name] = [] self.metadata[namespace][name].append((value, others)) def get_metadata(self, namespace, name): "Retrieve metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] return self.metadata[namespace].get(name, []) def set_unique_metadata(self, namespace, name, value, others=None): "Add metadata if metadata with this identifier does not already exist, otherwise update existing metadata." if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace in self.metadata and name in self.metadata[namespace]: self.metadata[namespace][name] = [(value, others)] else: self.add_metadata(namespace, name, value, others) def add_item(self, item): """ Add additional item to the book. If not defined, media type and chapter id will be defined for the item. :Args: - item: Item instance """ if item.media_type == '': (has_guessed, media_type) = guess_type(item.get_name().lower()) if has_guessed: if media_type is not None: item.media_type = media_type else: item.media_type = has_guessed else: item.media_type = 'application/octet-stream' if not item.get_id(): # make chapter_, image_ and static_ configurable if isinstance(item, EpubHtml): item.id = 'chapter_%d' % self._id_html self._id_html += 1 # If there's a page list, append it to the book's page list self.pages += item.pages elif isinstance(item, EpubImage): item.id = 'image_%d' % self._id_image self._id_image += 1 else: item.id = 'static_%d' % self._id_image self._id_image += 1 item.book = self self.items.append(item) return item def get_item_with_id(self, uid): """ Returns item for defined UID. >>> book.get_item_with_id('image_001') :Args: - uid: UID for the item :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.id == uid: return item return None def get_item_with_href(self, href): """ Returns item for defined HREF. >>> book.get_item_with_href('EPUB/document.xhtml') :Args: - href: HREF for the item we are searching for :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.get_name() == href: return item return None def get_items(self): """ Returns all items attached to this book. :Returns: Returns all items as tuple. """ return (item for item in self.items) def get_items_of_media_type(self, media_type): """ Returns all items of specified media type. :Args: - media_type: Media type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.media_type == media_type) def set_template(self, name, value): """ Defines templates which are used to generate certain types of pages. When defining new value for the template we have to use content of type 'str' (Python 2) or 'bytes' (Python 3). At the moment we use these templates: - ncx - nav - chapter - cover :Args: - name: Name for the template - value: Content for the template """ self.templates[name] = value def get_template(self, name): """ Returns value for the template. :Args: - name: template name :Returns: Value of the template. """ return self.templates.get(name) def add_prefix(self, name, uri): """ Appends custom prefix to be added to the content.opf document >>> epub_book.add_prefix('bkterms', 'http://booktype.org/') :Args: - name: namespave name - uri: URI for the namespace """ self.prefixes.append('%s: %s' % (name, uri))
aerkalov/ebooklib
ebooklib/epub.py
EpubBook.get_items_of_media_type
python
def get_items_of_media_type(self, media_type): return (item for item in self.items if item.media_type == media_type)
Returns all items of specified media type. :Args: - media_type: Media type for items we are searching for :Returns: Returns found items as tuple.
train
https://github.com/aerkalov/ebooklib/blob/305f2dd7f02923ffabf9586a5d16266113d00c4a/ebooklib/epub.py#L804-L814
null
class EpubBook(object): def __init__(self): self.EPUB_VERSION = None self.reset() # we should have options here def reset(self): "Initialises all needed variables to default values" self.metadata = {} self.items = [] self.spine = [] self.guide = [] self.pages = [] self.toc = [] self.bindings = [] self.IDENTIFIER_ID = 'id' self.FOLDER_NAME = 'EPUB' self._id_html = 0 self._id_image = 0 self._id_static = 0 self.title = '' self.language = 'en' self.direction = None self.templates = { 'ncx': NCX_XML, 'nav': NAV_XML, 'chapter': CHAPTER_XML, 'cover': COVER_XML } self.add_metadata('OPF', 'generator', '', { 'name': 'generator', 'content': 'Ebook-lib %s' % '.'.join([str(s) for s in VERSION]) }) # default to using a randomly-unique identifier if one is not specified manually self.set_identifier(str(uuid.uuid4())) # custom prefixes and namespaces to be set to the content.opf doc self.prefixes = [] self.namespaces = {} def set_identifier(self, uid): """ Sets unique id for this epub :Args: - uid: Value of unique identifier for this book """ self.uid = uid self.set_unique_metadata('DC', 'identifier', self.uid, {'id': self.IDENTIFIER_ID}) def set_title(self, title): """ Set title. You can set multiple titles. :Args: - title: Title value """ self.title = title self.add_metadata('DC', 'title', self.title) def set_language(self, lang): """ Set language for this epub. You can set multiple languages. Specific items in the book can have different language settings. :Args: - lang: Language code """ self.language = lang self.add_metadata('DC', 'language', lang) def set_direction(self, direction): """ :Args: - direction: Options are "ltr", "rtl" and "default" """ self.direction = direction def set_cover(self, file_name, content, create_page=True): """ Set cover and create cover document if needed. :Args: - file_name: file name of the cover page - content: Content for the cover image - create_page: Should cover page be defined. Defined as bool value (optional). Default value is True. """ # as it is now, it can only be called once c0 = EpubCover(file_name=file_name) c0.content = content self.add_item(c0) if create_page: c1 = EpubCoverHtml(image_name=file_name) self.add_item(c1) self.add_metadata(None, 'meta', '', OrderedDict([('name', 'cover'), ('content', 'cover-img')])) def add_author(self, author, file_as=None, role=None, uid='creator'): "Add author for this document" self.add_metadata('DC', 'creator', author, {'id': uid}) if file_as: self.add_metadata(None, 'meta', file_as, {'refines': '#' + uid, 'property': 'file-as', 'scheme': 'marc:relators'}) if role: self.add_metadata(None, 'meta', role, {'refines': '#' + uid, 'property': 'role', 'scheme': 'marc:relators'}) def add_metadata(self, namespace, name, value, others=None): "Add metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace not in self.metadata: self.metadata[namespace] = {} if name not in self.metadata[namespace]: self.metadata[namespace][name] = [] self.metadata[namespace][name].append((value, others)) def get_metadata(self, namespace, name): "Retrieve metadata" if namespace in NAMESPACES: namespace = NAMESPACES[namespace] return self.metadata[namespace].get(name, []) def set_unique_metadata(self, namespace, name, value, others=None): "Add metadata if metadata with this identifier does not already exist, otherwise update existing metadata." if namespace in NAMESPACES: namespace = NAMESPACES[namespace] if namespace in self.metadata and name in self.metadata[namespace]: self.metadata[namespace][name] = [(value, others)] else: self.add_metadata(namespace, name, value, others) def add_item(self, item): """ Add additional item to the book. If not defined, media type and chapter id will be defined for the item. :Args: - item: Item instance """ if item.media_type == '': (has_guessed, media_type) = guess_type(item.get_name().lower()) if has_guessed: if media_type is not None: item.media_type = media_type else: item.media_type = has_guessed else: item.media_type = 'application/octet-stream' if not item.get_id(): # make chapter_, image_ and static_ configurable if isinstance(item, EpubHtml): item.id = 'chapter_%d' % self._id_html self._id_html += 1 # If there's a page list, append it to the book's page list self.pages += item.pages elif isinstance(item, EpubImage): item.id = 'image_%d' % self._id_image self._id_image += 1 else: item.id = 'static_%d' % self._id_image self._id_image += 1 item.book = self self.items.append(item) return item def get_item_with_id(self, uid): """ Returns item for defined UID. >>> book.get_item_with_id('image_001') :Args: - uid: UID for the item :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.id == uid: return item return None def get_item_with_href(self, href): """ Returns item for defined HREF. >>> book.get_item_with_href('EPUB/document.xhtml') :Args: - href: HREF for the item we are searching for :Returns: Returns item object. Returns None if nothing was found. """ for item in self.get_items(): if item.get_name() == href: return item return None def get_items(self): """ Returns all items attached to this book. :Returns: Returns all items as tuple. """ return (item for item in self.items) def get_items_of_type(self, item_type): """ Returns all items of specified type. >>> book.get_items_of_type(epub.ITEM_IMAGE) :Args: - item_type: Type for items we are searching for :Returns: Returns found items as tuple. """ return (item for item in self.items if item.get_type() == item_type) def set_template(self, name, value): """ Defines templates which are used to generate certain types of pages. When defining new value for the template we have to use content of type 'str' (Python 2) or 'bytes' (Python 3). At the moment we use these templates: - ncx - nav - chapter - cover :Args: - name: Name for the template - value: Content for the template """ self.templates[name] = value def get_template(self, name): """ Returns value for the template. :Args: - name: template name :Returns: Value of the template. """ return self.templates.get(name) def add_prefix(self, name, uri): """ Appends custom prefix to be added to the content.opf document >>> epub_book.add_prefix('bkterms', 'http://booktype.org/') :Args: - name: namespave name - uri: URI for the namespace """ self.prefixes.append('%s: %s' % (name, uri))
MaxHalford/prince
prince/mfa.py
MFA.row_coordinates
python
def row_coordinates(self, X): utils.validation.check_is_fitted(self, 's_') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) return self._row_coordinates_from_global(self._build_X_global(X))
Returns the row principal coordinates.
train
https://github.com/MaxHalford/prince/blob/714c9cdfc4d9f8823eabf550a23ad01fe87c50d7/prince/mfa.py#L132-L143
null
class MFA(pca.PCA): def __init__(self, groups=None, normalize=True, n_components=2, n_iter=10, copy=True, check_input=True, random_state=None, engine='auto'): super().__init__( rescale_with_mean=False, rescale_with_std=False, n_components=n_components, n_iter=n_iter, copy=copy, check_input=check_input, random_state=random_state, engine=engine ) self.groups = groups self.normalize = normalize def fit(self, X, y=None): # Checks groups are provided if self.groups is None: raise ValueError('Groups have to be specified') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) # Check group types are consistent self.all_nums_ = {} for name, cols in sorted(self.groups.items()): all_num = all(pd.api.types.is_numeric_dtype(X[c]) for c in cols) all_cat = all(pd.api.types.is_string_dtype(X[c]) for c in cols) if not (all_num or all_cat): raise ValueError('Not all columns in "{}" group are of the same type'.format(name)) self.all_nums_[name] = all_num # Run a factor analysis in each group self.partial_factor_analysis_ = {} for name, cols in sorted(self.groups.items()): if self.all_nums_[name]: fa = pca.PCA( rescale_with_mean=False, rescale_with_std=False, n_components=self.n_components, n_iter=self.n_iter, copy=True, random_state=self.random_state, engine=self.engine ) else: fa = mca.MCA( n_components=self.n_components, n_iter=self.n_iter, copy=self.copy, random_state=self.random_state, engine=self.engine ) self.partial_factor_analysis_[name] = fa.fit(X.loc[:, cols]) # Fit the global PCA self.cat_one_hots_ = {} super().fit(self._build_X_global(X)) return self def _prepare_input(self, X): # Make sure X is a DataFrame for convenience if not isinstance(X, pd.DataFrame): X = pd.DataFrame(X) # Copy data if self.copy: X = X.copy() if self.normalize: # Scale continuous variables to unit variance num = X.select_dtypes(np.number).columns # If a column's cardinality is 1 then it's variance is 0 which can # can cause a division by 0 normalize = lambda x: x / (np.sqrt((x ** 2).sum()) or 1) X.loc[:, num] = (X.loc[:, num] - X.loc[:, num].mean()).apply(normalize, axis='rows') return X def _build_X_global(self, X): X_partials = [] for name, cols in sorted(self.groups.items()): X_partial = X.loc[:, cols] if not self.all_nums_[name]: if name in self.cat_one_hots_: oh = self.cat_one_hots_[name] else: oh = one_hot.OneHotEncoder().fit(X_partial) X_partial = oh.transform(X_partial) self.cat_one_hots_[name] = oh X_partials.append(X_partial / self.partial_factor_analysis_[name].s_[0]) X_global = pd.concat(X_partials, axis='columns') X_global.index = X.index return X_global def transform(self, X): """Returns the row principal coordinates of a dataset.""" return self.row_coordinates(X) def _row_coordinates_from_global(self, X_global): """Returns the row principal coordinates.""" return len(X_global) ** 0.5 * super().row_coordinates(X_global) def row_contributions(self, X): """Returns the row contributions towards each principal component.""" utils.validation.check_is_fitted(self, 's_') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) return super().row_contributions(self._build_X_global(X)) def partial_row_coordinates(self, X): """Returns the row coordinates for each group.""" utils.validation.check_is_fitted(self, 's_') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) # Define the projection matrix P P = len(X) ** 0.5 * self.U_ / self.s_ # Get the projections for each group coords = {} for name, cols in sorted(self.groups.items()): X_partial = X.loc[:, cols] if not self.all_nums_[name]: X_partial = self.cat_one_hots_[name].transform(X_partial) Z_partial = X_partial / self.partial_factor_analysis_[name].s_[0] coords[name] = len(self.groups) * (Z_partial @ Z_partial.T) @ P # Convert coords to a MultiIndex DataFrame coords = pd.DataFrame({ (name, i): group_coords.loc[:, i] for name, group_coords in coords.items() for i in range(group_coords.shape[1]) }) return coords def column_correlations(self, X): """Returns the column correlations.""" utils.validation.check_is_fitted(self, 's_') X_global = self._build_X_global(X) row_pc = self._row_coordinates_from_global(X_global) return pd.DataFrame({ component: { feature: row_pc[component].corr(X_global[feature].to_dense()) for feature in X_global.columns } for component in row_pc.columns }) def plot_partial_row_coordinates(self, X, ax=None, figsize=(6, 6), x_component=0, y_component=1, color_labels=None, **kwargs): """Plot the row principal coordinates.""" utils.validation.check_is_fitted(self, 's_') if ax is None: fig, ax = plt.subplots(figsize=figsize) # Add plotting style ax = plot.stylize_axis(ax) # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) # Retrieve partial coordinates coords = self.partial_row_coordinates(X) # Determine the color of each group if there are group labels if color_labels is not None: colors = {g: ax._get_lines.get_next_color() for g in sorted(list(set(color_labels)))} # Get the list of all possible markers marks = itertools.cycle(list(markers.MarkerStyle.markers.keys())) next(marks) # The first marker looks pretty shit so we skip it # Plot points for name in self.groups: mark = next(marks) x = coords[name][x_component] y = coords[name][y_component] if color_labels is None: ax.scatter(x, y, marker=mark, label=name, **kwargs) continue for color_label, color in sorted(colors.items()): mask = np.array(color_labels) == color_label label = '{} - {}'.format(name, color_label) ax.scatter(x[mask], y[mask], marker=mark, color=color, label=label, **kwargs) # Legend ax.legend() # Text ax.set_title('Partial row principal coordinates') ei = self.explained_inertia_ ax.set_xlabel('Component {} ({:.2f}% inertia)'.format(x_component, 100 * ei[x_component])) ax.set_ylabel('Component {} ({:.2f}% inertia)'.format(y_component, 100 * ei[y_component])) return ax
MaxHalford/prince
prince/mfa.py
MFA.row_contributions
python
def row_contributions(self, X): utils.validation.check_is_fitted(self, 's_') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) return super().row_contributions(self._build_X_global(X))
Returns the row contributions towards each principal component.
train
https://github.com/MaxHalford/prince/blob/714c9cdfc4d9f8823eabf550a23ad01fe87c50d7/prince/mfa.py#L145-L156
null
class MFA(pca.PCA): def __init__(self, groups=None, normalize=True, n_components=2, n_iter=10, copy=True, check_input=True, random_state=None, engine='auto'): super().__init__( rescale_with_mean=False, rescale_with_std=False, n_components=n_components, n_iter=n_iter, copy=copy, check_input=check_input, random_state=random_state, engine=engine ) self.groups = groups self.normalize = normalize def fit(self, X, y=None): # Checks groups are provided if self.groups is None: raise ValueError('Groups have to be specified') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) # Check group types are consistent self.all_nums_ = {} for name, cols in sorted(self.groups.items()): all_num = all(pd.api.types.is_numeric_dtype(X[c]) for c in cols) all_cat = all(pd.api.types.is_string_dtype(X[c]) for c in cols) if not (all_num or all_cat): raise ValueError('Not all columns in "{}" group are of the same type'.format(name)) self.all_nums_[name] = all_num # Run a factor analysis in each group self.partial_factor_analysis_ = {} for name, cols in sorted(self.groups.items()): if self.all_nums_[name]: fa = pca.PCA( rescale_with_mean=False, rescale_with_std=False, n_components=self.n_components, n_iter=self.n_iter, copy=True, random_state=self.random_state, engine=self.engine ) else: fa = mca.MCA( n_components=self.n_components, n_iter=self.n_iter, copy=self.copy, random_state=self.random_state, engine=self.engine ) self.partial_factor_analysis_[name] = fa.fit(X.loc[:, cols]) # Fit the global PCA self.cat_one_hots_ = {} super().fit(self._build_X_global(X)) return self def _prepare_input(self, X): # Make sure X is a DataFrame for convenience if not isinstance(X, pd.DataFrame): X = pd.DataFrame(X) # Copy data if self.copy: X = X.copy() if self.normalize: # Scale continuous variables to unit variance num = X.select_dtypes(np.number).columns # If a column's cardinality is 1 then it's variance is 0 which can # can cause a division by 0 normalize = lambda x: x / (np.sqrt((x ** 2).sum()) or 1) X.loc[:, num] = (X.loc[:, num] - X.loc[:, num].mean()).apply(normalize, axis='rows') return X def _build_X_global(self, X): X_partials = [] for name, cols in sorted(self.groups.items()): X_partial = X.loc[:, cols] if not self.all_nums_[name]: if name in self.cat_one_hots_: oh = self.cat_one_hots_[name] else: oh = one_hot.OneHotEncoder().fit(X_partial) X_partial = oh.transform(X_partial) self.cat_one_hots_[name] = oh X_partials.append(X_partial / self.partial_factor_analysis_[name].s_[0]) X_global = pd.concat(X_partials, axis='columns') X_global.index = X.index return X_global def transform(self, X): """Returns the row principal coordinates of a dataset.""" return self.row_coordinates(X) def _row_coordinates_from_global(self, X_global): """Returns the row principal coordinates.""" return len(X_global) ** 0.5 * super().row_coordinates(X_global) def row_coordinates(self, X): """Returns the row principal coordinates.""" utils.validation.check_is_fitted(self, 's_') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) return self._row_coordinates_from_global(self._build_X_global(X)) def partial_row_coordinates(self, X): """Returns the row coordinates for each group.""" utils.validation.check_is_fitted(self, 's_') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) # Define the projection matrix P P = len(X) ** 0.5 * self.U_ / self.s_ # Get the projections for each group coords = {} for name, cols in sorted(self.groups.items()): X_partial = X.loc[:, cols] if not self.all_nums_[name]: X_partial = self.cat_one_hots_[name].transform(X_partial) Z_partial = X_partial / self.partial_factor_analysis_[name].s_[0] coords[name] = len(self.groups) * (Z_partial @ Z_partial.T) @ P # Convert coords to a MultiIndex DataFrame coords = pd.DataFrame({ (name, i): group_coords.loc[:, i] for name, group_coords in coords.items() for i in range(group_coords.shape[1]) }) return coords def column_correlations(self, X): """Returns the column correlations.""" utils.validation.check_is_fitted(self, 's_') X_global = self._build_X_global(X) row_pc = self._row_coordinates_from_global(X_global) return pd.DataFrame({ component: { feature: row_pc[component].corr(X_global[feature].to_dense()) for feature in X_global.columns } for component in row_pc.columns }) def plot_partial_row_coordinates(self, X, ax=None, figsize=(6, 6), x_component=0, y_component=1, color_labels=None, **kwargs): """Plot the row principal coordinates.""" utils.validation.check_is_fitted(self, 's_') if ax is None: fig, ax = plt.subplots(figsize=figsize) # Add plotting style ax = plot.stylize_axis(ax) # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) # Retrieve partial coordinates coords = self.partial_row_coordinates(X) # Determine the color of each group if there are group labels if color_labels is not None: colors = {g: ax._get_lines.get_next_color() for g in sorted(list(set(color_labels)))} # Get the list of all possible markers marks = itertools.cycle(list(markers.MarkerStyle.markers.keys())) next(marks) # The first marker looks pretty shit so we skip it # Plot points for name in self.groups: mark = next(marks) x = coords[name][x_component] y = coords[name][y_component] if color_labels is None: ax.scatter(x, y, marker=mark, label=name, **kwargs) continue for color_label, color in sorted(colors.items()): mask = np.array(color_labels) == color_label label = '{} - {}'.format(name, color_label) ax.scatter(x[mask], y[mask], marker=mark, color=color, label=label, **kwargs) # Legend ax.legend() # Text ax.set_title('Partial row principal coordinates') ei = self.explained_inertia_ ax.set_xlabel('Component {} ({:.2f}% inertia)'.format(x_component, 100 * ei[x_component])) ax.set_ylabel('Component {} ({:.2f}% inertia)'.format(y_component, 100 * ei[y_component])) return ax
MaxHalford/prince
prince/mfa.py
MFA.partial_row_coordinates
python
def partial_row_coordinates(self, X): utils.validation.check_is_fitted(self, 's_') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) # Define the projection matrix P P = len(X) ** 0.5 * self.U_ / self.s_ # Get the projections for each group coords = {} for name, cols in sorted(self.groups.items()): X_partial = X.loc[:, cols] if not self.all_nums_[name]: X_partial = self.cat_one_hots_[name].transform(X_partial) Z_partial = X_partial / self.partial_factor_analysis_[name].s_[0] coords[name] = len(self.groups) * (Z_partial @ Z_partial.T) @ P # Convert coords to a MultiIndex DataFrame coords = pd.DataFrame({ (name, i): group_coords.loc[:, i] for name, group_coords in coords.items() for i in range(group_coords.shape[1]) }) return coords
Returns the row coordinates for each group.
train
https://github.com/MaxHalford/prince/blob/714c9cdfc4d9f8823eabf550a23ad01fe87c50d7/prince/mfa.py#L158-L190
null
class MFA(pca.PCA): def __init__(self, groups=None, normalize=True, n_components=2, n_iter=10, copy=True, check_input=True, random_state=None, engine='auto'): super().__init__( rescale_with_mean=False, rescale_with_std=False, n_components=n_components, n_iter=n_iter, copy=copy, check_input=check_input, random_state=random_state, engine=engine ) self.groups = groups self.normalize = normalize def fit(self, X, y=None): # Checks groups are provided if self.groups is None: raise ValueError('Groups have to be specified') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) # Check group types are consistent self.all_nums_ = {} for name, cols in sorted(self.groups.items()): all_num = all(pd.api.types.is_numeric_dtype(X[c]) for c in cols) all_cat = all(pd.api.types.is_string_dtype(X[c]) for c in cols) if not (all_num or all_cat): raise ValueError('Not all columns in "{}" group are of the same type'.format(name)) self.all_nums_[name] = all_num # Run a factor analysis in each group self.partial_factor_analysis_ = {} for name, cols in sorted(self.groups.items()): if self.all_nums_[name]: fa = pca.PCA( rescale_with_mean=False, rescale_with_std=False, n_components=self.n_components, n_iter=self.n_iter, copy=True, random_state=self.random_state, engine=self.engine ) else: fa = mca.MCA( n_components=self.n_components, n_iter=self.n_iter, copy=self.copy, random_state=self.random_state, engine=self.engine ) self.partial_factor_analysis_[name] = fa.fit(X.loc[:, cols]) # Fit the global PCA self.cat_one_hots_ = {} super().fit(self._build_X_global(X)) return self def _prepare_input(self, X): # Make sure X is a DataFrame for convenience if not isinstance(X, pd.DataFrame): X = pd.DataFrame(X) # Copy data if self.copy: X = X.copy() if self.normalize: # Scale continuous variables to unit variance num = X.select_dtypes(np.number).columns # If a column's cardinality is 1 then it's variance is 0 which can # can cause a division by 0 normalize = lambda x: x / (np.sqrt((x ** 2).sum()) or 1) X.loc[:, num] = (X.loc[:, num] - X.loc[:, num].mean()).apply(normalize, axis='rows') return X def _build_X_global(self, X): X_partials = [] for name, cols in sorted(self.groups.items()): X_partial = X.loc[:, cols] if not self.all_nums_[name]: if name in self.cat_one_hots_: oh = self.cat_one_hots_[name] else: oh = one_hot.OneHotEncoder().fit(X_partial) X_partial = oh.transform(X_partial) self.cat_one_hots_[name] = oh X_partials.append(X_partial / self.partial_factor_analysis_[name].s_[0]) X_global = pd.concat(X_partials, axis='columns') X_global.index = X.index return X_global def transform(self, X): """Returns the row principal coordinates of a dataset.""" return self.row_coordinates(X) def _row_coordinates_from_global(self, X_global): """Returns the row principal coordinates.""" return len(X_global) ** 0.5 * super().row_coordinates(X_global) def row_coordinates(self, X): """Returns the row principal coordinates.""" utils.validation.check_is_fitted(self, 's_') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) return self._row_coordinates_from_global(self._build_X_global(X)) def row_contributions(self, X): """Returns the row contributions towards each principal component.""" utils.validation.check_is_fitted(self, 's_') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) return super().row_contributions(self._build_X_global(X)) def column_correlations(self, X): """Returns the column correlations.""" utils.validation.check_is_fitted(self, 's_') X_global = self._build_X_global(X) row_pc = self._row_coordinates_from_global(X_global) return pd.DataFrame({ component: { feature: row_pc[component].corr(X_global[feature].to_dense()) for feature in X_global.columns } for component in row_pc.columns }) def plot_partial_row_coordinates(self, X, ax=None, figsize=(6, 6), x_component=0, y_component=1, color_labels=None, **kwargs): """Plot the row principal coordinates.""" utils.validation.check_is_fitted(self, 's_') if ax is None: fig, ax = plt.subplots(figsize=figsize) # Add plotting style ax = plot.stylize_axis(ax) # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) # Retrieve partial coordinates coords = self.partial_row_coordinates(X) # Determine the color of each group if there are group labels if color_labels is not None: colors = {g: ax._get_lines.get_next_color() for g in sorted(list(set(color_labels)))} # Get the list of all possible markers marks = itertools.cycle(list(markers.MarkerStyle.markers.keys())) next(marks) # The first marker looks pretty shit so we skip it # Plot points for name in self.groups: mark = next(marks) x = coords[name][x_component] y = coords[name][y_component] if color_labels is None: ax.scatter(x, y, marker=mark, label=name, **kwargs) continue for color_label, color in sorted(colors.items()): mask = np.array(color_labels) == color_label label = '{} - {}'.format(name, color_label) ax.scatter(x[mask], y[mask], marker=mark, color=color, label=label, **kwargs) # Legend ax.legend() # Text ax.set_title('Partial row principal coordinates') ei = self.explained_inertia_ ax.set_xlabel('Component {} ({:.2f}% inertia)'.format(x_component, 100 * ei[x_component])) ax.set_ylabel('Component {} ({:.2f}% inertia)'.format(y_component, 100 * ei[y_component])) return ax
MaxHalford/prince
prince/mfa.py
MFA.column_correlations
python
def column_correlations(self, X): utils.validation.check_is_fitted(self, 's_') X_global = self._build_X_global(X) row_pc = self._row_coordinates_from_global(X_global) return pd.DataFrame({ component: { feature: row_pc[component].corr(X_global[feature].to_dense()) for feature in X_global.columns } for component in row_pc.columns })
Returns the column correlations.
train
https://github.com/MaxHalford/prince/blob/714c9cdfc4d9f8823eabf550a23ad01fe87c50d7/prince/mfa.py#L192-L205
null
class MFA(pca.PCA): def __init__(self, groups=None, normalize=True, n_components=2, n_iter=10, copy=True, check_input=True, random_state=None, engine='auto'): super().__init__( rescale_with_mean=False, rescale_with_std=False, n_components=n_components, n_iter=n_iter, copy=copy, check_input=check_input, random_state=random_state, engine=engine ) self.groups = groups self.normalize = normalize def fit(self, X, y=None): # Checks groups are provided if self.groups is None: raise ValueError('Groups have to be specified') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) # Check group types are consistent self.all_nums_ = {} for name, cols in sorted(self.groups.items()): all_num = all(pd.api.types.is_numeric_dtype(X[c]) for c in cols) all_cat = all(pd.api.types.is_string_dtype(X[c]) for c in cols) if not (all_num or all_cat): raise ValueError('Not all columns in "{}" group are of the same type'.format(name)) self.all_nums_[name] = all_num # Run a factor analysis in each group self.partial_factor_analysis_ = {} for name, cols in sorted(self.groups.items()): if self.all_nums_[name]: fa = pca.PCA( rescale_with_mean=False, rescale_with_std=False, n_components=self.n_components, n_iter=self.n_iter, copy=True, random_state=self.random_state, engine=self.engine ) else: fa = mca.MCA( n_components=self.n_components, n_iter=self.n_iter, copy=self.copy, random_state=self.random_state, engine=self.engine ) self.partial_factor_analysis_[name] = fa.fit(X.loc[:, cols]) # Fit the global PCA self.cat_one_hots_ = {} super().fit(self._build_X_global(X)) return self def _prepare_input(self, X): # Make sure X is a DataFrame for convenience if not isinstance(X, pd.DataFrame): X = pd.DataFrame(X) # Copy data if self.copy: X = X.copy() if self.normalize: # Scale continuous variables to unit variance num = X.select_dtypes(np.number).columns # If a column's cardinality is 1 then it's variance is 0 which can # can cause a division by 0 normalize = lambda x: x / (np.sqrt((x ** 2).sum()) or 1) X.loc[:, num] = (X.loc[:, num] - X.loc[:, num].mean()).apply(normalize, axis='rows') return X def _build_X_global(self, X): X_partials = [] for name, cols in sorted(self.groups.items()): X_partial = X.loc[:, cols] if not self.all_nums_[name]: if name in self.cat_one_hots_: oh = self.cat_one_hots_[name] else: oh = one_hot.OneHotEncoder().fit(X_partial) X_partial = oh.transform(X_partial) self.cat_one_hots_[name] = oh X_partials.append(X_partial / self.partial_factor_analysis_[name].s_[0]) X_global = pd.concat(X_partials, axis='columns') X_global.index = X.index return X_global def transform(self, X): """Returns the row principal coordinates of a dataset.""" return self.row_coordinates(X) def _row_coordinates_from_global(self, X_global): """Returns the row principal coordinates.""" return len(X_global) ** 0.5 * super().row_coordinates(X_global) def row_coordinates(self, X): """Returns the row principal coordinates.""" utils.validation.check_is_fitted(self, 's_') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) return self._row_coordinates_from_global(self._build_X_global(X)) def row_contributions(self, X): """Returns the row contributions towards each principal component.""" utils.validation.check_is_fitted(self, 's_') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) return super().row_contributions(self._build_X_global(X)) def partial_row_coordinates(self, X): """Returns the row coordinates for each group.""" utils.validation.check_is_fitted(self, 's_') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) # Define the projection matrix P P = len(X) ** 0.5 * self.U_ / self.s_ # Get the projections for each group coords = {} for name, cols in sorted(self.groups.items()): X_partial = X.loc[:, cols] if not self.all_nums_[name]: X_partial = self.cat_one_hots_[name].transform(X_partial) Z_partial = X_partial / self.partial_factor_analysis_[name].s_[0] coords[name] = len(self.groups) * (Z_partial @ Z_partial.T) @ P # Convert coords to a MultiIndex DataFrame coords = pd.DataFrame({ (name, i): group_coords.loc[:, i] for name, group_coords in coords.items() for i in range(group_coords.shape[1]) }) return coords def plot_partial_row_coordinates(self, X, ax=None, figsize=(6, 6), x_component=0, y_component=1, color_labels=None, **kwargs): """Plot the row principal coordinates.""" utils.validation.check_is_fitted(self, 's_') if ax is None: fig, ax = plt.subplots(figsize=figsize) # Add plotting style ax = plot.stylize_axis(ax) # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) # Retrieve partial coordinates coords = self.partial_row_coordinates(X) # Determine the color of each group if there are group labels if color_labels is not None: colors = {g: ax._get_lines.get_next_color() for g in sorted(list(set(color_labels)))} # Get the list of all possible markers marks = itertools.cycle(list(markers.MarkerStyle.markers.keys())) next(marks) # The first marker looks pretty shit so we skip it # Plot points for name in self.groups: mark = next(marks) x = coords[name][x_component] y = coords[name][y_component] if color_labels is None: ax.scatter(x, y, marker=mark, label=name, **kwargs) continue for color_label, color in sorted(colors.items()): mask = np.array(color_labels) == color_label label = '{} - {}'.format(name, color_label) ax.scatter(x[mask], y[mask], marker=mark, color=color, label=label, **kwargs) # Legend ax.legend() # Text ax.set_title('Partial row principal coordinates') ei = self.explained_inertia_ ax.set_xlabel('Component {} ({:.2f}% inertia)'.format(x_component, 100 * ei[x_component])) ax.set_ylabel('Component {} ({:.2f}% inertia)'.format(y_component, 100 * ei[y_component])) return ax
MaxHalford/prince
prince/mfa.py
MFA.plot_partial_row_coordinates
python
def plot_partial_row_coordinates(self, X, ax=None, figsize=(6, 6), x_component=0, y_component=1, color_labels=None, **kwargs): utils.validation.check_is_fitted(self, 's_') if ax is None: fig, ax = plt.subplots(figsize=figsize) # Add plotting style ax = plot.stylize_axis(ax) # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) # Retrieve partial coordinates coords = self.partial_row_coordinates(X) # Determine the color of each group if there are group labels if color_labels is not None: colors = {g: ax._get_lines.get_next_color() for g in sorted(list(set(color_labels)))} # Get the list of all possible markers marks = itertools.cycle(list(markers.MarkerStyle.markers.keys())) next(marks) # The first marker looks pretty shit so we skip it # Plot points for name in self.groups: mark = next(marks) x = coords[name][x_component] y = coords[name][y_component] if color_labels is None: ax.scatter(x, y, marker=mark, label=name, **kwargs) continue for color_label, color in sorted(colors.items()): mask = np.array(color_labels) == color_label label = '{} - {}'.format(name, color_label) ax.scatter(x[mask], y[mask], marker=mark, color=color, label=label, **kwargs) # Legend ax.legend() # Text ax.set_title('Partial row principal coordinates') ei = self.explained_inertia_ ax.set_xlabel('Component {} ({:.2f}% inertia)'.format(x_component, 100 * ei[x_component])) ax.set_ylabel('Component {} ({:.2f}% inertia)'.format(y_component, 100 * ei[y_component])) return ax
Plot the row principal coordinates.
train
https://github.com/MaxHalford/prince/blob/714c9cdfc4d9f8823eabf550a23ad01fe87c50d7/prince/mfa.py#L207-L262
[ "def stylize_axis(ax, grid=True):\n\n if grid:\n ax.grid()\n\n ax.xaxis.set_ticks_position('none')\n ax.yaxis.set_ticks_position('none')\n\n ax.axhline(y=0, linestyle='-', linewidth=1.2, color=GRAY['dark'], alpha=0.6)\n ax.axvline(x=0, linestyle='-', linewidth=1.2, color=GRAY['dark'], alpha=0.6)\n\n return ax\n" ]
class MFA(pca.PCA): def __init__(self, groups=None, normalize=True, n_components=2, n_iter=10, copy=True, check_input=True, random_state=None, engine='auto'): super().__init__( rescale_with_mean=False, rescale_with_std=False, n_components=n_components, n_iter=n_iter, copy=copy, check_input=check_input, random_state=random_state, engine=engine ) self.groups = groups self.normalize = normalize def fit(self, X, y=None): # Checks groups are provided if self.groups is None: raise ValueError('Groups have to be specified') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) # Check group types are consistent self.all_nums_ = {} for name, cols in sorted(self.groups.items()): all_num = all(pd.api.types.is_numeric_dtype(X[c]) for c in cols) all_cat = all(pd.api.types.is_string_dtype(X[c]) for c in cols) if not (all_num or all_cat): raise ValueError('Not all columns in "{}" group are of the same type'.format(name)) self.all_nums_[name] = all_num # Run a factor analysis in each group self.partial_factor_analysis_ = {} for name, cols in sorted(self.groups.items()): if self.all_nums_[name]: fa = pca.PCA( rescale_with_mean=False, rescale_with_std=False, n_components=self.n_components, n_iter=self.n_iter, copy=True, random_state=self.random_state, engine=self.engine ) else: fa = mca.MCA( n_components=self.n_components, n_iter=self.n_iter, copy=self.copy, random_state=self.random_state, engine=self.engine ) self.partial_factor_analysis_[name] = fa.fit(X.loc[:, cols]) # Fit the global PCA self.cat_one_hots_ = {} super().fit(self._build_X_global(X)) return self def _prepare_input(self, X): # Make sure X is a DataFrame for convenience if not isinstance(X, pd.DataFrame): X = pd.DataFrame(X) # Copy data if self.copy: X = X.copy() if self.normalize: # Scale continuous variables to unit variance num = X.select_dtypes(np.number).columns # If a column's cardinality is 1 then it's variance is 0 which can # can cause a division by 0 normalize = lambda x: x / (np.sqrt((x ** 2).sum()) or 1) X.loc[:, num] = (X.loc[:, num] - X.loc[:, num].mean()).apply(normalize, axis='rows') return X def _build_X_global(self, X): X_partials = [] for name, cols in sorted(self.groups.items()): X_partial = X.loc[:, cols] if not self.all_nums_[name]: if name in self.cat_one_hots_: oh = self.cat_one_hots_[name] else: oh = one_hot.OneHotEncoder().fit(X_partial) X_partial = oh.transform(X_partial) self.cat_one_hots_[name] = oh X_partials.append(X_partial / self.partial_factor_analysis_[name].s_[0]) X_global = pd.concat(X_partials, axis='columns') X_global.index = X.index return X_global def transform(self, X): """Returns the row principal coordinates of a dataset.""" return self.row_coordinates(X) def _row_coordinates_from_global(self, X_global): """Returns the row principal coordinates.""" return len(X_global) ** 0.5 * super().row_coordinates(X_global) def row_coordinates(self, X): """Returns the row principal coordinates.""" utils.validation.check_is_fitted(self, 's_') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) return self._row_coordinates_from_global(self._build_X_global(X)) def row_contributions(self, X): """Returns the row contributions towards each principal component.""" utils.validation.check_is_fitted(self, 's_') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) return super().row_contributions(self._build_X_global(X)) def partial_row_coordinates(self, X): """Returns the row coordinates for each group.""" utils.validation.check_is_fitted(self, 's_') # Check input if self.check_input: utils.check_array(X, dtype=[str, np.number]) # Prepare input X = self._prepare_input(X) # Define the projection matrix P P = len(X) ** 0.5 * self.U_ / self.s_ # Get the projections for each group coords = {} for name, cols in sorted(self.groups.items()): X_partial = X.loc[:, cols] if not self.all_nums_[name]: X_partial = self.cat_one_hots_[name].transform(X_partial) Z_partial = X_partial / self.partial_factor_analysis_[name].s_[0] coords[name] = len(self.groups) * (Z_partial @ Z_partial.T) @ P # Convert coords to a MultiIndex DataFrame coords = pd.DataFrame({ (name, i): group_coords.loc[:, i] for name, group_coords in coords.items() for i in range(group_coords.shape[1]) }) return coords def column_correlations(self, X): """Returns the column correlations.""" utils.validation.check_is_fitted(self, 's_') X_global = self._build_X_global(X) row_pc = self._row_coordinates_from_global(X_global) return pd.DataFrame({ component: { feature: row_pc[component].corr(X_global[feature].to_dense()) for feature in X_global.columns } for component in row_pc.columns })
MaxHalford/prince
prince/ca.py
CA.transform
python
def transform(self, X): utils.validation.check_is_fitted(self, 's_') if self.check_input: utils.check_array(X) return self.row_coordinates(X)
Computes the row principal coordinates of a dataset. Same as calling `row_coordinates`. In most cases you should be using the same dataset as you did when calling the `fit` method. You might however also want to included supplementary data.
train
https://github.com/MaxHalford/prince/blob/714c9cdfc4d9f8823eabf550a23ad01fe87c50d7/prince/ca.py#L69-L79
null
class CA(base.BaseEstimator, base.TransformerMixin): def __init__(self, n_components=2, n_iter=10, copy=True, check_input=True, random_state=None, engine='auto'): self.n_components = n_components self.n_iter = n_iter self.copy = copy self.check_input = check_input self.random_state = random_state self.engine = engine def fit(self, X, y=None): # Check input if self.check_input: utils.check_array(X) # Check all values are positive if (X < 0).any().any(): raise ValueError("All values in X should be positive") _, row_names, _, col_names = util.make_labels_and_names(X) if isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = np.copy(X) # Compute the correspondence matrix which contains the relative frequencies X = X / np.sum(X) # Compute row and column masses self.row_masses_ = pd.Series(X.sum(axis=1), index=row_names) self.col_masses_ = pd.Series(X.sum(axis=0), index=col_names) # Compute standardised residuals r = self.row_masses_.to_numpy() c = self.col_masses_.to_numpy() S = sparse.diags(r ** -0.5) @ (X - np.outer(r, c)) @ sparse.diags(c ** -0.5) # Compute SVD on the standardised residuals self.U_, self.s_, self.V_ = svd.compute_svd( X=S, n_components=self.n_components, n_iter=self.n_iter, random_state=self.random_state, engine=self.engine ) # Compute total inertia self.total_inertia_ = np.einsum('ij,ji->', S, S.T) return self @property def eigenvalues_(self): """The eigenvalues associated with each principal component.""" utils.validation.check_is_fitted(self, 's_') return np.square(self.s_).tolist() @property def explained_inertia_(self): """The percentage of explained inertia per principal component.""" utils.validation.check_is_fitted(self, 'total_inertia_') return [eig / self.total_inertia_ for eig in self.eigenvalues_] def row_coordinates(self, X): """The row principal coordinates.""" utils.validation.check_is_fitted(self, 'V_') _, row_names, _, _ = util.make_labels_and_names(X) if isinstance(X, pd.SparseDataFrame): X = X.to_coo().astype(float) elif isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = X.copy() # Normalise the rows so that they sum up to 1 if isinstance(X, np.ndarray): X = X / X.sum(axis=1)[:, None] else: X = X / X.sum(axis=1) return pd.DataFrame( data=X @ sparse.diags(self.col_masses_.to_numpy() ** -0.5) @ self.V_.T, index=row_names ) def column_coordinates(self, X): """The column principal coordinates.""" utils.validation.check_is_fitted(self, 'V_') _, _, _, col_names = util.make_labels_and_names(X) if isinstance(X, pd.SparseDataFrame): X = X.to_coo() elif isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = X.copy() # Transpose and make sure the rows sum up to 1 if isinstance(X, np.ndarray): X = X.T / X.T.sum(axis=1)[:, None] else: X = X.T / X.T.sum(axis=1) return pd.DataFrame( data=X @ sparse.diags(self.row_masses_.to_numpy() ** -0.5) @ self.U_, index=col_names ) def plot_coordinates(self, X, ax=None, figsize=(6, 6), x_component=0, y_component=1, show_row_labels=True, show_col_labels=True, **kwargs): """Plot the principal coordinates.""" utils.validation.check_is_fitted(self, 's_') if ax is None: fig, ax = plt.subplots(figsize=figsize) # Add style ax = plot.stylize_axis(ax) # Get labels and names row_label, row_names, col_label, col_names = util.make_labels_and_names(X) # Plot row principal coordinates row_coords = self.row_coordinates(X) ax.scatter( row_coords[x_component], row_coords[y_component], **kwargs, label=row_label ) # Plot column principal coordinates col_coords = self.column_coordinates(X) ax.scatter( col_coords[x_component], col_coords[y_component], **kwargs, label=col_label ) # Add row labels if show_row_labels: x = row_coords[x_component] y = row_coords[y_component] for i, label in enumerate(row_names): ax.annotate(label, (x[i], y[i])) # Add column labels if show_col_labels: x = col_coords[x_component] y = col_coords[y_component] for i, label in enumerate(col_names): ax.annotate(label, (x[i], y[i])) # Legend ax.legend() # Text ax.set_title('Principal coordinates') ei = self.explained_inertia_ ax.set_xlabel('Component {} ({:.2f}% inertia)'.format(x_component, 100 * ei[x_component])) ax.set_ylabel('Component {} ({:.2f}% inertia)'.format(y_component, 100 * ei[y_component])) return ax
MaxHalford/prince
prince/ca.py
CA.eigenvalues_
python
def eigenvalues_(self): utils.validation.check_is_fitted(self, 's_') return np.square(self.s_).tolist()
The eigenvalues associated with each principal component.
train
https://github.com/MaxHalford/prince/blob/714c9cdfc4d9f8823eabf550a23ad01fe87c50d7/prince/ca.py#L82-L85
null
class CA(base.BaseEstimator, base.TransformerMixin): def __init__(self, n_components=2, n_iter=10, copy=True, check_input=True, random_state=None, engine='auto'): self.n_components = n_components self.n_iter = n_iter self.copy = copy self.check_input = check_input self.random_state = random_state self.engine = engine def fit(self, X, y=None): # Check input if self.check_input: utils.check_array(X) # Check all values are positive if (X < 0).any().any(): raise ValueError("All values in X should be positive") _, row_names, _, col_names = util.make_labels_and_names(X) if isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = np.copy(X) # Compute the correspondence matrix which contains the relative frequencies X = X / np.sum(X) # Compute row and column masses self.row_masses_ = pd.Series(X.sum(axis=1), index=row_names) self.col_masses_ = pd.Series(X.sum(axis=0), index=col_names) # Compute standardised residuals r = self.row_masses_.to_numpy() c = self.col_masses_.to_numpy() S = sparse.diags(r ** -0.5) @ (X - np.outer(r, c)) @ sparse.diags(c ** -0.5) # Compute SVD on the standardised residuals self.U_, self.s_, self.V_ = svd.compute_svd( X=S, n_components=self.n_components, n_iter=self.n_iter, random_state=self.random_state, engine=self.engine ) # Compute total inertia self.total_inertia_ = np.einsum('ij,ji->', S, S.T) return self def transform(self, X): """Computes the row principal coordinates of a dataset. Same as calling `row_coordinates`. In most cases you should be using the same dataset as you did when calling the `fit` method. You might however also want to included supplementary data. """ utils.validation.check_is_fitted(self, 's_') if self.check_input: utils.check_array(X) return self.row_coordinates(X) @property @property def explained_inertia_(self): """The percentage of explained inertia per principal component.""" utils.validation.check_is_fitted(self, 'total_inertia_') return [eig / self.total_inertia_ for eig in self.eigenvalues_] def row_coordinates(self, X): """The row principal coordinates.""" utils.validation.check_is_fitted(self, 'V_') _, row_names, _, _ = util.make_labels_and_names(X) if isinstance(X, pd.SparseDataFrame): X = X.to_coo().astype(float) elif isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = X.copy() # Normalise the rows so that they sum up to 1 if isinstance(X, np.ndarray): X = X / X.sum(axis=1)[:, None] else: X = X / X.sum(axis=1) return pd.DataFrame( data=X @ sparse.diags(self.col_masses_.to_numpy() ** -0.5) @ self.V_.T, index=row_names ) def column_coordinates(self, X): """The column principal coordinates.""" utils.validation.check_is_fitted(self, 'V_') _, _, _, col_names = util.make_labels_and_names(X) if isinstance(X, pd.SparseDataFrame): X = X.to_coo() elif isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = X.copy() # Transpose and make sure the rows sum up to 1 if isinstance(X, np.ndarray): X = X.T / X.T.sum(axis=1)[:, None] else: X = X.T / X.T.sum(axis=1) return pd.DataFrame( data=X @ sparse.diags(self.row_masses_.to_numpy() ** -0.5) @ self.U_, index=col_names ) def plot_coordinates(self, X, ax=None, figsize=(6, 6), x_component=0, y_component=1, show_row_labels=True, show_col_labels=True, **kwargs): """Plot the principal coordinates.""" utils.validation.check_is_fitted(self, 's_') if ax is None: fig, ax = plt.subplots(figsize=figsize) # Add style ax = plot.stylize_axis(ax) # Get labels and names row_label, row_names, col_label, col_names = util.make_labels_and_names(X) # Plot row principal coordinates row_coords = self.row_coordinates(X) ax.scatter( row_coords[x_component], row_coords[y_component], **kwargs, label=row_label ) # Plot column principal coordinates col_coords = self.column_coordinates(X) ax.scatter( col_coords[x_component], col_coords[y_component], **kwargs, label=col_label ) # Add row labels if show_row_labels: x = row_coords[x_component] y = row_coords[y_component] for i, label in enumerate(row_names): ax.annotate(label, (x[i], y[i])) # Add column labels if show_col_labels: x = col_coords[x_component] y = col_coords[y_component] for i, label in enumerate(col_names): ax.annotate(label, (x[i], y[i])) # Legend ax.legend() # Text ax.set_title('Principal coordinates') ei = self.explained_inertia_ ax.set_xlabel('Component {} ({:.2f}% inertia)'.format(x_component, 100 * ei[x_component])) ax.set_ylabel('Component {} ({:.2f}% inertia)'.format(y_component, 100 * ei[y_component])) return ax
MaxHalford/prince
prince/ca.py
CA.explained_inertia_
python
def explained_inertia_(self): utils.validation.check_is_fitted(self, 'total_inertia_') return [eig / self.total_inertia_ for eig in self.eigenvalues_]
The percentage of explained inertia per principal component.
train
https://github.com/MaxHalford/prince/blob/714c9cdfc4d9f8823eabf550a23ad01fe87c50d7/prince/ca.py#L88-L91
null
class CA(base.BaseEstimator, base.TransformerMixin): def __init__(self, n_components=2, n_iter=10, copy=True, check_input=True, random_state=None, engine='auto'): self.n_components = n_components self.n_iter = n_iter self.copy = copy self.check_input = check_input self.random_state = random_state self.engine = engine def fit(self, X, y=None): # Check input if self.check_input: utils.check_array(X) # Check all values are positive if (X < 0).any().any(): raise ValueError("All values in X should be positive") _, row_names, _, col_names = util.make_labels_and_names(X) if isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = np.copy(X) # Compute the correspondence matrix which contains the relative frequencies X = X / np.sum(X) # Compute row and column masses self.row_masses_ = pd.Series(X.sum(axis=1), index=row_names) self.col_masses_ = pd.Series(X.sum(axis=0), index=col_names) # Compute standardised residuals r = self.row_masses_.to_numpy() c = self.col_masses_.to_numpy() S = sparse.diags(r ** -0.5) @ (X - np.outer(r, c)) @ sparse.diags(c ** -0.5) # Compute SVD on the standardised residuals self.U_, self.s_, self.V_ = svd.compute_svd( X=S, n_components=self.n_components, n_iter=self.n_iter, random_state=self.random_state, engine=self.engine ) # Compute total inertia self.total_inertia_ = np.einsum('ij,ji->', S, S.T) return self def transform(self, X): """Computes the row principal coordinates of a dataset. Same as calling `row_coordinates`. In most cases you should be using the same dataset as you did when calling the `fit` method. You might however also want to included supplementary data. """ utils.validation.check_is_fitted(self, 's_') if self.check_input: utils.check_array(X) return self.row_coordinates(X) @property def eigenvalues_(self): """The eigenvalues associated with each principal component.""" utils.validation.check_is_fitted(self, 's_') return np.square(self.s_).tolist() @property def row_coordinates(self, X): """The row principal coordinates.""" utils.validation.check_is_fitted(self, 'V_') _, row_names, _, _ = util.make_labels_and_names(X) if isinstance(X, pd.SparseDataFrame): X = X.to_coo().astype(float) elif isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = X.copy() # Normalise the rows so that they sum up to 1 if isinstance(X, np.ndarray): X = X / X.sum(axis=1)[:, None] else: X = X / X.sum(axis=1) return pd.DataFrame( data=X @ sparse.diags(self.col_masses_.to_numpy() ** -0.5) @ self.V_.T, index=row_names ) def column_coordinates(self, X): """The column principal coordinates.""" utils.validation.check_is_fitted(self, 'V_') _, _, _, col_names = util.make_labels_and_names(X) if isinstance(X, pd.SparseDataFrame): X = X.to_coo() elif isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = X.copy() # Transpose and make sure the rows sum up to 1 if isinstance(X, np.ndarray): X = X.T / X.T.sum(axis=1)[:, None] else: X = X.T / X.T.sum(axis=1) return pd.DataFrame( data=X @ sparse.diags(self.row_masses_.to_numpy() ** -0.5) @ self.U_, index=col_names ) def plot_coordinates(self, X, ax=None, figsize=(6, 6), x_component=0, y_component=1, show_row_labels=True, show_col_labels=True, **kwargs): """Plot the principal coordinates.""" utils.validation.check_is_fitted(self, 's_') if ax is None: fig, ax = plt.subplots(figsize=figsize) # Add style ax = plot.stylize_axis(ax) # Get labels and names row_label, row_names, col_label, col_names = util.make_labels_and_names(X) # Plot row principal coordinates row_coords = self.row_coordinates(X) ax.scatter( row_coords[x_component], row_coords[y_component], **kwargs, label=row_label ) # Plot column principal coordinates col_coords = self.column_coordinates(X) ax.scatter( col_coords[x_component], col_coords[y_component], **kwargs, label=col_label ) # Add row labels if show_row_labels: x = row_coords[x_component] y = row_coords[y_component] for i, label in enumerate(row_names): ax.annotate(label, (x[i], y[i])) # Add column labels if show_col_labels: x = col_coords[x_component] y = col_coords[y_component] for i, label in enumerate(col_names): ax.annotate(label, (x[i], y[i])) # Legend ax.legend() # Text ax.set_title('Principal coordinates') ei = self.explained_inertia_ ax.set_xlabel('Component {} ({:.2f}% inertia)'.format(x_component, 100 * ei[x_component])) ax.set_ylabel('Component {} ({:.2f}% inertia)'.format(y_component, 100 * ei[y_component])) return ax
MaxHalford/prince
prince/ca.py
CA.row_coordinates
python
def row_coordinates(self, X): utils.validation.check_is_fitted(self, 'V_') _, row_names, _, _ = util.make_labels_and_names(X) if isinstance(X, pd.SparseDataFrame): X = X.to_coo().astype(float) elif isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = X.copy() # Normalise the rows so that they sum up to 1 if isinstance(X, np.ndarray): X = X / X.sum(axis=1)[:, None] else: X = X / X.sum(axis=1) return pd.DataFrame( data=X @ sparse.diags(self.col_masses_.to_numpy() ** -0.5) @ self.V_.T, index=row_names )
The row principal coordinates.
train
https://github.com/MaxHalford/prince/blob/714c9cdfc4d9f8823eabf550a23ad01fe87c50d7/prince/ca.py#L93-L116
[ "def make_labels_and_names(X):\n\n if isinstance(X, pd.DataFrame):\n row_label = X.index.name if X.index.name else 'Rows'\n row_names = X.index.tolist()\n col_label = X.columns.name if X.columns.name else 'Columns'\n col_names = X.columns.tolist()\n else:\n row_label = 'Rows'\n row_names = list(range(X.shape[0]))\n col_label = 'Columns'\n col_names = list(range(X.shape[1]))\n\n return row_label, row_names, col_label, col_names\n" ]
class CA(base.BaseEstimator, base.TransformerMixin): def __init__(self, n_components=2, n_iter=10, copy=True, check_input=True, random_state=None, engine='auto'): self.n_components = n_components self.n_iter = n_iter self.copy = copy self.check_input = check_input self.random_state = random_state self.engine = engine def fit(self, X, y=None): # Check input if self.check_input: utils.check_array(X) # Check all values are positive if (X < 0).any().any(): raise ValueError("All values in X should be positive") _, row_names, _, col_names = util.make_labels_and_names(X) if isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = np.copy(X) # Compute the correspondence matrix which contains the relative frequencies X = X / np.sum(X) # Compute row and column masses self.row_masses_ = pd.Series(X.sum(axis=1), index=row_names) self.col_masses_ = pd.Series(X.sum(axis=0), index=col_names) # Compute standardised residuals r = self.row_masses_.to_numpy() c = self.col_masses_.to_numpy() S = sparse.diags(r ** -0.5) @ (X - np.outer(r, c)) @ sparse.diags(c ** -0.5) # Compute SVD on the standardised residuals self.U_, self.s_, self.V_ = svd.compute_svd( X=S, n_components=self.n_components, n_iter=self.n_iter, random_state=self.random_state, engine=self.engine ) # Compute total inertia self.total_inertia_ = np.einsum('ij,ji->', S, S.T) return self def transform(self, X): """Computes the row principal coordinates of a dataset. Same as calling `row_coordinates`. In most cases you should be using the same dataset as you did when calling the `fit` method. You might however also want to included supplementary data. """ utils.validation.check_is_fitted(self, 's_') if self.check_input: utils.check_array(X) return self.row_coordinates(X) @property def eigenvalues_(self): """The eigenvalues associated with each principal component.""" utils.validation.check_is_fitted(self, 's_') return np.square(self.s_).tolist() @property def explained_inertia_(self): """The percentage of explained inertia per principal component.""" utils.validation.check_is_fitted(self, 'total_inertia_') return [eig / self.total_inertia_ for eig in self.eigenvalues_] def column_coordinates(self, X): """The column principal coordinates.""" utils.validation.check_is_fitted(self, 'V_') _, _, _, col_names = util.make_labels_and_names(X) if isinstance(X, pd.SparseDataFrame): X = X.to_coo() elif isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = X.copy() # Transpose and make sure the rows sum up to 1 if isinstance(X, np.ndarray): X = X.T / X.T.sum(axis=1)[:, None] else: X = X.T / X.T.sum(axis=1) return pd.DataFrame( data=X @ sparse.diags(self.row_masses_.to_numpy() ** -0.5) @ self.U_, index=col_names ) def plot_coordinates(self, X, ax=None, figsize=(6, 6), x_component=0, y_component=1, show_row_labels=True, show_col_labels=True, **kwargs): """Plot the principal coordinates.""" utils.validation.check_is_fitted(self, 's_') if ax is None: fig, ax = plt.subplots(figsize=figsize) # Add style ax = plot.stylize_axis(ax) # Get labels and names row_label, row_names, col_label, col_names = util.make_labels_and_names(X) # Plot row principal coordinates row_coords = self.row_coordinates(X) ax.scatter( row_coords[x_component], row_coords[y_component], **kwargs, label=row_label ) # Plot column principal coordinates col_coords = self.column_coordinates(X) ax.scatter( col_coords[x_component], col_coords[y_component], **kwargs, label=col_label ) # Add row labels if show_row_labels: x = row_coords[x_component] y = row_coords[y_component] for i, label in enumerate(row_names): ax.annotate(label, (x[i], y[i])) # Add column labels if show_col_labels: x = col_coords[x_component] y = col_coords[y_component] for i, label in enumerate(col_names): ax.annotate(label, (x[i], y[i])) # Legend ax.legend() # Text ax.set_title('Principal coordinates') ei = self.explained_inertia_ ax.set_xlabel('Component {} ({:.2f}% inertia)'.format(x_component, 100 * ei[x_component])) ax.set_ylabel('Component {} ({:.2f}% inertia)'.format(y_component, 100 * ei[y_component])) return ax
MaxHalford/prince
prince/ca.py
CA.column_coordinates
python
def column_coordinates(self, X): utils.validation.check_is_fitted(self, 'V_') _, _, _, col_names = util.make_labels_and_names(X) if isinstance(X, pd.SparseDataFrame): X = X.to_coo() elif isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = X.copy() # Transpose and make sure the rows sum up to 1 if isinstance(X, np.ndarray): X = X.T / X.T.sum(axis=1)[:, None] else: X = X.T / X.T.sum(axis=1) return pd.DataFrame( data=X @ sparse.diags(self.row_masses_.to_numpy() ** -0.5) @ self.U_, index=col_names )
The column principal coordinates.
train
https://github.com/MaxHalford/prince/blob/714c9cdfc4d9f8823eabf550a23ad01fe87c50d7/prince/ca.py#L118-L141
[ "def make_labels_and_names(X):\n\n if isinstance(X, pd.DataFrame):\n row_label = X.index.name if X.index.name else 'Rows'\n row_names = X.index.tolist()\n col_label = X.columns.name if X.columns.name else 'Columns'\n col_names = X.columns.tolist()\n else:\n row_label = 'Rows'\n row_names = list(range(X.shape[0]))\n col_label = 'Columns'\n col_names = list(range(X.shape[1]))\n\n return row_label, row_names, col_label, col_names\n" ]
class CA(base.BaseEstimator, base.TransformerMixin): def __init__(self, n_components=2, n_iter=10, copy=True, check_input=True, random_state=None, engine='auto'): self.n_components = n_components self.n_iter = n_iter self.copy = copy self.check_input = check_input self.random_state = random_state self.engine = engine def fit(self, X, y=None): # Check input if self.check_input: utils.check_array(X) # Check all values are positive if (X < 0).any().any(): raise ValueError("All values in X should be positive") _, row_names, _, col_names = util.make_labels_and_names(X) if isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = np.copy(X) # Compute the correspondence matrix which contains the relative frequencies X = X / np.sum(X) # Compute row and column masses self.row_masses_ = pd.Series(X.sum(axis=1), index=row_names) self.col_masses_ = pd.Series(X.sum(axis=0), index=col_names) # Compute standardised residuals r = self.row_masses_.to_numpy() c = self.col_masses_.to_numpy() S = sparse.diags(r ** -0.5) @ (X - np.outer(r, c)) @ sparse.diags(c ** -0.5) # Compute SVD on the standardised residuals self.U_, self.s_, self.V_ = svd.compute_svd( X=S, n_components=self.n_components, n_iter=self.n_iter, random_state=self.random_state, engine=self.engine ) # Compute total inertia self.total_inertia_ = np.einsum('ij,ji->', S, S.T) return self def transform(self, X): """Computes the row principal coordinates of a dataset. Same as calling `row_coordinates`. In most cases you should be using the same dataset as you did when calling the `fit` method. You might however also want to included supplementary data. """ utils.validation.check_is_fitted(self, 's_') if self.check_input: utils.check_array(X) return self.row_coordinates(X) @property def eigenvalues_(self): """The eigenvalues associated with each principal component.""" utils.validation.check_is_fitted(self, 's_') return np.square(self.s_).tolist() @property def explained_inertia_(self): """The percentage of explained inertia per principal component.""" utils.validation.check_is_fitted(self, 'total_inertia_') return [eig / self.total_inertia_ for eig in self.eigenvalues_] def row_coordinates(self, X): """The row principal coordinates.""" utils.validation.check_is_fitted(self, 'V_') _, row_names, _, _ = util.make_labels_and_names(X) if isinstance(X, pd.SparseDataFrame): X = X.to_coo().astype(float) elif isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = X.copy() # Normalise the rows so that they sum up to 1 if isinstance(X, np.ndarray): X = X / X.sum(axis=1)[:, None] else: X = X / X.sum(axis=1) return pd.DataFrame( data=X @ sparse.diags(self.col_masses_.to_numpy() ** -0.5) @ self.V_.T, index=row_names ) def plot_coordinates(self, X, ax=None, figsize=(6, 6), x_component=0, y_component=1, show_row_labels=True, show_col_labels=True, **kwargs): """Plot the principal coordinates.""" utils.validation.check_is_fitted(self, 's_') if ax is None: fig, ax = plt.subplots(figsize=figsize) # Add style ax = plot.stylize_axis(ax) # Get labels and names row_label, row_names, col_label, col_names = util.make_labels_and_names(X) # Plot row principal coordinates row_coords = self.row_coordinates(X) ax.scatter( row_coords[x_component], row_coords[y_component], **kwargs, label=row_label ) # Plot column principal coordinates col_coords = self.column_coordinates(X) ax.scatter( col_coords[x_component], col_coords[y_component], **kwargs, label=col_label ) # Add row labels if show_row_labels: x = row_coords[x_component] y = row_coords[y_component] for i, label in enumerate(row_names): ax.annotate(label, (x[i], y[i])) # Add column labels if show_col_labels: x = col_coords[x_component] y = col_coords[y_component] for i, label in enumerate(col_names): ax.annotate(label, (x[i], y[i])) # Legend ax.legend() # Text ax.set_title('Principal coordinates') ei = self.explained_inertia_ ax.set_xlabel('Component {} ({:.2f}% inertia)'.format(x_component, 100 * ei[x_component])) ax.set_ylabel('Component {} ({:.2f}% inertia)'.format(y_component, 100 * ei[y_component])) return ax
MaxHalford/prince
prince/ca.py
CA.plot_coordinates
python
def plot_coordinates(self, X, ax=None, figsize=(6, 6), x_component=0, y_component=1, show_row_labels=True, show_col_labels=True, **kwargs): utils.validation.check_is_fitted(self, 's_') if ax is None: fig, ax = plt.subplots(figsize=figsize) # Add style ax = plot.stylize_axis(ax) # Get labels and names row_label, row_names, col_label, col_names = util.make_labels_and_names(X) # Plot row principal coordinates row_coords = self.row_coordinates(X) ax.scatter( row_coords[x_component], row_coords[y_component], **kwargs, label=row_label ) # Plot column principal coordinates col_coords = self.column_coordinates(X) ax.scatter( col_coords[x_component], col_coords[y_component], **kwargs, label=col_label ) # Add row labels if show_row_labels: x = row_coords[x_component] y = row_coords[y_component] for i, label in enumerate(row_names): ax.annotate(label, (x[i], y[i])) # Add column labels if show_col_labels: x = col_coords[x_component] y = col_coords[y_component] for i, label in enumerate(col_names): ax.annotate(label, (x[i], y[i])) # Legend ax.legend() # Text ax.set_title('Principal coordinates') ei = self.explained_inertia_ ax.set_xlabel('Component {} ({:.2f}% inertia)'.format(x_component, 100 * ei[x_component])) ax.set_ylabel('Component {} ({:.2f}% inertia)'.format(y_component, 100 * ei[y_component])) return ax
Plot the principal coordinates.
train
https://github.com/MaxHalford/prince/blob/714c9cdfc4d9f8823eabf550a23ad01fe87c50d7/prince/ca.py#L143-L199
[ "def make_labels_and_names(X):\n\n if isinstance(X, pd.DataFrame):\n row_label = X.index.name if X.index.name else 'Rows'\n row_names = X.index.tolist()\n col_label = X.columns.name if X.columns.name else 'Columns'\n col_names = X.columns.tolist()\n else:\n row_label = 'Rows'\n row_names = list(range(X.shape[0]))\n col_label = 'Columns'\n col_names = list(range(X.shape[1]))\n\n return row_label, row_names, col_label, col_names\n", "def stylize_axis(ax, grid=True):\n\n if grid:\n ax.grid()\n\n ax.xaxis.set_ticks_position('none')\n ax.yaxis.set_ticks_position('none')\n\n ax.axhline(y=0, linestyle='-', linewidth=1.2, color=GRAY['dark'], alpha=0.6)\n ax.axvline(x=0, linestyle='-', linewidth=1.2, color=GRAY['dark'], alpha=0.6)\n\n return ax\n" ]
class CA(base.BaseEstimator, base.TransformerMixin): def __init__(self, n_components=2, n_iter=10, copy=True, check_input=True, random_state=None, engine='auto'): self.n_components = n_components self.n_iter = n_iter self.copy = copy self.check_input = check_input self.random_state = random_state self.engine = engine def fit(self, X, y=None): # Check input if self.check_input: utils.check_array(X) # Check all values are positive if (X < 0).any().any(): raise ValueError("All values in X should be positive") _, row_names, _, col_names = util.make_labels_and_names(X) if isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = np.copy(X) # Compute the correspondence matrix which contains the relative frequencies X = X / np.sum(X) # Compute row and column masses self.row_masses_ = pd.Series(X.sum(axis=1), index=row_names) self.col_masses_ = pd.Series(X.sum(axis=0), index=col_names) # Compute standardised residuals r = self.row_masses_.to_numpy() c = self.col_masses_.to_numpy() S = sparse.diags(r ** -0.5) @ (X - np.outer(r, c)) @ sparse.diags(c ** -0.5) # Compute SVD on the standardised residuals self.U_, self.s_, self.V_ = svd.compute_svd( X=S, n_components=self.n_components, n_iter=self.n_iter, random_state=self.random_state, engine=self.engine ) # Compute total inertia self.total_inertia_ = np.einsum('ij,ji->', S, S.T) return self def transform(self, X): """Computes the row principal coordinates of a dataset. Same as calling `row_coordinates`. In most cases you should be using the same dataset as you did when calling the `fit` method. You might however also want to included supplementary data. """ utils.validation.check_is_fitted(self, 's_') if self.check_input: utils.check_array(X) return self.row_coordinates(X) @property def eigenvalues_(self): """The eigenvalues associated with each principal component.""" utils.validation.check_is_fitted(self, 's_') return np.square(self.s_).tolist() @property def explained_inertia_(self): """The percentage of explained inertia per principal component.""" utils.validation.check_is_fitted(self, 'total_inertia_') return [eig / self.total_inertia_ for eig in self.eigenvalues_] def row_coordinates(self, X): """The row principal coordinates.""" utils.validation.check_is_fitted(self, 'V_') _, row_names, _, _ = util.make_labels_and_names(X) if isinstance(X, pd.SparseDataFrame): X = X.to_coo().astype(float) elif isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = X.copy() # Normalise the rows so that they sum up to 1 if isinstance(X, np.ndarray): X = X / X.sum(axis=1)[:, None] else: X = X / X.sum(axis=1) return pd.DataFrame( data=X @ sparse.diags(self.col_masses_.to_numpy() ** -0.5) @ self.V_.T, index=row_names ) def column_coordinates(self, X): """The column principal coordinates.""" utils.validation.check_is_fitted(self, 'V_') _, _, _, col_names = util.make_labels_and_names(X) if isinstance(X, pd.SparseDataFrame): X = X.to_coo() elif isinstance(X, pd.DataFrame): X = X.to_numpy() if self.copy: X = X.copy() # Transpose and make sure the rows sum up to 1 if isinstance(X, np.ndarray): X = X.T / X.T.sum(axis=1)[:, None] else: X = X.T / X.T.sum(axis=1) return pd.DataFrame( data=X @ sparse.diags(self.row_masses_.to_numpy() ** -0.5) @ self.U_, index=col_names )