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
xtream1101/cutil
cutil/database.py
Database.update
python
def update(self, table, data_list, matched_field=None, return_cols='id'): data_list = copy.deepcopy(data_list) # Create deepcopy so the original list does not get modified if matched_field is None: # Assume the id field logger.info("Matched field not defined, assuming the `id` field") matched_field = 'id' # Make sure that `data_list` is a list if not isinstance(data_list, list): data_list = [data_list] if len(data_list) == 0: # No need to continue return [] # Make sure return_cols is a list if return_cols is None or len(return_cols) == 0 or return_cols[0] is None: return_cols = '' elif not isinstance(return_cols, list): return_cols = [return_cols] if len(return_cols) > 0: return_cols = 'RETURNING ' + ','.join(return_cols) # Data in the list must be dicts (just check the first one) if not isinstance(data_list[0], dict): logger.critical("Data must be a list of dicts") # Do not return here, let the exception handle the error that will be thrown when the query runs try: with self.getcursor() as cur: query_list = [] # TODO: change to return data from the database, not just what you passed in return_list = [] for row in data_list: if row.get(matched_field) is None: logger.debug("Cannot update row. Missing field {field} in data {data}" .format(field=matched_field, data=row)) logger.error("Cannot update row. Missing field {field} in data".format(field=matched_field)) continue # Pull matched_value from data to be updated and remove that key matched_value = row.get(matched_field) del row[matched_field] query = "UPDATE {table} SET {data} WHERE {matched_field}=%s {return_cols}"\ .format(table=table, data=','.join("%s=%%s" % u for u in row.keys()), matched_field=matched_field, return_cols=return_cols ) values = list(row.values()) values.append(matched_value) values = _check_values(values) query = cur.mogrify(query, values) query_list.append(query) return_list.append(matched_value) finial_query = b';'.join(query_list) cur.execute(finial_query) try: return cur.fetchall() except Exception: return None except Exception as e: logger.exception("Error updating data") logger.debug("Error updating data: {data}".format(data=data_list)) raise e.with_traceback(sys.exc_info()[2])
Create a bulk insert statement which is much faster (~2x in tests with 10k & 100k rows and 4 cols) for inserting data then executemany() TODO: Is there a limit of length the query can be? If so handle it.
train
https://github.com/xtream1101/cutil/blob/2e4d1f00e66154b44d4ccffb9b1db3f37e87f2e8/cutil/database.py#L213-L289
[ "def _check_values(in_values):\n \"\"\" Check if values need to be converted before they get mogrify'd\n \"\"\"\n out_values = []\n for value in in_values:\n # if isinstance(value, (dict, list)):\n # out_values.append(json.dumps(value))\n # else:\n out_values.append(value)\n\n return tuple(out_values)\n" ]
class Database: def __init__(self, db_config, table_raw=None, max_connections=10): from psycopg2.pool import ThreadedConnectionPool self.table_raw = table_raw try: # Set default port is port is not passed if 'db_port' not in db_config: db_config['db_port'] = 5432 self.pool = ThreadedConnectionPool(minconn=1, maxconn=max_connections, dsn="dbname={db_name} user={db_user} host={db_host} password={db_pass} port={db_port}" .format(**db_config)) except Exception: logger.exception("Error in db connection") sys.exit(1) logger.debug("Connected to database: {host}".format(host=db_config['db_host'])) @contextmanager def getcursor(self, **kwargs): conn = self.pool.getconn() try: yield conn.cursor(**kwargs) conn.commit() except Exception as e: conn.rollback() raise e.with_traceback(sys.exc_info()[2]) finally: self.pool.putconn(conn) def close(self): self.pool.closeall() def insert(self, table, data_list, return_cols='id'): """ Create a bulk insert statement which is much faster (~2x in tests with 10k & 100k rows and n cols) for inserting data then executemany() TODO: Is there a limit of length the query can be? If so handle it. """ data_list = copy.deepcopy(data_list) # Create deepcopy so the original list does not get modified # Make sure that `data_list` is a list if not isinstance(data_list, list): data_list = [data_list] # Make sure data_list has content if len(data_list) == 0: # No need to continue return [] # Data in the list must be dicts (just check the first one) if not isinstance(data_list[0], dict): logger.critical("Data must be a list of dicts") # Do not return here, let the exception handle the error that will be thrown when the query runs # Make sure return_cols is a list if return_cols is None or len(return_cols) == 0 or return_cols[0] is None: return_cols = '' elif not isinstance(return_cols, list): return_cols = [return_cols] if len(return_cols) > 0: return_cols = 'RETURNING ' + ','.join(return_cols) try: with self.getcursor() as cur: query = "INSERT INTO {table} ({fields}) VALUES {values} {return_cols}"\ .format(table=table, fields='"{0}"'.format('", "'.join(data_list[0].keys())), values=','.join(['%s'] * len(data_list)), return_cols=return_cols, ) values = [] for row in [tuple(v.values()) for v in data_list]: values.append(_check_values(row)) query = cur.mogrify(query, values) cur.execute(query) try: return cur.fetchall() except Exception: return None except Exception as e: logger.exception("Error inserting data") logger.debug("Error inserting data: {data}".format(data=data_list)) raise e.with_traceback(sys.exc_info()[2]) def upsert(self, table, data_list, on_conflict_fields, on_conflict_action='update', update_fields=None, return_cols='id'): """ Create a bulk upsert statement which is much faster (~6x in tests with 10k & 100k rows and n cols) for upserting data then executemany() TODO: Is there a limit of length the query can be? If so handle it. """ data_list = copy.deepcopy(data_list) # Create deepcopy so the original list does not get modified # Make sure that `data_list` is a list if not isinstance(data_list, list): data_list = [data_list] # Make sure data_list has content if len(data_list) == 0: # No need to continue return [] # Data in the list must be dicts (just check the first one) if not isinstance(data_list[0], dict): logger.critical("Data must be a list of dicts") # TODO: raise some error here rather then returning None return None # Make sure on_conflict_fields is a list if not isinstance(on_conflict_fields, list): on_conflict_fields = [on_conflict_fields] # Make sure on_conflict_fields has data if len(on_conflict_fields) == 0 or on_conflict_fields[0] is None: # No need to continue logger.critical("Must pass in `on_conflict_fields` argument") # TODO: raise some error here rather then returning None return None # Make sure return_cols is a list if return_cols is None or len(return_cols) == 0 or return_cols[0] is None: return_cols = '' elif not isinstance(return_cols, list): return_cols = [return_cols] if len(return_cols) > 0: return_cols = 'RETURNING ' + ','.join(return_cols) # Make sure update_fields is a list/valid if on_conflict_action == 'update': if not isinstance(update_fields, list): update_fields = [update_fields] # If noting is passed in, set `update_fields` to all (data_list-on_conflict_fields) if len(update_fields) == 0 or update_fields[0] is None: update_fields = list(set(data_list[0].keys()) - set(on_conflict_fields)) # If update_fields is empty here that could only mean that all fields are set as conflict_fields if len(update_fields) == 0: logger.critical("Not all the fields can be `on_conflict_fields` when doing an update") # TODO: raise some error here rather then returning None return None # If everything is good to go with the update fields fields_update_tmp = [] for key in data_list[0].keys(): fields_update_tmp.append('"{0}"="excluded"."{0}"'.format(key)) conflict_action_sql = 'UPDATE SET {update_fields}'\ .format(update_fields=', '.join(fields_update_tmp)) else: # Do nothing on conflict conflict_action_sql = 'NOTHING' try: with self.getcursor() as cur: query = """INSERT INTO {table} ({insert_fields}) VALUES {values} ON CONFLICT ({on_conflict_fields}) DO {conflict_action_sql} {return_cols} """.format(table=table, insert_fields='"{0}"'.format('","'.join(data_list[0].keys())), values=','.join(['%s'] * len(data_list)), on_conflict_fields=','.join(on_conflict_fields), conflict_action_sql=conflict_action_sql, return_cols=return_cols, ) # Get all the values for each row and create a lists of lists values = [] for row in [list(v.values()) for v in data_list]: values.append(_check_values(row)) query = cur.mogrify(query, values) cur.execute(query) try: return cur.fetchall() except Exception: return None except Exception as e: logger.exception("Error upserting data") logger.debug("Error upserting data: {data}".format(data=data_list)) raise e.with_traceback(sys.exc_info()[2])
xtream1101/cutil
cutil/config.py
Config.load_configs
python
def load_configs(self, conf_file): with open(conf_file) as stream: lines = itertools.chain(("[global]",), stream) self._config.read_file(lines) return self._config['global']
Assumes that the config file does not have any sections, so throw it all in global
train
https://github.com/xtream1101/cutil/blob/2e4d1f00e66154b44d4ccffb9b1db3f37e87f2e8/cutil/config.py#L12-L19
null
class Config: def __init__(self, conf_file=None): self._config = configparser.ConfigParser() self._config.optionxform = str # Keep case of keys self.config_values = self.remove_quotes(self.load_configs(conf_file)) def remove_quotes(self, configs): """ Because some values are wraped in single quotes """ for key in configs: value = configs[key] if value[0] == "'" and value[-1] == "'": configs[key] = value[1:-1] return configs
xtream1101/cutil
cutil/config.py
Config.remove_quotes
python
def remove_quotes(self, configs): for key in configs: value = configs[key] if value[0] == "'" and value[-1] == "'": configs[key] = value[1:-1] return configs
Because some values are wraped in single quotes
train
https://github.com/xtream1101/cutil/blob/2e4d1f00e66154b44d4ccffb9b1db3f37e87f2e8/cutil/config.py#L21-L29
null
class Config: def __init__(self, conf_file=None): self._config = configparser.ConfigParser() self._config.optionxform = str # Keep case of keys self.config_values = self.remove_quotes(self.load_configs(conf_file)) def load_configs(self, conf_file): """ Assumes that the config file does not have any sections, so throw it all in global """ with open(conf_file) as stream: lines = itertools.chain(("[global]",), stream) self._config.read_file(lines) return self._config['global']
xtream1101/cutil
cutil/__init__.py
multikey_sort
python
def multikey_sort(items, columns): comparers = [ ((itemgetter(col[1:].strip()), -1) if col.startswith('-') else (itemgetter(col.strip()), 1)) for col in columns ] def cmp(a, b): return (a > b) - (a < b) def comparer(left, right): comparer_iter = ( cmp(fn(left), fn(right)) * mult for fn, mult in comparers ) return next((result for result in comparer_iter if result), 0) return sorted(items, key=cmp_to_key(comparer))
Source: https://stackoverflow.com/questions/1143671/python-sorting-list-of-dictionaries-by-multiple-keys
train
https://github.com/xtream1101/cutil/blob/2e4d1f00e66154b44d4ccffb9b1db3f37e87f2e8/cutil/__init__.py#L108-L125
null
from cutil.database import Database # noqa: F401 from cutil.config import Config # noqa: F401 from cutil.custom_terminal import CustomTerminal # noqa: F401 from cutil.repeating_timer import RepeatingTimer # noqa: F401 import os import re import sys import uuid import math import time import pytz import json import queue import random import socket import urllib import hashlib import logging import datetime import threading import collections from hashids import Hashids from functools import wraps from operator import itemgetter from functools import cmp_to_key logger = logging.getLogger(__name__) #### # Time related functions #### def get_epoch(): """ :return: time as epoch """ return int(time.time()) def get_datetime(): """ :return: datetime object """ return datetime.datetime.now() def datetime_to_str(timestamp): return timestamp.isoformat() + "+0000" def datetime_to_utc(timestamp): return timestamp.astimezone(pytz.timezone('UTC')) def str_to_date(timestamp, formats=["%Y-%m-%dT%H:%M:%S.%f%z", "%Y-%m-%dT%H:%M:%S%z"]): rdata = None if timestamp is None: return None for time_format in formats: try: rdata = datetime.datetime.strptime(timestamp, time_format) except ValueError: rdata = None else: # If we did not raise an exception break return rdata ### # Threading ### def threads(num_threads, data, callback, *args, **kwargs): q = queue.Queue() item_list = [] def _thread_run(): while True: item = q.get() try: item_list.append(callback(item, *args, **kwargs)) except Exception: logger.exception("Error in _thread_run callback {} with item:\n{}".format(callback.__name__, item)) q.task_done() for i in range(num_threads): t = threading.Thread(target=_thread_run) t.daemon = True t.start() # Fill the Queue with the data to process for item in data: q.put(item) # Start processing the data q.join() return item_list #### # Other functions #### def get_internal_ip(): return socket.gethostbyname(socket.gethostname()) def generate_key(value=None, salt=None, size=8): if value is None: value = random.randint(0, 999999) if salt is None: salt = random.randint(0, 999999) if not isinstance(size, int): # Enforce that size must be an int size = 8 hashids = Hashids(salt=str(salt), min_length=size) return hashids.encode(value) def create_uid(): # Named uid and not uuid because this function does not have to use uuid's uid = uuid.uuid4().hex logger.debug("Created new uid: {uid}".format(uid=uid)) return uid def make_url_safe(string): # Convert special chars to % chars return urllib.parse.quote_plus(string) def sanitize(string): """ Catch and replace invalid path chars [replace, with] """ replace_chars = [ ['\\', '-'], [':', '-'], ['/', '-'], ['?', ''], ['<', ''], ['>', ''], ['`', '`'], ['|', '-'], ['*', '`'], ['"', '\''], ['.', ''], ['&', 'and'] ] for ch in replace_chars: string = string.replace(ch[0], ch[1]) return string def rreplace(s, old, new, occurrence): """ Taken from: http://stackoverflow.com/a/2556252 """ li = s.rsplit(old, occurrence) return new.join(li) def flatten(dict_obj, prev_key='', sep='_'): items = {} for key, value in dict_obj.items(): new_key = prev_key + sep + key if prev_key != '' else key if isinstance(value, dict): items.update(flatten(value, new_key)) else: items[new_key] = value return items def update_dict(d, u): """ Source: https://stackoverflow.com/questions/3232943/update-value-of-a-nested-dictionary-of-varying-depth """ for k, v in u.items(): if isinstance(v, collections.Mapping): r = update_dict(d.get(k, {}), v) d[k] = r else: d[k] = u[k] return d def get_script_name(ext=False): name = os.path.basename(sys.argv[0]) if ext is False: name = name.split('.')[0] return name def chunks_of(max_chunk_size, list_to_chunk): """ Yields the list with a max size of max_chunk_size """ for i in range(0, len(list_to_chunk), max_chunk_size): yield list_to_chunk[i:i + max_chunk_size] def split_into(max_num_chunks, list_to_chunk): """ Yields the list with a max total size of max_num_chunks """ max_chunk_size = math.ceil(len(list_to_chunk) / max_num_chunks) return chunks_of(max_chunk_size, list_to_chunk) #### # File/filesystem related function #### def get_file_ext(file): file_name, file_extension = os.path.splitext(file) return file_extension def norm_path(path): """ :return: Proper path for os with vars expanded out """ # path = os.path.normcase(path) path = os.path.expanduser(path) path = os.path.expandvars(path) path = os.path.normpath(path) return path def create_hashed_path(base_path, name, depth=2): """ Create a directory structure using the hashed filename :return: string of the path to save to not including filename/ext """ if depth > 16: logger.warning("depth cannot be greater then 16, setting to 16") depth = 16 name_hash = hashlib.md5(str(name).encode('utf-8')).hexdigest() if base_path.endswith(os.path.sep): save_path = base_path else: save_path = base_path + os.path.sep for i in range(1, depth + 1): end = i * 2 start = end - 2 save_path += name_hash[start:end] + os.path.sep return {'path': save_path, 'hash': name_hash, } def create_path(path, is_dir=False): """ Check if path exists, if not create it :param path: path or file to create directory for :param is_dir: pass True if we are passing in a directory, default = False :return: os safe path from `path` """ path = norm_path(path) path_check = path if not is_dir: path_check = os.path.dirname(path) does_path_exists = os.path.exists(path_check) if does_path_exists: return path try: os.makedirs(path_check) except OSError: pass return path def dump_json(file_, data, **kwargs): json_args = {'sort_keys': True, 'indent': 4} json_args.update(**kwargs) create_path(file_) if not file_.endswith('.json'): file_ += '.json' with open(file_, 'w') as outfile: json.dump(data, outfile, **json_args) # Lets only do this once price_pattern = re.compile('(?P<low>[\d,.\s]+)(?:\D*(?P<high>[\d,.\s]+))?') def parse_price(price): found_price = {'low': None, 'high': None } price_raw = re.search(price_pattern, price) if price_raw: matched = price_raw.groupdict() found_price['low'] = matched.get('low') found_price['high'] = matched.get('high') for key, value in found_price.items(): if value is not None: value = value.strip() new_value = value.replace(',', '').replace('.', '').replace(' ', '') try: # Check if price has cents if value[-3] in [',', '.']: # Add . for cents back new_value = new_value[:-2] + '.' + new_value[-2:] except IndexError: # Price is 99 or less with no cents pass if new_value != '': found_price[key] = float(new_value) else: found_price[key] = None return found_price def get_image_dimension(url): import requests from PIL import Image # pip install pillow from io import BytesIO size = {'width': None, 'height': None, } try: if url.startswith('//'): url = "http:" + url data = requests.get(url, timeout=15).content im = Image.open(BytesIO(data)) size['width'], size['height'] = im.size except Exception: logger.debug("Error getting image size {url}".format(url=url), exc_info=True) logger.warning("Error getting image size {url}".format(url=url)) return size def crop_image(image_file, output_file=None, height=None, width=None, x=None, y=None): from PIL import Image # pip install pillow if output_file is None or height is None or width is None or x is None or y is None: logger.error("Must pass in all params: output_file, height, width, x, and y as named args") return None if width <= 0 or height <= 0: logger.warning("Width and height must be 1 or greater") raise ValueError image = Image.open(image_file) image.crop((x, y, width + x, height + y)).save(output_file) return output_file #### # Decorators #### def rate_limited(num_calls=1, every=1.0): """ Source: https://github.com/tomasbasham/ratelimit/tree/0ca5a616fa6d184fa180b9ad0b6fd0cf54c46936 Need to make a few changes that included having num_calls be a float Prevent a method from being called if it was previously called before a time widows has elapsed. Keyword Arguments: num_calls (float): Maximum method invocations within a period. Must be greater than 0. every (float): A dampening factor (in seconds). Can be any number greater than 0. Return: function: Decorated function that will forward method invocations if the time window has elapsed. """ frequency = abs(every) / float(num_calls) def decorator(func): """ Extend the behaviour of the following function, forwarding method invocations if the time window hes elapsed. Arguments: func (function): The function to decorate Returns: function: Decorated function """ # To get around issues with function local scope # and reassigning variables, we wrap the time # within a list. When updating the value we're # not reassigning `last_called`, which would not # work, but instead reassigning the value at a # particular index. last_called = [0.0] # Add thread safety lock = threading.RLock() def wrapper(*args, **kargs): """Decorator wrapper function""" with lock: elapsed = time.time() - last_called[0] left_to_wait = frequency - elapsed if left_to_wait > 0: time.sleep(left_to_wait) last_called[0] = time.time() return func(*args, **kargs) return wrapper return decorator def rate_limited_old(max_per_second): """ Source: https://gist.github.com/gregburek/1441055 """ lock = threading.Lock() min_interval = 1.0 / max_per_second def decorate(func): last_time_called = time.perf_counter() @wraps(func) def rate_limited_function(*args, **kwargs): lock.acquire() nonlocal last_time_called try: elapsed = time.perf_counter() - last_time_called left_to_wait = min_interval - elapsed if left_to_wait > 0: time.sleep(left_to_wait) return func(*args, **kwargs) finally: last_time_called = time.perf_counter() lock.release() return rate_limited_function return decorate def timeit(stat_tracker_func, name): """ Pass in a function and the name of the stat Will time the function that this is a decorator to and send the `name` as well as the value (in seconds) to `stat_tracker_func` `stat_tracker_func` can be used to either print out the data or save it """ def _timeit(func): def wrapper(*args, **kw): start_time = time.time() result = func(*args, **kw) stop_time = time.time() stat_tracker_func(name, stop_time - start_time) return result return wrapper return _timeit #### # Regex #### # Keep re.compile's outside of fn as to only create it once proxy_parts_pattern = re.compile('^(?:(?P<schema>\w+):\/\/)(?:(?P<user>.*):(?P<password>.*)@)?(?P<host>[^:]*)(?::(?P<port>\d+))?$') def get_proxy_parts(proxy): """ Take a proxy url and break it up to its parts """ proxy_parts = {'schema': None, 'user': None, 'password': None, 'host': None, 'port': None, } # Find parts results = re.match(proxy_parts_pattern, proxy) if results: matched = results.groupdict() for key in proxy_parts: proxy_parts[key] = matched.get(key) else: logger.error("Invalid proxy format `{proxy}`".format(proxy=proxy)) if proxy_parts['port'] is None: proxy_parts['port'] = '80' return proxy_parts def remove_html_tag(input_str='', tag=None): """ Returns a string with the html tag and all its contents from a string """ result = input_str if tag is not None: pattern = re.compile('<{tag}[\s\S]+?/{tag}>'.format(tag=tag)) result = re.sub(pattern, '', str(input_str)) return result
xtream1101/cutil
cutil/__init__.py
sanitize
python
def sanitize(string): replace_chars = [ ['\\', '-'], [':', '-'], ['/', '-'], ['?', ''], ['<', ''], ['>', ''], ['`', '`'], ['|', '-'], ['*', '`'], ['"', '\''], ['.', ''], ['&', 'and'] ] for ch in replace_chars: string = string.replace(ch[0], ch[1]) return string
Catch and replace invalid path chars [replace, with]
train
https://github.com/xtream1101/cutil/blob/2e4d1f00e66154b44d4ccffb9b1db3f37e87f2e8/cutil/__init__.py#L159-L172
null
from cutil.database import Database # noqa: F401 from cutil.config import Config # noqa: F401 from cutil.custom_terminal import CustomTerminal # noqa: F401 from cutil.repeating_timer import RepeatingTimer # noqa: F401 import os import re import sys import uuid import math import time import pytz import json import queue import random import socket import urllib import hashlib import logging import datetime import threading import collections from hashids import Hashids from functools import wraps from operator import itemgetter from functools import cmp_to_key logger = logging.getLogger(__name__) #### # Time related functions #### def get_epoch(): """ :return: time as epoch """ return int(time.time()) def get_datetime(): """ :return: datetime object """ return datetime.datetime.now() def datetime_to_str(timestamp): return timestamp.isoformat() + "+0000" def datetime_to_utc(timestamp): return timestamp.astimezone(pytz.timezone('UTC')) def str_to_date(timestamp, formats=["%Y-%m-%dT%H:%M:%S.%f%z", "%Y-%m-%dT%H:%M:%S%z"]): rdata = None if timestamp is None: return None for time_format in formats: try: rdata = datetime.datetime.strptime(timestamp, time_format) except ValueError: rdata = None else: # If we did not raise an exception break return rdata ### # Threading ### def threads(num_threads, data, callback, *args, **kwargs): q = queue.Queue() item_list = [] def _thread_run(): while True: item = q.get() try: item_list.append(callback(item, *args, **kwargs)) except Exception: logger.exception("Error in _thread_run callback {} with item:\n{}".format(callback.__name__, item)) q.task_done() for i in range(num_threads): t = threading.Thread(target=_thread_run) t.daemon = True t.start() # Fill the Queue with the data to process for item in data: q.put(item) # Start processing the data q.join() return item_list #### # Other functions #### def multikey_sort(items, columns): """Source: https://stackoverflow.com/questions/1143671/python-sorting-list-of-dictionaries-by-multiple-keys """ comparers = [ ((itemgetter(col[1:].strip()), -1) if col.startswith('-') else (itemgetter(col.strip()), 1)) for col in columns ] def cmp(a, b): return (a > b) - (a < b) def comparer(left, right): comparer_iter = ( cmp(fn(left), fn(right)) * mult for fn, mult in comparers ) return next((result for result in comparer_iter if result), 0) return sorted(items, key=cmp_to_key(comparer)) def get_internal_ip(): return socket.gethostbyname(socket.gethostname()) def generate_key(value=None, salt=None, size=8): if value is None: value = random.randint(0, 999999) if salt is None: salt = random.randint(0, 999999) if not isinstance(size, int): # Enforce that size must be an int size = 8 hashids = Hashids(salt=str(salt), min_length=size) return hashids.encode(value) def create_uid(): # Named uid and not uuid because this function does not have to use uuid's uid = uuid.uuid4().hex logger.debug("Created new uid: {uid}".format(uid=uid)) return uid def make_url_safe(string): # Convert special chars to % chars return urllib.parse.quote_plus(string) def rreplace(s, old, new, occurrence): """ Taken from: http://stackoverflow.com/a/2556252 """ li = s.rsplit(old, occurrence) return new.join(li) def flatten(dict_obj, prev_key='', sep='_'): items = {} for key, value in dict_obj.items(): new_key = prev_key + sep + key if prev_key != '' else key if isinstance(value, dict): items.update(flatten(value, new_key)) else: items[new_key] = value return items def update_dict(d, u): """ Source: https://stackoverflow.com/questions/3232943/update-value-of-a-nested-dictionary-of-varying-depth """ for k, v in u.items(): if isinstance(v, collections.Mapping): r = update_dict(d.get(k, {}), v) d[k] = r else: d[k] = u[k] return d def get_script_name(ext=False): name = os.path.basename(sys.argv[0]) if ext is False: name = name.split('.')[0] return name def chunks_of(max_chunk_size, list_to_chunk): """ Yields the list with a max size of max_chunk_size """ for i in range(0, len(list_to_chunk), max_chunk_size): yield list_to_chunk[i:i + max_chunk_size] def split_into(max_num_chunks, list_to_chunk): """ Yields the list with a max total size of max_num_chunks """ max_chunk_size = math.ceil(len(list_to_chunk) / max_num_chunks) return chunks_of(max_chunk_size, list_to_chunk) #### # File/filesystem related function #### def get_file_ext(file): file_name, file_extension = os.path.splitext(file) return file_extension def norm_path(path): """ :return: Proper path for os with vars expanded out """ # path = os.path.normcase(path) path = os.path.expanduser(path) path = os.path.expandvars(path) path = os.path.normpath(path) return path def create_hashed_path(base_path, name, depth=2): """ Create a directory structure using the hashed filename :return: string of the path to save to not including filename/ext """ if depth > 16: logger.warning("depth cannot be greater then 16, setting to 16") depth = 16 name_hash = hashlib.md5(str(name).encode('utf-8')).hexdigest() if base_path.endswith(os.path.sep): save_path = base_path else: save_path = base_path + os.path.sep for i in range(1, depth + 1): end = i * 2 start = end - 2 save_path += name_hash[start:end] + os.path.sep return {'path': save_path, 'hash': name_hash, } def create_path(path, is_dir=False): """ Check if path exists, if not create it :param path: path or file to create directory for :param is_dir: pass True if we are passing in a directory, default = False :return: os safe path from `path` """ path = norm_path(path) path_check = path if not is_dir: path_check = os.path.dirname(path) does_path_exists = os.path.exists(path_check) if does_path_exists: return path try: os.makedirs(path_check) except OSError: pass return path def dump_json(file_, data, **kwargs): json_args = {'sort_keys': True, 'indent': 4} json_args.update(**kwargs) create_path(file_) if not file_.endswith('.json'): file_ += '.json' with open(file_, 'w') as outfile: json.dump(data, outfile, **json_args) # Lets only do this once price_pattern = re.compile('(?P<low>[\d,.\s]+)(?:\D*(?P<high>[\d,.\s]+))?') def parse_price(price): found_price = {'low': None, 'high': None } price_raw = re.search(price_pattern, price) if price_raw: matched = price_raw.groupdict() found_price['low'] = matched.get('low') found_price['high'] = matched.get('high') for key, value in found_price.items(): if value is not None: value = value.strip() new_value = value.replace(',', '').replace('.', '').replace(' ', '') try: # Check if price has cents if value[-3] in [',', '.']: # Add . for cents back new_value = new_value[:-2] + '.' + new_value[-2:] except IndexError: # Price is 99 or less with no cents pass if new_value != '': found_price[key] = float(new_value) else: found_price[key] = None return found_price def get_image_dimension(url): import requests from PIL import Image # pip install pillow from io import BytesIO size = {'width': None, 'height': None, } try: if url.startswith('//'): url = "http:" + url data = requests.get(url, timeout=15).content im = Image.open(BytesIO(data)) size['width'], size['height'] = im.size except Exception: logger.debug("Error getting image size {url}".format(url=url), exc_info=True) logger.warning("Error getting image size {url}".format(url=url)) return size def crop_image(image_file, output_file=None, height=None, width=None, x=None, y=None): from PIL import Image # pip install pillow if output_file is None or height is None or width is None or x is None or y is None: logger.error("Must pass in all params: output_file, height, width, x, and y as named args") return None if width <= 0 or height <= 0: logger.warning("Width and height must be 1 or greater") raise ValueError image = Image.open(image_file) image.crop((x, y, width + x, height + y)).save(output_file) return output_file #### # Decorators #### def rate_limited(num_calls=1, every=1.0): """ Source: https://github.com/tomasbasham/ratelimit/tree/0ca5a616fa6d184fa180b9ad0b6fd0cf54c46936 Need to make a few changes that included having num_calls be a float Prevent a method from being called if it was previously called before a time widows has elapsed. Keyword Arguments: num_calls (float): Maximum method invocations within a period. Must be greater than 0. every (float): A dampening factor (in seconds). Can be any number greater than 0. Return: function: Decorated function that will forward method invocations if the time window has elapsed. """ frequency = abs(every) / float(num_calls) def decorator(func): """ Extend the behaviour of the following function, forwarding method invocations if the time window hes elapsed. Arguments: func (function): The function to decorate Returns: function: Decorated function """ # To get around issues with function local scope # and reassigning variables, we wrap the time # within a list. When updating the value we're # not reassigning `last_called`, which would not # work, but instead reassigning the value at a # particular index. last_called = [0.0] # Add thread safety lock = threading.RLock() def wrapper(*args, **kargs): """Decorator wrapper function""" with lock: elapsed = time.time() - last_called[0] left_to_wait = frequency - elapsed if left_to_wait > 0: time.sleep(left_to_wait) last_called[0] = time.time() return func(*args, **kargs) return wrapper return decorator def rate_limited_old(max_per_second): """ Source: https://gist.github.com/gregburek/1441055 """ lock = threading.Lock() min_interval = 1.0 / max_per_second def decorate(func): last_time_called = time.perf_counter() @wraps(func) def rate_limited_function(*args, **kwargs): lock.acquire() nonlocal last_time_called try: elapsed = time.perf_counter() - last_time_called left_to_wait = min_interval - elapsed if left_to_wait > 0: time.sleep(left_to_wait) return func(*args, **kwargs) finally: last_time_called = time.perf_counter() lock.release() return rate_limited_function return decorate def timeit(stat_tracker_func, name): """ Pass in a function and the name of the stat Will time the function that this is a decorator to and send the `name` as well as the value (in seconds) to `stat_tracker_func` `stat_tracker_func` can be used to either print out the data or save it """ def _timeit(func): def wrapper(*args, **kw): start_time = time.time() result = func(*args, **kw) stop_time = time.time() stat_tracker_func(name, stop_time - start_time) return result return wrapper return _timeit #### # Regex #### # Keep re.compile's outside of fn as to only create it once proxy_parts_pattern = re.compile('^(?:(?P<schema>\w+):\/\/)(?:(?P<user>.*):(?P<password>.*)@)?(?P<host>[^:]*)(?::(?P<port>\d+))?$') def get_proxy_parts(proxy): """ Take a proxy url and break it up to its parts """ proxy_parts = {'schema': None, 'user': None, 'password': None, 'host': None, 'port': None, } # Find parts results = re.match(proxy_parts_pattern, proxy) if results: matched = results.groupdict() for key in proxy_parts: proxy_parts[key] = matched.get(key) else: logger.error("Invalid proxy format `{proxy}`".format(proxy=proxy)) if proxy_parts['port'] is None: proxy_parts['port'] = '80' return proxy_parts def remove_html_tag(input_str='', tag=None): """ Returns a string with the html tag and all its contents from a string """ result = input_str if tag is not None: pattern = re.compile('<{tag}[\s\S]+?/{tag}>'.format(tag=tag)) result = re.sub(pattern, '', str(input_str)) return result
xtream1101/cutil
cutil/__init__.py
chunks_of
python
def chunks_of(max_chunk_size, list_to_chunk): for i in range(0, len(list_to_chunk), max_chunk_size): yield list_to_chunk[i:i + max_chunk_size]
Yields the list with a max size of max_chunk_size
train
https://github.com/xtream1101/cutil/blob/2e4d1f00e66154b44d4ccffb9b1db3f37e87f2e8/cutil/__init__.py#L216-L221
null
from cutil.database import Database # noqa: F401 from cutil.config import Config # noqa: F401 from cutil.custom_terminal import CustomTerminal # noqa: F401 from cutil.repeating_timer import RepeatingTimer # noqa: F401 import os import re import sys import uuid import math import time import pytz import json import queue import random import socket import urllib import hashlib import logging import datetime import threading import collections from hashids import Hashids from functools import wraps from operator import itemgetter from functools import cmp_to_key logger = logging.getLogger(__name__) #### # Time related functions #### def get_epoch(): """ :return: time as epoch """ return int(time.time()) def get_datetime(): """ :return: datetime object """ return datetime.datetime.now() def datetime_to_str(timestamp): return timestamp.isoformat() + "+0000" def datetime_to_utc(timestamp): return timestamp.astimezone(pytz.timezone('UTC')) def str_to_date(timestamp, formats=["%Y-%m-%dT%H:%M:%S.%f%z", "%Y-%m-%dT%H:%M:%S%z"]): rdata = None if timestamp is None: return None for time_format in formats: try: rdata = datetime.datetime.strptime(timestamp, time_format) except ValueError: rdata = None else: # If we did not raise an exception break return rdata ### # Threading ### def threads(num_threads, data, callback, *args, **kwargs): q = queue.Queue() item_list = [] def _thread_run(): while True: item = q.get() try: item_list.append(callback(item, *args, **kwargs)) except Exception: logger.exception("Error in _thread_run callback {} with item:\n{}".format(callback.__name__, item)) q.task_done() for i in range(num_threads): t = threading.Thread(target=_thread_run) t.daemon = True t.start() # Fill the Queue with the data to process for item in data: q.put(item) # Start processing the data q.join() return item_list #### # Other functions #### def multikey_sort(items, columns): """Source: https://stackoverflow.com/questions/1143671/python-sorting-list-of-dictionaries-by-multiple-keys """ comparers = [ ((itemgetter(col[1:].strip()), -1) if col.startswith('-') else (itemgetter(col.strip()), 1)) for col in columns ] def cmp(a, b): return (a > b) - (a < b) def comparer(left, right): comparer_iter = ( cmp(fn(left), fn(right)) * mult for fn, mult in comparers ) return next((result for result in comparer_iter if result), 0) return sorted(items, key=cmp_to_key(comparer)) def get_internal_ip(): return socket.gethostbyname(socket.gethostname()) def generate_key(value=None, salt=None, size=8): if value is None: value = random.randint(0, 999999) if salt is None: salt = random.randint(0, 999999) if not isinstance(size, int): # Enforce that size must be an int size = 8 hashids = Hashids(salt=str(salt), min_length=size) return hashids.encode(value) def create_uid(): # Named uid and not uuid because this function does not have to use uuid's uid = uuid.uuid4().hex logger.debug("Created new uid: {uid}".format(uid=uid)) return uid def make_url_safe(string): # Convert special chars to % chars return urllib.parse.quote_plus(string) def sanitize(string): """ Catch and replace invalid path chars [replace, with] """ replace_chars = [ ['\\', '-'], [':', '-'], ['/', '-'], ['?', ''], ['<', ''], ['>', ''], ['`', '`'], ['|', '-'], ['*', '`'], ['"', '\''], ['.', ''], ['&', 'and'] ] for ch in replace_chars: string = string.replace(ch[0], ch[1]) return string def rreplace(s, old, new, occurrence): """ Taken from: http://stackoverflow.com/a/2556252 """ li = s.rsplit(old, occurrence) return new.join(li) def flatten(dict_obj, prev_key='', sep='_'): items = {} for key, value in dict_obj.items(): new_key = prev_key + sep + key if prev_key != '' else key if isinstance(value, dict): items.update(flatten(value, new_key)) else: items[new_key] = value return items def update_dict(d, u): """ Source: https://stackoverflow.com/questions/3232943/update-value-of-a-nested-dictionary-of-varying-depth """ for k, v in u.items(): if isinstance(v, collections.Mapping): r = update_dict(d.get(k, {}), v) d[k] = r else: d[k] = u[k] return d def get_script_name(ext=False): name = os.path.basename(sys.argv[0]) if ext is False: name = name.split('.')[0] return name def split_into(max_num_chunks, list_to_chunk): """ Yields the list with a max total size of max_num_chunks """ max_chunk_size = math.ceil(len(list_to_chunk) / max_num_chunks) return chunks_of(max_chunk_size, list_to_chunk) #### # File/filesystem related function #### def get_file_ext(file): file_name, file_extension = os.path.splitext(file) return file_extension def norm_path(path): """ :return: Proper path for os with vars expanded out """ # path = os.path.normcase(path) path = os.path.expanduser(path) path = os.path.expandvars(path) path = os.path.normpath(path) return path def create_hashed_path(base_path, name, depth=2): """ Create a directory structure using the hashed filename :return: string of the path to save to not including filename/ext """ if depth > 16: logger.warning("depth cannot be greater then 16, setting to 16") depth = 16 name_hash = hashlib.md5(str(name).encode('utf-8')).hexdigest() if base_path.endswith(os.path.sep): save_path = base_path else: save_path = base_path + os.path.sep for i in range(1, depth + 1): end = i * 2 start = end - 2 save_path += name_hash[start:end] + os.path.sep return {'path': save_path, 'hash': name_hash, } def create_path(path, is_dir=False): """ Check if path exists, if not create it :param path: path or file to create directory for :param is_dir: pass True if we are passing in a directory, default = False :return: os safe path from `path` """ path = norm_path(path) path_check = path if not is_dir: path_check = os.path.dirname(path) does_path_exists = os.path.exists(path_check) if does_path_exists: return path try: os.makedirs(path_check) except OSError: pass return path def dump_json(file_, data, **kwargs): json_args = {'sort_keys': True, 'indent': 4} json_args.update(**kwargs) create_path(file_) if not file_.endswith('.json'): file_ += '.json' with open(file_, 'w') as outfile: json.dump(data, outfile, **json_args) # Lets only do this once price_pattern = re.compile('(?P<low>[\d,.\s]+)(?:\D*(?P<high>[\d,.\s]+))?') def parse_price(price): found_price = {'low': None, 'high': None } price_raw = re.search(price_pattern, price) if price_raw: matched = price_raw.groupdict() found_price['low'] = matched.get('low') found_price['high'] = matched.get('high') for key, value in found_price.items(): if value is not None: value = value.strip() new_value = value.replace(',', '').replace('.', '').replace(' ', '') try: # Check if price has cents if value[-3] in [',', '.']: # Add . for cents back new_value = new_value[:-2] + '.' + new_value[-2:] except IndexError: # Price is 99 or less with no cents pass if new_value != '': found_price[key] = float(new_value) else: found_price[key] = None return found_price def get_image_dimension(url): import requests from PIL import Image # pip install pillow from io import BytesIO size = {'width': None, 'height': None, } try: if url.startswith('//'): url = "http:" + url data = requests.get(url, timeout=15).content im = Image.open(BytesIO(data)) size['width'], size['height'] = im.size except Exception: logger.debug("Error getting image size {url}".format(url=url), exc_info=True) logger.warning("Error getting image size {url}".format(url=url)) return size def crop_image(image_file, output_file=None, height=None, width=None, x=None, y=None): from PIL import Image # pip install pillow if output_file is None or height is None or width is None or x is None or y is None: logger.error("Must pass in all params: output_file, height, width, x, and y as named args") return None if width <= 0 or height <= 0: logger.warning("Width and height must be 1 or greater") raise ValueError image = Image.open(image_file) image.crop((x, y, width + x, height + y)).save(output_file) return output_file #### # Decorators #### def rate_limited(num_calls=1, every=1.0): """ Source: https://github.com/tomasbasham/ratelimit/tree/0ca5a616fa6d184fa180b9ad0b6fd0cf54c46936 Need to make a few changes that included having num_calls be a float Prevent a method from being called if it was previously called before a time widows has elapsed. Keyword Arguments: num_calls (float): Maximum method invocations within a period. Must be greater than 0. every (float): A dampening factor (in seconds). Can be any number greater than 0. Return: function: Decorated function that will forward method invocations if the time window has elapsed. """ frequency = abs(every) / float(num_calls) def decorator(func): """ Extend the behaviour of the following function, forwarding method invocations if the time window hes elapsed. Arguments: func (function): The function to decorate Returns: function: Decorated function """ # To get around issues with function local scope # and reassigning variables, we wrap the time # within a list. When updating the value we're # not reassigning `last_called`, which would not # work, but instead reassigning the value at a # particular index. last_called = [0.0] # Add thread safety lock = threading.RLock() def wrapper(*args, **kargs): """Decorator wrapper function""" with lock: elapsed = time.time() - last_called[0] left_to_wait = frequency - elapsed if left_to_wait > 0: time.sleep(left_to_wait) last_called[0] = time.time() return func(*args, **kargs) return wrapper return decorator def rate_limited_old(max_per_second): """ Source: https://gist.github.com/gregburek/1441055 """ lock = threading.Lock() min_interval = 1.0 / max_per_second def decorate(func): last_time_called = time.perf_counter() @wraps(func) def rate_limited_function(*args, **kwargs): lock.acquire() nonlocal last_time_called try: elapsed = time.perf_counter() - last_time_called left_to_wait = min_interval - elapsed if left_to_wait > 0: time.sleep(left_to_wait) return func(*args, **kwargs) finally: last_time_called = time.perf_counter() lock.release() return rate_limited_function return decorate def timeit(stat_tracker_func, name): """ Pass in a function and the name of the stat Will time the function that this is a decorator to and send the `name` as well as the value (in seconds) to `stat_tracker_func` `stat_tracker_func` can be used to either print out the data or save it """ def _timeit(func): def wrapper(*args, **kw): start_time = time.time() result = func(*args, **kw) stop_time = time.time() stat_tracker_func(name, stop_time - start_time) return result return wrapper return _timeit #### # Regex #### # Keep re.compile's outside of fn as to only create it once proxy_parts_pattern = re.compile('^(?:(?P<schema>\w+):\/\/)(?:(?P<user>.*):(?P<password>.*)@)?(?P<host>[^:]*)(?::(?P<port>\d+))?$') def get_proxy_parts(proxy): """ Take a proxy url and break it up to its parts """ proxy_parts = {'schema': None, 'user': None, 'password': None, 'host': None, 'port': None, } # Find parts results = re.match(proxy_parts_pattern, proxy) if results: matched = results.groupdict() for key in proxy_parts: proxy_parts[key] = matched.get(key) else: logger.error("Invalid proxy format `{proxy}`".format(proxy=proxy)) if proxy_parts['port'] is None: proxy_parts['port'] = '80' return proxy_parts def remove_html_tag(input_str='', tag=None): """ Returns a string with the html tag and all its contents from a string """ result = input_str if tag is not None: pattern = re.compile('<{tag}[\s\S]+?/{tag}>'.format(tag=tag)) result = re.sub(pattern, '', str(input_str)) return result
xtream1101/cutil
cutil/__init__.py
split_into
python
def split_into(max_num_chunks, list_to_chunk): max_chunk_size = math.ceil(len(list_to_chunk) / max_num_chunks) return chunks_of(max_chunk_size, list_to_chunk)
Yields the list with a max total size of max_num_chunks
train
https://github.com/xtream1101/cutil/blob/2e4d1f00e66154b44d4ccffb9b1db3f37e87f2e8/cutil/__init__.py#L224-L229
[ "def chunks_of(max_chunk_size, list_to_chunk):\n \"\"\"\n Yields the list with a max size of max_chunk_size\n \"\"\"\n for i in range(0, len(list_to_chunk), max_chunk_size):\n yield list_to_chunk[i:i + max_chunk_size]\n" ]
from cutil.database import Database # noqa: F401 from cutil.config import Config # noqa: F401 from cutil.custom_terminal import CustomTerminal # noqa: F401 from cutil.repeating_timer import RepeatingTimer # noqa: F401 import os import re import sys import uuid import math import time import pytz import json import queue import random import socket import urllib import hashlib import logging import datetime import threading import collections from hashids import Hashids from functools import wraps from operator import itemgetter from functools import cmp_to_key logger = logging.getLogger(__name__) #### # Time related functions #### def get_epoch(): """ :return: time as epoch """ return int(time.time()) def get_datetime(): """ :return: datetime object """ return datetime.datetime.now() def datetime_to_str(timestamp): return timestamp.isoformat() + "+0000" def datetime_to_utc(timestamp): return timestamp.astimezone(pytz.timezone('UTC')) def str_to_date(timestamp, formats=["%Y-%m-%dT%H:%M:%S.%f%z", "%Y-%m-%dT%H:%M:%S%z"]): rdata = None if timestamp is None: return None for time_format in formats: try: rdata = datetime.datetime.strptime(timestamp, time_format) except ValueError: rdata = None else: # If we did not raise an exception break return rdata ### # Threading ### def threads(num_threads, data, callback, *args, **kwargs): q = queue.Queue() item_list = [] def _thread_run(): while True: item = q.get() try: item_list.append(callback(item, *args, **kwargs)) except Exception: logger.exception("Error in _thread_run callback {} with item:\n{}".format(callback.__name__, item)) q.task_done() for i in range(num_threads): t = threading.Thread(target=_thread_run) t.daemon = True t.start() # Fill the Queue with the data to process for item in data: q.put(item) # Start processing the data q.join() return item_list #### # Other functions #### def multikey_sort(items, columns): """Source: https://stackoverflow.com/questions/1143671/python-sorting-list-of-dictionaries-by-multiple-keys """ comparers = [ ((itemgetter(col[1:].strip()), -1) if col.startswith('-') else (itemgetter(col.strip()), 1)) for col in columns ] def cmp(a, b): return (a > b) - (a < b) def comparer(left, right): comparer_iter = ( cmp(fn(left), fn(right)) * mult for fn, mult in comparers ) return next((result for result in comparer_iter if result), 0) return sorted(items, key=cmp_to_key(comparer)) def get_internal_ip(): return socket.gethostbyname(socket.gethostname()) def generate_key(value=None, salt=None, size=8): if value is None: value = random.randint(0, 999999) if salt is None: salt = random.randint(0, 999999) if not isinstance(size, int): # Enforce that size must be an int size = 8 hashids = Hashids(salt=str(salt), min_length=size) return hashids.encode(value) def create_uid(): # Named uid and not uuid because this function does not have to use uuid's uid = uuid.uuid4().hex logger.debug("Created new uid: {uid}".format(uid=uid)) return uid def make_url_safe(string): # Convert special chars to % chars return urllib.parse.quote_plus(string) def sanitize(string): """ Catch and replace invalid path chars [replace, with] """ replace_chars = [ ['\\', '-'], [':', '-'], ['/', '-'], ['?', ''], ['<', ''], ['>', ''], ['`', '`'], ['|', '-'], ['*', '`'], ['"', '\''], ['.', ''], ['&', 'and'] ] for ch in replace_chars: string = string.replace(ch[0], ch[1]) return string def rreplace(s, old, new, occurrence): """ Taken from: http://stackoverflow.com/a/2556252 """ li = s.rsplit(old, occurrence) return new.join(li) def flatten(dict_obj, prev_key='', sep='_'): items = {} for key, value in dict_obj.items(): new_key = prev_key + sep + key if prev_key != '' else key if isinstance(value, dict): items.update(flatten(value, new_key)) else: items[new_key] = value return items def update_dict(d, u): """ Source: https://stackoverflow.com/questions/3232943/update-value-of-a-nested-dictionary-of-varying-depth """ for k, v in u.items(): if isinstance(v, collections.Mapping): r = update_dict(d.get(k, {}), v) d[k] = r else: d[k] = u[k] return d def get_script_name(ext=False): name = os.path.basename(sys.argv[0]) if ext is False: name = name.split('.')[0] return name def chunks_of(max_chunk_size, list_to_chunk): """ Yields the list with a max size of max_chunk_size """ for i in range(0, len(list_to_chunk), max_chunk_size): yield list_to_chunk[i:i + max_chunk_size] #### # File/filesystem related function #### def get_file_ext(file): file_name, file_extension = os.path.splitext(file) return file_extension def norm_path(path): """ :return: Proper path for os with vars expanded out """ # path = os.path.normcase(path) path = os.path.expanduser(path) path = os.path.expandvars(path) path = os.path.normpath(path) return path def create_hashed_path(base_path, name, depth=2): """ Create a directory structure using the hashed filename :return: string of the path to save to not including filename/ext """ if depth > 16: logger.warning("depth cannot be greater then 16, setting to 16") depth = 16 name_hash = hashlib.md5(str(name).encode('utf-8')).hexdigest() if base_path.endswith(os.path.sep): save_path = base_path else: save_path = base_path + os.path.sep for i in range(1, depth + 1): end = i * 2 start = end - 2 save_path += name_hash[start:end] + os.path.sep return {'path': save_path, 'hash': name_hash, } def create_path(path, is_dir=False): """ Check if path exists, if not create it :param path: path or file to create directory for :param is_dir: pass True if we are passing in a directory, default = False :return: os safe path from `path` """ path = norm_path(path) path_check = path if not is_dir: path_check = os.path.dirname(path) does_path_exists = os.path.exists(path_check) if does_path_exists: return path try: os.makedirs(path_check) except OSError: pass return path def dump_json(file_, data, **kwargs): json_args = {'sort_keys': True, 'indent': 4} json_args.update(**kwargs) create_path(file_) if not file_.endswith('.json'): file_ += '.json' with open(file_, 'w') as outfile: json.dump(data, outfile, **json_args) # Lets only do this once price_pattern = re.compile('(?P<low>[\d,.\s]+)(?:\D*(?P<high>[\d,.\s]+))?') def parse_price(price): found_price = {'low': None, 'high': None } price_raw = re.search(price_pattern, price) if price_raw: matched = price_raw.groupdict() found_price['low'] = matched.get('low') found_price['high'] = matched.get('high') for key, value in found_price.items(): if value is not None: value = value.strip() new_value = value.replace(',', '').replace('.', '').replace(' ', '') try: # Check if price has cents if value[-3] in [',', '.']: # Add . for cents back new_value = new_value[:-2] + '.' + new_value[-2:] except IndexError: # Price is 99 or less with no cents pass if new_value != '': found_price[key] = float(new_value) else: found_price[key] = None return found_price def get_image_dimension(url): import requests from PIL import Image # pip install pillow from io import BytesIO size = {'width': None, 'height': None, } try: if url.startswith('//'): url = "http:" + url data = requests.get(url, timeout=15).content im = Image.open(BytesIO(data)) size['width'], size['height'] = im.size except Exception: logger.debug("Error getting image size {url}".format(url=url), exc_info=True) logger.warning("Error getting image size {url}".format(url=url)) return size def crop_image(image_file, output_file=None, height=None, width=None, x=None, y=None): from PIL import Image # pip install pillow if output_file is None or height is None or width is None or x is None or y is None: logger.error("Must pass in all params: output_file, height, width, x, and y as named args") return None if width <= 0 or height <= 0: logger.warning("Width and height must be 1 or greater") raise ValueError image = Image.open(image_file) image.crop((x, y, width + x, height + y)).save(output_file) return output_file #### # Decorators #### def rate_limited(num_calls=1, every=1.0): """ Source: https://github.com/tomasbasham/ratelimit/tree/0ca5a616fa6d184fa180b9ad0b6fd0cf54c46936 Need to make a few changes that included having num_calls be a float Prevent a method from being called if it was previously called before a time widows has elapsed. Keyword Arguments: num_calls (float): Maximum method invocations within a period. Must be greater than 0. every (float): A dampening factor (in seconds). Can be any number greater than 0. Return: function: Decorated function that will forward method invocations if the time window has elapsed. """ frequency = abs(every) / float(num_calls) def decorator(func): """ Extend the behaviour of the following function, forwarding method invocations if the time window hes elapsed. Arguments: func (function): The function to decorate Returns: function: Decorated function """ # To get around issues with function local scope # and reassigning variables, we wrap the time # within a list. When updating the value we're # not reassigning `last_called`, which would not # work, but instead reassigning the value at a # particular index. last_called = [0.0] # Add thread safety lock = threading.RLock() def wrapper(*args, **kargs): """Decorator wrapper function""" with lock: elapsed = time.time() - last_called[0] left_to_wait = frequency - elapsed if left_to_wait > 0: time.sleep(left_to_wait) last_called[0] = time.time() return func(*args, **kargs) return wrapper return decorator def rate_limited_old(max_per_second): """ Source: https://gist.github.com/gregburek/1441055 """ lock = threading.Lock() min_interval = 1.0 / max_per_second def decorate(func): last_time_called = time.perf_counter() @wraps(func) def rate_limited_function(*args, **kwargs): lock.acquire() nonlocal last_time_called try: elapsed = time.perf_counter() - last_time_called left_to_wait = min_interval - elapsed if left_to_wait > 0: time.sleep(left_to_wait) return func(*args, **kwargs) finally: last_time_called = time.perf_counter() lock.release() return rate_limited_function return decorate def timeit(stat_tracker_func, name): """ Pass in a function and the name of the stat Will time the function that this is a decorator to and send the `name` as well as the value (in seconds) to `stat_tracker_func` `stat_tracker_func` can be used to either print out the data or save it """ def _timeit(func): def wrapper(*args, **kw): start_time = time.time() result = func(*args, **kw) stop_time = time.time() stat_tracker_func(name, stop_time - start_time) return result return wrapper return _timeit #### # Regex #### # Keep re.compile's outside of fn as to only create it once proxy_parts_pattern = re.compile('^(?:(?P<schema>\w+):\/\/)(?:(?P<user>.*):(?P<password>.*)@)?(?P<host>[^:]*)(?::(?P<port>\d+))?$') def get_proxy_parts(proxy): """ Take a proxy url and break it up to its parts """ proxy_parts = {'schema': None, 'user': None, 'password': None, 'host': None, 'port': None, } # Find parts results = re.match(proxy_parts_pattern, proxy) if results: matched = results.groupdict() for key in proxy_parts: proxy_parts[key] = matched.get(key) else: logger.error("Invalid proxy format `{proxy}`".format(proxy=proxy)) if proxy_parts['port'] is None: proxy_parts['port'] = '80' return proxy_parts def remove_html_tag(input_str='', tag=None): """ Returns a string with the html tag and all its contents from a string """ result = input_str if tag is not None: pattern = re.compile('<{tag}[\s\S]+?/{tag}>'.format(tag=tag)) result = re.sub(pattern, '', str(input_str)) return result
xtream1101/cutil
cutil/__init__.py
norm_path
python
def norm_path(path): # path = os.path.normcase(path) path = os.path.expanduser(path) path = os.path.expandvars(path) path = os.path.normpath(path) return path
:return: Proper path for os with vars expanded out
train
https://github.com/xtream1101/cutil/blob/2e4d1f00e66154b44d4ccffb9b1db3f37e87f2e8/cutil/__init__.py#L240-L248
null
from cutil.database import Database # noqa: F401 from cutil.config import Config # noqa: F401 from cutil.custom_terminal import CustomTerminal # noqa: F401 from cutil.repeating_timer import RepeatingTimer # noqa: F401 import os import re import sys import uuid import math import time import pytz import json import queue import random import socket import urllib import hashlib import logging import datetime import threading import collections from hashids import Hashids from functools import wraps from operator import itemgetter from functools import cmp_to_key logger = logging.getLogger(__name__) #### # Time related functions #### def get_epoch(): """ :return: time as epoch """ return int(time.time()) def get_datetime(): """ :return: datetime object """ return datetime.datetime.now() def datetime_to_str(timestamp): return timestamp.isoformat() + "+0000" def datetime_to_utc(timestamp): return timestamp.astimezone(pytz.timezone('UTC')) def str_to_date(timestamp, formats=["%Y-%m-%dT%H:%M:%S.%f%z", "%Y-%m-%dT%H:%M:%S%z"]): rdata = None if timestamp is None: return None for time_format in formats: try: rdata = datetime.datetime.strptime(timestamp, time_format) except ValueError: rdata = None else: # If we did not raise an exception break return rdata ### # Threading ### def threads(num_threads, data, callback, *args, **kwargs): q = queue.Queue() item_list = [] def _thread_run(): while True: item = q.get() try: item_list.append(callback(item, *args, **kwargs)) except Exception: logger.exception("Error in _thread_run callback {} with item:\n{}".format(callback.__name__, item)) q.task_done() for i in range(num_threads): t = threading.Thread(target=_thread_run) t.daemon = True t.start() # Fill the Queue with the data to process for item in data: q.put(item) # Start processing the data q.join() return item_list #### # Other functions #### def multikey_sort(items, columns): """Source: https://stackoverflow.com/questions/1143671/python-sorting-list-of-dictionaries-by-multiple-keys """ comparers = [ ((itemgetter(col[1:].strip()), -1) if col.startswith('-') else (itemgetter(col.strip()), 1)) for col in columns ] def cmp(a, b): return (a > b) - (a < b) def comparer(left, right): comparer_iter = ( cmp(fn(left), fn(right)) * mult for fn, mult in comparers ) return next((result for result in comparer_iter if result), 0) return sorted(items, key=cmp_to_key(comparer)) def get_internal_ip(): return socket.gethostbyname(socket.gethostname()) def generate_key(value=None, salt=None, size=8): if value is None: value = random.randint(0, 999999) if salt is None: salt = random.randint(0, 999999) if not isinstance(size, int): # Enforce that size must be an int size = 8 hashids = Hashids(salt=str(salt), min_length=size) return hashids.encode(value) def create_uid(): # Named uid and not uuid because this function does not have to use uuid's uid = uuid.uuid4().hex logger.debug("Created new uid: {uid}".format(uid=uid)) return uid def make_url_safe(string): # Convert special chars to % chars return urllib.parse.quote_plus(string) def sanitize(string): """ Catch and replace invalid path chars [replace, with] """ replace_chars = [ ['\\', '-'], [':', '-'], ['/', '-'], ['?', ''], ['<', ''], ['>', ''], ['`', '`'], ['|', '-'], ['*', '`'], ['"', '\''], ['.', ''], ['&', 'and'] ] for ch in replace_chars: string = string.replace(ch[0], ch[1]) return string def rreplace(s, old, new, occurrence): """ Taken from: http://stackoverflow.com/a/2556252 """ li = s.rsplit(old, occurrence) return new.join(li) def flatten(dict_obj, prev_key='', sep='_'): items = {} for key, value in dict_obj.items(): new_key = prev_key + sep + key if prev_key != '' else key if isinstance(value, dict): items.update(flatten(value, new_key)) else: items[new_key] = value return items def update_dict(d, u): """ Source: https://stackoverflow.com/questions/3232943/update-value-of-a-nested-dictionary-of-varying-depth """ for k, v in u.items(): if isinstance(v, collections.Mapping): r = update_dict(d.get(k, {}), v) d[k] = r else: d[k] = u[k] return d def get_script_name(ext=False): name = os.path.basename(sys.argv[0]) if ext is False: name = name.split('.')[0] return name def chunks_of(max_chunk_size, list_to_chunk): """ Yields the list with a max size of max_chunk_size """ for i in range(0, len(list_to_chunk), max_chunk_size): yield list_to_chunk[i:i + max_chunk_size] def split_into(max_num_chunks, list_to_chunk): """ Yields the list with a max total size of max_num_chunks """ max_chunk_size = math.ceil(len(list_to_chunk) / max_num_chunks) return chunks_of(max_chunk_size, list_to_chunk) #### # File/filesystem related function #### def get_file_ext(file): file_name, file_extension = os.path.splitext(file) return file_extension def create_hashed_path(base_path, name, depth=2): """ Create a directory structure using the hashed filename :return: string of the path to save to not including filename/ext """ if depth > 16: logger.warning("depth cannot be greater then 16, setting to 16") depth = 16 name_hash = hashlib.md5(str(name).encode('utf-8')).hexdigest() if base_path.endswith(os.path.sep): save_path = base_path else: save_path = base_path + os.path.sep for i in range(1, depth + 1): end = i * 2 start = end - 2 save_path += name_hash[start:end] + os.path.sep return {'path': save_path, 'hash': name_hash, } def create_path(path, is_dir=False): """ Check if path exists, if not create it :param path: path or file to create directory for :param is_dir: pass True if we are passing in a directory, default = False :return: os safe path from `path` """ path = norm_path(path) path_check = path if not is_dir: path_check = os.path.dirname(path) does_path_exists = os.path.exists(path_check) if does_path_exists: return path try: os.makedirs(path_check) except OSError: pass return path def dump_json(file_, data, **kwargs): json_args = {'sort_keys': True, 'indent': 4} json_args.update(**kwargs) create_path(file_) if not file_.endswith('.json'): file_ += '.json' with open(file_, 'w') as outfile: json.dump(data, outfile, **json_args) # Lets only do this once price_pattern = re.compile('(?P<low>[\d,.\s]+)(?:\D*(?P<high>[\d,.\s]+))?') def parse_price(price): found_price = {'low': None, 'high': None } price_raw = re.search(price_pattern, price) if price_raw: matched = price_raw.groupdict() found_price['low'] = matched.get('low') found_price['high'] = matched.get('high') for key, value in found_price.items(): if value is not None: value = value.strip() new_value = value.replace(',', '').replace('.', '').replace(' ', '') try: # Check if price has cents if value[-3] in [',', '.']: # Add . for cents back new_value = new_value[:-2] + '.' + new_value[-2:] except IndexError: # Price is 99 or less with no cents pass if new_value != '': found_price[key] = float(new_value) else: found_price[key] = None return found_price def get_image_dimension(url): import requests from PIL import Image # pip install pillow from io import BytesIO size = {'width': None, 'height': None, } try: if url.startswith('//'): url = "http:" + url data = requests.get(url, timeout=15).content im = Image.open(BytesIO(data)) size['width'], size['height'] = im.size except Exception: logger.debug("Error getting image size {url}".format(url=url), exc_info=True) logger.warning("Error getting image size {url}".format(url=url)) return size def crop_image(image_file, output_file=None, height=None, width=None, x=None, y=None): from PIL import Image # pip install pillow if output_file is None or height is None or width is None or x is None or y is None: logger.error("Must pass in all params: output_file, height, width, x, and y as named args") return None if width <= 0 or height <= 0: logger.warning("Width and height must be 1 or greater") raise ValueError image = Image.open(image_file) image.crop((x, y, width + x, height + y)).save(output_file) return output_file #### # Decorators #### def rate_limited(num_calls=1, every=1.0): """ Source: https://github.com/tomasbasham/ratelimit/tree/0ca5a616fa6d184fa180b9ad0b6fd0cf54c46936 Need to make a few changes that included having num_calls be a float Prevent a method from being called if it was previously called before a time widows has elapsed. Keyword Arguments: num_calls (float): Maximum method invocations within a period. Must be greater than 0. every (float): A dampening factor (in seconds). Can be any number greater than 0. Return: function: Decorated function that will forward method invocations if the time window has elapsed. """ frequency = abs(every) / float(num_calls) def decorator(func): """ Extend the behaviour of the following function, forwarding method invocations if the time window hes elapsed. Arguments: func (function): The function to decorate Returns: function: Decorated function """ # To get around issues with function local scope # and reassigning variables, we wrap the time # within a list. When updating the value we're # not reassigning `last_called`, which would not # work, but instead reassigning the value at a # particular index. last_called = [0.0] # Add thread safety lock = threading.RLock() def wrapper(*args, **kargs): """Decorator wrapper function""" with lock: elapsed = time.time() - last_called[0] left_to_wait = frequency - elapsed if left_to_wait > 0: time.sleep(left_to_wait) last_called[0] = time.time() return func(*args, **kargs) return wrapper return decorator def rate_limited_old(max_per_second): """ Source: https://gist.github.com/gregburek/1441055 """ lock = threading.Lock() min_interval = 1.0 / max_per_second def decorate(func): last_time_called = time.perf_counter() @wraps(func) def rate_limited_function(*args, **kwargs): lock.acquire() nonlocal last_time_called try: elapsed = time.perf_counter() - last_time_called left_to_wait = min_interval - elapsed if left_to_wait > 0: time.sleep(left_to_wait) return func(*args, **kwargs) finally: last_time_called = time.perf_counter() lock.release() return rate_limited_function return decorate def timeit(stat_tracker_func, name): """ Pass in a function and the name of the stat Will time the function that this is a decorator to and send the `name` as well as the value (in seconds) to `stat_tracker_func` `stat_tracker_func` can be used to either print out the data or save it """ def _timeit(func): def wrapper(*args, **kw): start_time = time.time() result = func(*args, **kw) stop_time = time.time() stat_tracker_func(name, stop_time - start_time) return result return wrapper return _timeit #### # Regex #### # Keep re.compile's outside of fn as to only create it once proxy_parts_pattern = re.compile('^(?:(?P<schema>\w+):\/\/)(?:(?P<user>.*):(?P<password>.*)@)?(?P<host>[^:]*)(?::(?P<port>\d+))?$') def get_proxy_parts(proxy): """ Take a proxy url and break it up to its parts """ proxy_parts = {'schema': None, 'user': None, 'password': None, 'host': None, 'port': None, } # Find parts results = re.match(proxy_parts_pattern, proxy) if results: matched = results.groupdict() for key in proxy_parts: proxy_parts[key] = matched.get(key) else: logger.error("Invalid proxy format `{proxy}`".format(proxy=proxy)) if proxy_parts['port'] is None: proxy_parts['port'] = '80' return proxy_parts def remove_html_tag(input_str='', tag=None): """ Returns a string with the html tag and all its contents from a string """ result = input_str if tag is not None: pattern = re.compile('<{tag}[\s\S]+?/{tag}>'.format(tag=tag)) result = re.sub(pattern, '', str(input_str)) return result
xtream1101/cutil
cutil/__init__.py
create_hashed_path
python
def create_hashed_path(base_path, name, depth=2): if depth > 16: logger.warning("depth cannot be greater then 16, setting to 16") depth = 16 name_hash = hashlib.md5(str(name).encode('utf-8')).hexdigest() if base_path.endswith(os.path.sep): save_path = base_path else: save_path = base_path + os.path.sep for i in range(1, depth + 1): end = i * 2 start = end - 2 save_path += name_hash[start:end] + os.path.sep return {'path': save_path, 'hash': name_hash, }
Create a directory structure using the hashed filename :return: string of the path to save to not including filename/ext
train
https://github.com/xtream1101/cutil/blob/2e4d1f00e66154b44d4ccffb9b1db3f37e87f2e8/cutil/__init__.py#L251-L272
null
from cutil.database import Database # noqa: F401 from cutil.config import Config # noqa: F401 from cutil.custom_terminal import CustomTerminal # noqa: F401 from cutil.repeating_timer import RepeatingTimer # noqa: F401 import os import re import sys import uuid import math import time import pytz import json import queue import random import socket import urllib import hashlib import logging import datetime import threading import collections from hashids import Hashids from functools import wraps from operator import itemgetter from functools import cmp_to_key logger = logging.getLogger(__name__) #### # Time related functions #### def get_epoch(): """ :return: time as epoch """ return int(time.time()) def get_datetime(): """ :return: datetime object """ return datetime.datetime.now() def datetime_to_str(timestamp): return timestamp.isoformat() + "+0000" def datetime_to_utc(timestamp): return timestamp.astimezone(pytz.timezone('UTC')) def str_to_date(timestamp, formats=["%Y-%m-%dT%H:%M:%S.%f%z", "%Y-%m-%dT%H:%M:%S%z"]): rdata = None if timestamp is None: return None for time_format in formats: try: rdata = datetime.datetime.strptime(timestamp, time_format) except ValueError: rdata = None else: # If we did not raise an exception break return rdata ### # Threading ### def threads(num_threads, data, callback, *args, **kwargs): q = queue.Queue() item_list = [] def _thread_run(): while True: item = q.get() try: item_list.append(callback(item, *args, **kwargs)) except Exception: logger.exception("Error in _thread_run callback {} with item:\n{}".format(callback.__name__, item)) q.task_done() for i in range(num_threads): t = threading.Thread(target=_thread_run) t.daemon = True t.start() # Fill the Queue with the data to process for item in data: q.put(item) # Start processing the data q.join() return item_list #### # Other functions #### def multikey_sort(items, columns): """Source: https://stackoverflow.com/questions/1143671/python-sorting-list-of-dictionaries-by-multiple-keys """ comparers = [ ((itemgetter(col[1:].strip()), -1) if col.startswith('-') else (itemgetter(col.strip()), 1)) for col in columns ] def cmp(a, b): return (a > b) - (a < b) def comparer(left, right): comparer_iter = ( cmp(fn(left), fn(right)) * mult for fn, mult in comparers ) return next((result for result in comparer_iter if result), 0) return sorted(items, key=cmp_to_key(comparer)) def get_internal_ip(): return socket.gethostbyname(socket.gethostname()) def generate_key(value=None, salt=None, size=8): if value is None: value = random.randint(0, 999999) if salt is None: salt = random.randint(0, 999999) if not isinstance(size, int): # Enforce that size must be an int size = 8 hashids = Hashids(salt=str(salt), min_length=size) return hashids.encode(value) def create_uid(): # Named uid and not uuid because this function does not have to use uuid's uid = uuid.uuid4().hex logger.debug("Created new uid: {uid}".format(uid=uid)) return uid def make_url_safe(string): # Convert special chars to % chars return urllib.parse.quote_plus(string) def sanitize(string): """ Catch and replace invalid path chars [replace, with] """ replace_chars = [ ['\\', '-'], [':', '-'], ['/', '-'], ['?', ''], ['<', ''], ['>', ''], ['`', '`'], ['|', '-'], ['*', '`'], ['"', '\''], ['.', ''], ['&', 'and'] ] for ch in replace_chars: string = string.replace(ch[0], ch[1]) return string def rreplace(s, old, new, occurrence): """ Taken from: http://stackoverflow.com/a/2556252 """ li = s.rsplit(old, occurrence) return new.join(li) def flatten(dict_obj, prev_key='', sep='_'): items = {} for key, value in dict_obj.items(): new_key = prev_key + sep + key if prev_key != '' else key if isinstance(value, dict): items.update(flatten(value, new_key)) else: items[new_key] = value return items def update_dict(d, u): """ Source: https://stackoverflow.com/questions/3232943/update-value-of-a-nested-dictionary-of-varying-depth """ for k, v in u.items(): if isinstance(v, collections.Mapping): r = update_dict(d.get(k, {}), v) d[k] = r else: d[k] = u[k] return d def get_script_name(ext=False): name = os.path.basename(sys.argv[0]) if ext is False: name = name.split('.')[0] return name def chunks_of(max_chunk_size, list_to_chunk): """ Yields the list with a max size of max_chunk_size """ for i in range(0, len(list_to_chunk), max_chunk_size): yield list_to_chunk[i:i + max_chunk_size] def split_into(max_num_chunks, list_to_chunk): """ Yields the list with a max total size of max_num_chunks """ max_chunk_size = math.ceil(len(list_to_chunk) / max_num_chunks) return chunks_of(max_chunk_size, list_to_chunk) #### # File/filesystem related function #### def get_file_ext(file): file_name, file_extension = os.path.splitext(file) return file_extension def norm_path(path): """ :return: Proper path for os with vars expanded out """ # path = os.path.normcase(path) path = os.path.expanduser(path) path = os.path.expandvars(path) path = os.path.normpath(path) return path def create_path(path, is_dir=False): """ Check if path exists, if not create it :param path: path or file to create directory for :param is_dir: pass True if we are passing in a directory, default = False :return: os safe path from `path` """ path = norm_path(path) path_check = path if not is_dir: path_check = os.path.dirname(path) does_path_exists = os.path.exists(path_check) if does_path_exists: return path try: os.makedirs(path_check) except OSError: pass return path def dump_json(file_, data, **kwargs): json_args = {'sort_keys': True, 'indent': 4} json_args.update(**kwargs) create_path(file_) if not file_.endswith('.json'): file_ += '.json' with open(file_, 'w') as outfile: json.dump(data, outfile, **json_args) # Lets only do this once price_pattern = re.compile('(?P<low>[\d,.\s]+)(?:\D*(?P<high>[\d,.\s]+))?') def parse_price(price): found_price = {'low': None, 'high': None } price_raw = re.search(price_pattern, price) if price_raw: matched = price_raw.groupdict() found_price['low'] = matched.get('low') found_price['high'] = matched.get('high') for key, value in found_price.items(): if value is not None: value = value.strip() new_value = value.replace(',', '').replace('.', '').replace(' ', '') try: # Check if price has cents if value[-3] in [',', '.']: # Add . for cents back new_value = new_value[:-2] + '.' + new_value[-2:] except IndexError: # Price is 99 or less with no cents pass if new_value != '': found_price[key] = float(new_value) else: found_price[key] = None return found_price def get_image_dimension(url): import requests from PIL import Image # pip install pillow from io import BytesIO size = {'width': None, 'height': None, } try: if url.startswith('//'): url = "http:" + url data = requests.get(url, timeout=15).content im = Image.open(BytesIO(data)) size['width'], size['height'] = im.size except Exception: logger.debug("Error getting image size {url}".format(url=url), exc_info=True) logger.warning("Error getting image size {url}".format(url=url)) return size def crop_image(image_file, output_file=None, height=None, width=None, x=None, y=None): from PIL import Image # pip install pillow if output_file is None or height is None or width is None or x is None or y is None: logger.error("Must pass in all params: output_file, height, width, x, and y as named args") return None if width <= 0 or height <= 0: logger.warning("Width and height must be 1 or greater") raise ValueError image = Image.open(image_file) image.crop((x, y, width + x, height + y)).save(output_file) return output_file #### # Decorators #### def rate_limited(num_calls=1, every=1.0): """ Source: https://github.com/tomasbasham/ratelimit/tree/0ca5a616fa6d184fa180b9ad0b6fd0cf54c46936 Need to make a few changes that included having num_calls be a float Prevent a method from being called if it was previously called before a time widows has elapsed. Keyword Arguments: num_calls (float): Maximum method invocations within a period. Must be greater than 0. every (float): A dampening factor (in seconds). Can be any number greater than 0. Return: function: Decorated function that will forward method invocations if the time window has elapsed. """ frequency = abs(every) / float(num_calls) def decorator(func): """ Extend the behaviour of the following function, forwarding method invocations if the time window hes elapsed. Arguments: func (function): The function to decorate Returns: function: Decorated function """ # To get around issues with function local scope # and reassigning variables, we wrap the time # within a list. When updating the value we're # not reassigning `last_called`, which would not # work, but instead reassigning the value at a # particular index. last_called = [0.0] # Add thread safety lock = threading.RLock() def wrapper(*args, **kargs): """Decorator wrapper function""" with lock: elapsed = time.time() - last_called[0] left_to_wait = frequency - elapsed if left_to_wait > 0: time.sleep(left_to_wait) last_called[0] = time.time() return func(*args, **kargs) return wrapper return decorator def rate_limited_old(max_per_second): """ Source: https://gist.github.com/gregburek/1441055 """ lock = threading.Lock() min_interval = 1.0 / max_per_second def decorate(func): last_time_called = time.perf_counter() @wraps(func) def rate_limited_function(*args, **kwargs): lock.acquire() nonlocal last_time_called try: elapsed = time.perf_counter() - last_time_called left_to_wait = min_interval - elapsed if left_to_wait > 0: time.sleep(left_to_wait) return func(*args, **kwargs) finally: last_time_called = time.perf_counter() lock.release() return rate_limited_function return decorate def timeit(stat_tracker_func, name): """ Pass in a function and the name of the stat Will time the function that this is a decorator to and send the `name` as well as the value (in seconds) to `stat_tracker_func` `stat_tracker_func` can be used to either print out the data or save it """ def _timeit(func): def wrapper(*args, **kw): start_time = time.time() result = func(*args, **kw) stop_time = time.time() stat_tracker_func(name, stop_time - start_time) return result return wrapper return _timeit #### # Regex #### # Keep re.compile's outside of fn as to only create it once proxy_parts_pattern = re.compile('^(?:(?P<schema>\w+):\/\/)(?:(?P<user>.*):(?P<password>.*)@)?(?P<host>[^:]*)(?::(?P<port>\d+))?$') def get_proxy_parts(proxy): """ Take a proxy url and break it up to its parts """ proxy_parts = {'schema': None, 'user': None, 'password': None, 'host': None, 'port': None, } # Find parts results = re.match(proxy_parts_pattern, proxy) if results: matched = results.groupdict() for key in proxy_parts: proxy_parts[key] = matched.get(key) else: logger.error("Invalid proxy format `{proxy}`".format(proxy=proxy)) if proxy_parts['port'] is None: proxy_parts['port'] = '80' return proxy_parts def remove_html_tag(input_str='', tag=None): """ Returns a string with the html tag and all its contents from a string """ result = input_str if tag is not None: pattern = re.compile('<{tag}[\s\S]+?/{tag}>'.format(tag=tag)) result = re.sub(pattern, '', str(input_str)) return result
xtream1101/cutil
cutil/__init__.py
create_path
python
def create_path(path, is_dir=False): path = norm_path(path) path_check = path if not is_dir: path_check = os.path.dirname(path) does_path_exists = os.path.exists(path_check) if does_path_exists: return path try: os.makedirs(path_check) except OSError: pass return path
Check if path exists, if not create it :param path: path or file to create directory for :param is_dir: pass True if we are passing in a directory, default = False :return: os safe path from `path`
train
https://github.com/xtream1101/cutil/blob/2e4d1f00e66154b44d4ccffb9b1db3f37e87f2e8/cutil/__init__.py#L275-L297
[ "def norm_path(path):\n \"\"\"\n :return: Proper path for os with vars expanded out\n \"\"\"\n # path = os.path.normcase(path)\n path = os.path.expanduser(path)\n path = os.path.expandvars(path)\n path = os.path.normpath(path)\n return path\n" ]
from cutil.database import Database # noqa: F401 from cutil.config import Config # noqa: F401 from cutil.custom_terminal import CustomTerminal # noqa: F401 from cutil.repeating_timer import RepeatingTimer # noqa: F401 import os import re import sys import uuid import math import time import pytz import json import queue import random import socket import urllib import hashlib import logging import datetime import threading import collections from hashids import Hashids from functools import wraps from operator import itemgetter from functools import cmp_to_key logger = logging.getLogger(__name__) #### # Time related functions #### def get_epoch(): """ :return: time as epoch """ return int(time.time()) def get_datetime(): """ :return: datetime object """ return datetime.datetime.now() def datetime_to_str(timestamp): return timestamp.isoformat() + "+0000" def datetime_to_utc(timestamp): return timestamp.astimezone(pytz.timezone('UTC')) def str_to_date(timestamp, formats=["%Y-%m-%dT%H:%M:%S.%f%z", "%Y-%m-%dT%H:%M:%S%z"]): rdata = None if timestamp is None: return None for time_format in formats: try: rdata = datetime.datetime.strptime(timestamp, time_format) except ValueError: rdata = None else: # If we did not raise an exception break return rdata ### # Threading ### def threads(num_threads, data, callback, *args, **kwargs): q = queue.Queue() item_list = [] def _thread_run(): while True: item = q.get() try: item_list.append(callback(item, *args, **kwargs)) except Exception: logger.exception("Error in _thread_run callback {} with item:\n{}".format(callback.__name__, item)) q.task_done() for i in range(num_threads): t = threading.Thread(target=_thread_run) t.daemon = True t.start() # Fill the Queue with the data to process for item in data: q.put(item) # Start processing the data q.join() return item_list #### # Other functions #### def multikey_sort(items, columns): """Source: https://stackoverflow.com/questions/1143671/python-sorting-list-of-dictionaries-by-multiple-keys """ comparers = [ ((itemgetter(col[1:].strip()), -1) if col.startswith('-') else (itemgetter(col.strip()), 1)) for col in columns ] def cmp(a, b): return (a > b) - (a < b) def comparer(left, right): comparer_iter = ( cmp(fn(left), fn(right)) * mult for fn, mult in comparers ) return next((result for result in comparer_iter if result), 0) return sorted(items, key=cmp_to_key(comparer)) def get_internal_ip(): return socket.gethostbyname(socket.gethostname()) def generate_key(value=None, salt=None, size=8): if value is None: value = random.randint(0, 999999) if salt is None: salt = random.randint(0, 999999) if not isinstance(size, int): # Enforce that size must be an int size = 8 hashids = Hashids(salt=str(salt), min_length=size) return hashids.encode(value) def create_uid(): # Named uid and not uuid because this function does not have to use uuid's uid = uuid.uuid4().hex logger.debug("Created new uid: {uid}".format(uid=uid)) return uid def make_url_safe(string): # Convert special chars to % chars return urllib.parse.quote_plus(string) def sanitize(string): """ Catch and replace invalid path chars [replace, with] """ replace_chars = [ ['\\', '-'], [':', '-'], ['/', '-'], ['?', ''], ['<', ''], ['>', ''], ['`', '`'], ['|', '-'], ['*', '`'], ['"', '\''], ['.', ''], ['&', 'and'] ] for ch in replace_chars: string = string.replace(ch[0], ch[1]) return string def rreplace(s, old, new, occurrence): """ Taken from: http://stackoverflow.com/a/2556252 """ li = s.rsplit(old, occurrence) return new.join(li) def flatten(dict_obj, prev_key='', sep='_'): items = {} for key, value in dict_obj.items(): new_key = prev_key + sep + key if prev_key != '' else key if isinstance(value, dict): items.update(flatten(value, new_key)) else: items[new_key] = value return items def update_dict(d, u): """ Source: https://stackoverflow.com/questions/3232943/update-value-of-a-nested-dictionary-of-varying-depth """ for k, v in u.items(): if isinstance(v, collections.Mapping): r = update_dict(d.get(k, {}), v) d[k] = r else: d[k] = u[k] return d def get_script_name(ext=False): name = os.path.basename(sys.argv[0]) if ext is False: name = name.split('.')[0] return name def chunks_of(max_chunk_size, list_to_chunk): """ Yields the list with a max size of max_chunk_size """ for i in range(0, len(list_to_chunk), max_chunk_size): yield list_to_chunk[i:i + max_chunk_size] def split_into(max_num_chunks, list_to_chunk): """ Yields the list with a max total size of max_num_chunks """ max_chunk_size = math.ceil(len(list_to_chunk) / max_num_chunks) return chunks_of(max_chunk_size, list_to_chunk) #### # File/filesystem related function #### def get_file_ext(file): file_name, file_extension = os.path.splitext(file) return file_extension def norm_path(path): """ :return: Proper path for os with vars expanded out """ # path = os.path.normcase(path) path = os.path.expanduser(path) path = os.path.expandvars(path) path = os.path.normpath(path) return path def create_hashed_path(base_path, name, depth=2): """ Create a directory structure using the hashed filename :return: string of the path to save to not including filename/ext """ if depth > 16: logger.warning("depth cannot be greater then 16, setting to 16") depth = 16 name_hash = hashlib.md5(str(name).encode('utf-8')).hexdigest() if base_path.endswith(os.path.sep): save_path = base_path else: save_path = base_path + os.path.sep for i in range(1, depth + 1): end = i * 2 start = end - 2 save_path += name_hash[start:end] + os.path.sep return {'path': save_path, 'hash': name_hash, } def dump_json(file_, data, **kwargs): json_args = {'sort_keys': True, 'indent': 4} json_args.update(**kwargs) create_path(file_) if not file_.endswith('.json'): file_ += '.json' with open(file_, 'w') as outfile: json.dump(data, outfile, **json_args) # Lets only do this once price_pattern = re.compile('(?P<low>[\d,.\s]+)(?:\D*(?P<high>[\d,.\s]+))?') def parse_price(price): found_price = {'low': None, 'high': None } price_raw = re.search(price_pattern, price) if price_raw: matched = price_raw.groupdict() found_price['low'] = matched.get('low') found_price['high'] = matched.get('high') for key, value in found_price.items(): if value is not None: value = value.strip() new_value = value.replace(',', '').replace('.', '').replace(' ', '') try: # Check if price has cents if value[-3] in [',', '.']: # Add . for cents back new_value = new_value[:-2] + '.' + new_value[-2:] except IndexError: # Price is 99 or less with no cents pass if new_value != '': found_price[key] = float(new_value) else: found_price[key] = None return found_price def get_image_dimension(url): import requests from PIL import Image # pip install pillow from io import BytesIO size = {'width': None, 'height': None, } try: if url.startswith('//'): url = "http:" + url data = requests.get(url, timeout=15).content im = Image.open(BytesIO(data)) size['width'], size['height'] = im.size except Exception: logger.debug("Error getting image size {url}".format(url=url), exc_info=True) logger.warning("Error getting image size {url}".format(url=url)) return size def crop_image(image_file, output_file=None, height=None, width=None, x=None, y=None): from PIL import Image # pip install pillow if output_file is None or height is None or width is None or x is None or y is None: logger.error("Must pass in all params: output_file, height, width, x, and y as named args") return None if width <= 0 or height <= 0: logger.warning("Width and height must be 1 or greater") raise ValueError image = Image.open(image_file) image.crop((x, y, width + x, height + y)).save(output_file) return output_file #### # Decorators #### def rate_limited(num_calls=1, every=1.0): """ Source: https://github.com/tomasbasham/ratelimit/tree/0ca5a616fa6d184fa180b9ad0b6fd0cf54c46936 Need to make a few changes that included having num_calls be a float Prevent a method from being called if it was previously called before a time widows has elapsed. Keyword Arguments: num_calls (float): Maximum method invocations within a period. Must be greater than 0. every (float): A dampening factor (in seconds). Can be any number greater than 0. Return: function: Decorated function that will forward method invocations if the time window has elapsed. """ frequency = abs(every) / float(num_calls) def decorator(func): """ Extend the behaviour of the following function, forwarding method invocations if the time window hes elapsed. Arguments: func (function): The function to decorate Returns: function: Decorated function """ # To get around issues with function local scope # and reassigning variables, we wrap the time # within a list. When updating the value we're # not reassigning `last_called`, which would not # work, but instead reassigning the value at a # particular index. last_called = [0.0] # Add thread safety lock = threading.RLock() def wrapper(*args, **kargs): """Decorator wrapper function""" with lock: elapsed = time.time() - last_called[0] left_to_wait = frequency - elapsed if left_to_wait > 0: time.sleep(left_to_wait) last_called[0] = time.time() return func(*args, **kargs) return wrapper return decorator def rate_limited_old(max_per_second): """ Source: https://gist.github.com/gregburek/1441055 """ lock = threading.Lock() min_interval = 1.0 / max_per_second def decorate(func): last_time_called = time.perf_counter() @wraps(func) def rate_limited_function(*args, **kwargs): lock.acquire() nonlocal last_time_called try: elapsed = time.perf_counter() - last_time_called left_to_wait = min_interval - elapsed if left_to_wait > 0: time.sleep(left_to_wait) return func(*args, **kwargs) finally: last_time_called = time.perf_counter() lock.release() return rate_limited_function return decorate def timeit(stat_tracker_func, name): """ Pass in a function and the name of the stat Will time the function that this is a decorator to and send the `name` as well as the value (in seconds) to `stat_tracker_func` `stat_tracker_func` can be used to either print out the data or save it """ def _timeit(func): def wrapper(*args, **kw): start_time = time.time() result = func(*args, **kw) stop_time = time.time() stat_tracker_func(name, stop_time - start_time) return result return wrapper return _timeit #### # Regex #### # Keep re.compile's outside of fn as to only create it once proxy_parts_pattern = re.compile('^(?:(?P<schema>\w+):\/\/)(?:(?P<user>.*):(?P<password>.*)@)?(?P<host>[^:]*)(?::(?P<port>\d+))?$') def get_proxy_parts(proxy): """ Take a proxy url and break it up to its parts """ proxy_parts = {'schema': None, 'user': None, 'password': None, 'host': None, 'port': None, } # Find parts results = re.match(proxy_parts_pattern, proxy) if results: matched = results.groupdict() for key in proxy_parts: proxy_parts[key] = matched.get(key) else: logger.error("Invalid proxy format `{proxy}`".format(proxy=proxy)) if proxy_parts['port'] is None: proxy_parts['port'] = '80' return proxy_parts def remove_html_tag(input_str='', tag=None): """ Returns a string with the html tag and all its contents from a string """ result = input_str if tag is not None: pattern = re.compile('<{tag}[\s\S]+?/{tag}>'.format(tag=tag)) result = re.sub(pattern, '', str(input_str)) return result
xtream1101/cutil
cutil/__init__.py
rate_limited
python
def rate_limited(num_calls=1, every=1.0): frequency = abs(every) / float(num_calls) def decorator(func): """ Extend the behaviour of the following function, forwarding method invocations if the time window hes elapsed. Arguments: func (function): The function to decorate Returns: function: Decorated function """ # To get around issues with function local scope # and reassigning variables, we wrap the time # within a list. When updating the value we're # not reassigning `last_called`, which would not # work, but instead reassigning the value at a # particular index. last_called = [0.0] # Add thread safety lock = threading.RLock() def wrapper(*args, **kargs): """Decorator wrapper function""" with lock: elapsed = time.time() - last_called[0] left_to_wait = frequency - elapsed if left_to_wait > 0: time.sleep(left_to_wait) last_called[0] = time.time() return func(*args, **kargs) return wrapper return decorator
Source: https://github.com/tomasbasham/ratelimit/tree/0ca5a616fa6d184fa180b9ad0b6fd0cf54c46936 Need to make a few changes that included having num_calls be a float Prevent a method from being called if it was previously called before a time widows has elapsed. Keyword Arguments: num_calls (float): Maximum method invocations within a period. Must be greater than 0. every (float): A dampening factor (in seconds). Can be any number greater than 0. Return: function: Decorated function that will forward method invocations if the time window has elapsed.
train
https://github.com/xtream1101/cutil/blob/2e4d1f00e66154b44d4ccffb9b1db3f37e87f2e8/cutil/__init__.py#L388-L440
null
from cutil.database import Database # noqa: F401 from cutil.config import Config # noqa: F401 from cutil.custom_terminal import CustomTerminal # noqa: F401 from cutil.repeating_timer import RepeatingTimer # noqa: F401 import os import re import sys import uuid import math import time import pytz import json import queue import random import socket import urllib import hashlib import logging import datetime import threading import collections from hashids import Hashids from functools import wraps from operator import itemgetter from functools import cmp_to_key logger = logging.getLogger(__name__) #### # Time related functions #### def get_epoch(): """ :return: time as epoch """ return int(time.time()) def get_datetime(): """ :return: datetime object """ return datetime.datetime.now() def datetime_to_str(timestamp): return timestamp.isoformat() + "+0000" def datetime_to_utc(timestamp): return timestamp.astimezone(pytz.timezone('UTC')) def str_to_date(timestamp, formats=["%Y-%m-%dT%H:%M:%S.%f%z", "%Y-%m-%dT%H:%M:%S%z"]): rdata = None if timestamp is None: return None for time_format in formats: try: rdata = datetime.datetime.strptime(timestamp, time_format) except ValueError: rdata = None else: # If we did not raise an exception break return rdata ### # Threading ### def threads(num_threads, data, callback, *args, **kwargs): q = queue.Queue() item_list = [] def _thread_run(): while True: item = q.get() try: item_list.append(callback(item, *args, **kwargs)) except Exception: logger.exception("Error in _thread_run callback {} with item:\n{}".format(callback.__name__, item)) q.task_done() for i in range(num_threads): t = threading.Thread(target=_thread_run) t.daemon = True t.start() # Fill the Queue with the data to process for item in data: q.put(item) # Start processing the data q.join() return item_list #### # Other functions #### def multikey_sort(items, columns): """Source: https://stackoverflow.com/questions/1143671/python-sorting-list-of-dictionaries-by-multiple-keys """ comparers = [ ((itemgetter(col[1:].strip()), -1) if col.startswith('-') else (itemgetter(col.strip()), 1)) for col in columns ] def cmp(a, b): return (a > b) - (a < b) def comparer(left, right): comparer_iter = ( cmp(fn(left), fn(right)) * mult for fn, mult in comparers ) return next((result for result in comparer_iter if result), 0) return sorted(items, key=cmp_to_key(comparer)) def get_internal_ip(): return socket.gethostbyname(socket.gethostname()) def generate_key(value=None, salt=None, size=8): if value is None: value = random.randint(0, 999999) if salt is None: salt = random.randint(0, 999999) if not isinstance(size, int): # Enforce that size must be an int size = 8 hashids = Hashids(salt=str(salt), min_length=size) return hashids.encode(value) def create_uid(): # Named uid and not uuid because this function does not have to use uuid's uid = uuid.uuid4().hex logger.debug("Created new uid: {uid}".format(uid=uid)) return uid def make_url_safe(string): # Convert special chars to % chars return urllib.parse.quote_plus(string) def sanitize(string): """ Catch and replace invalid path chars [replace, with] """ replace_chars = [ ['\\', '-'], [':', '-'], ['/', '-'], ['?', ''], ['<', ''], ['>', ''], ['`', '`'], ['|', '-'], ['*', '`'], ['"', '\''], ['.', ''], ['&', 'and'] ] for ch in replace_chars: string = string.replace(ch[0], ch[1]) return string def rreplace(s, old, new, occurrence): """ Taken from: http://stackoverflow.com/a/2556252 """ li = s.rsplit(old, occurrence) return new.join(li) def flatten(dict_obj, prev_key='', sep='_'): items = {} for key, value in dict_obj.items(): new_key = prev_key + sep + key if prev_key != '' else key if isinstance(value, dict): items.update(flatten(value, new_key)) else: items[new_key] = value return items def update_dict(d, u): """ Source: https://stackoverflow.com/questions/3232943/update-value-of-a-nested-dictionary-of-varying-depth """ for k, v in u.items(): if isinstance(v, collections.Mapping): r = update_dict(d.get(k, {}), v) d[k] = r else: d[k] = u[k] return d def get_script_name(ext=False): name = os.path.basename(sys.argv[0]) if ext is False: name = name.split('.')[0] return name def chunks_of(max_chunk_size, list_to_chunk): """ Yields the list with a max size of max_chunk_size """ for i in range(0, len(list_to_chunk), max_chunk_size): yield list_to_chunk[i:i + max_chunk_size] def split_into(max_num_chunks, list_to_chunk): """ Yields the list with a max total size of max_num_chunks """ max_chunk_size = math.ceil(len(list_to_chunk) / max_num_chunks) return chunks_of(max_chunk_size, list_to_chunk) #### # File/filesystem related function #### def get_file_ext(file): file_name, file_extension = os.path.splitext(file) return file_extension def norm_path(path): """ :return: Proper path for os with vars expanded out """ # path = os.path.normcase(path) path = os.path.expanduser(path) path = os.path.expandvars(path) path = os.path.normpath(path) return path def create_hashed_path(base_path, name, depth=2): """ Create a directory structure using the hashed filename :return: string of the path to save to not including filename/ext """ if depth > 16: logger.warning("depth cannot be greater then 16, setting to 16") depth = 16 name_hash = hashlib.md5(str(name).encode('utf-8')).hexdigest() if base_path.endswith(os.path.sep): save_path = base_path else: save_path = base_path + os.path.sep for i in range(1, depth + 1): end = i * 2 start = end - 2 save_path += name_hash[start:end] + os.path.sep return {'path': save_path, 'hash': name_hash, } def create_path(path, is_dir=False): """ Check if path exists, if not create it :param path: path or file to create directory for :param is_dir: pass True if we are passing in a directory, default = False :return: os safe path from `path` """ path = norm_path(path) path_check = path if not is_dir: path_check = os.path.dirname(path) does_path_exists = os.path.exists(path_check) if does_path_exists: return path try: os.makedirs(path_check) except OSError: pass return path def dump_json(file_, data, **kwargs): json_args = {'sort_keys': True, 'indent': 4} json_args.update(**kwargs) create_path(file_) if not file_.endswith('.json'): file_ += '.json' with open(file_, 'w') as outfile: json.dump(data, outfile, **json_args) # Lets only do this once price_pattern = re.compile('(?P<low>[\d,.\s]+)(?:\D*(?P<high>[\d,.\s]+))?') def parse_price(price): found_price = {'low': None, 'high': None } price_raw = re.search(price_pattern, price) if price_raw: matched = price_raw.groupdict() found_price['low'] = matched.get('low') found_price['high'] = matched.get('high') for key, value in found_price.items(): if value is not None: value = value.strip() new_value = value.replace(',', '').replace('.', '').replace(' ', '') try: # Check if price has cents if value[-3] in [',', '.']: # Add . for cents back new_value = new_value[:-2] + '.' + new_value[-2:] except IndexError: # Price is 99 or less with no cents pass if new_value != '': found_price[key] = float(new_value) else: found_price[key] = None return found_price def get_image_dimension(url): import requests from PIL import Image # pip install pillow from io import BytesIO size = {'width': None, 'height': None, } try: if url.startswith('//'): url = "http:" + url data = requests.get(url, timeout=15).content im = Image.open(BytesIO(data)) size['width'], size['height'] = im.size except Exception: logger.debug("Error getting image size {url}".format(url=url), exc_info=True) logger.warning("Error getting image size {url}".format(url=url)) return size def crop_image(image_file, output_file=None, height=None, width=None, x=None, y=None): from PIL import Image # pip install pillow if output_file is None or height is None or width is None or x is None or y is None: logger.error("Must pass in all params: output_file, height, width, x, and y as named args") return None if width <= 0 or height <= 0: logger.warning("Width and height must be 1 or greater") raise ValueError image = Image.open(image_file) image.crop((x, y, width + x, height + y)).save(output_file) return output_file #### # Decorators #### def rate_limited_old(max_per_second): """ Source: https://gist.github.com/gregburek/1441055 """ lock = threading.Lock() min_interval = 1.0 / max_per_second def decorate(func): last_time_called = time.perf_counter() @wraps(func) def rate_limited_function(*args, **kwargs): lock.acquire() nonlocal last_time_called try: elapsed = time.perf_counter() - last_time_called left_to_wait = min_interval - elapsed if left_to_wait > 0: time.sleep(left_to_wait) return func(*args, **kwargs) finally: last_time_called = time.perf_counter() lock.release() return rate_limited_function return decorate def timeit(stat_tracker_func, name): """ Pass in a function and the name of the stat Will time the function that this is a decorator to and send the `name` as well as the value (in seconds) to `stat_tracker_func` `stat_tracker_func` can be used to either print out the data or save it """ def _timeit(func): def wrapper(*args, **kw): start_time = time.time() result = func(*args, **kw) stop_time = time.time() stat_tracker_func(name, stop_time - start_time) return result return wrapper return _timeit #### # Regex #### # Keep re.compile's outside of fn as to only create it once proxy_parts_pattern = re.compile('^(?:(?P<schema>\w+):\/\/)(?:(?P<user>.*):(?P<password>.*)@)?(?P<host>[^:]*)(?::(?P<port>\d+))?$') def get_proxy_parts(proxy): """ Take a proxy url and break it up to its parts """ proxy_parts = {'schema': None, 'user': None, 'password': None, 'host': None, 'port': None, } # Find parts results = re.match(proxy_parts_pattern, proxy) if results: matched = results.groupdict() for key in proxy_parts: proxy_parts[key] = matched.get(key) else: logger.error("Invalid proxy format `{proxy}`".format(proxy=proxy)) if proxy_parts['port'] is None: proxy_parts['port'] = '80' return proxy_parts def remove_html_tag(input_str='', tag=None): """ Returns a string with the html tag and all its contents from a string """ result = input_str if tag is not None: pattern = re.compile('<{tag}[\s\S]+?/{tag}>'.format(tag=tag)) result = re.sub(pattern, '', str(input_str)) return result
xtream1101/cutil
cutil/__init__.py
rate_limited_old
python
def rate_limited_old(max_per_second): lock = threading.Lock() min_interval = 1.0 / max_per_second def decorate(func): last_time_called = time.perf_counter() @wraps(func) def rate_limited_function(*args, **kwargs): lock.acquire() nonlocal last_time_called try: elapsed = time.perf_counter() - last_time_called left_to_wait = min_interval - elapsed if left_to_wait > 0: time.sleep(left_to_wait) return func(*args, **kwargs) finally: last_time_called = time.perf_counter() lock.release() return rate_limited_function return decorate
Source: https://gist.github.com/gregburek/1441055
train
https://github.com/xtream1101/cutil/blob/2e4d1f00e66154b44d4ccffb9b1db3f37e87f2e8/cutil/__init__.py#L443-L470
null
from cutil.database import Database # noqa: F401 from cutil.config import Config # noqa: F401 from cutil.custom_terminal import CustomTerminal # noqa: F401 from cutil.repeating_timer import RepeatingTimer # noqa: F401 import os import re import sys import uuid import math import time import pytz import json import queue import random import socket import urllib import hashlib import logging import datetime import threading import collections from hashids import Hashids from functools import wraps from operator import itemgetter from functools import cmp_to_key logger = logging.getLogger(__name__) #### # Time related functions #### def get_epoch(): """ :return: time as epoch """ return int(time.time()) def get_datetime(): """ :return: datetime object """ return datetime.datetime.now() def datetime_to_str(timestamp): return timestamp.isoformat() + "+0000" def datetime_to_utc(timestamp): return timestamp.astimezone(pytz.timezone('UTC')) def str_to_date(timestamp, formats=["%Y-%m-%dT%H:%M:%S.%f%z", "%Y-%m-%dT%H:%M:%S%z"]): rdata = None if timestamp is None: return None for time_format in formats: try: rdata = datetime.datetime.strptime(timestamp, time_format) except ValueError: rdata = None else: # If we did not raise an exception break return rdata ### # Threading ### def threads(num_threads, data, callback, *args, **kwargs): q = queue.Queue() item_list = [] def _thread_run(): while True: item = q.get() try: item_list.append(callback(item, *args, **kwargs)) except Exception: logger.exception("Error in _thread_run callback {} with item:\n{}".format(callback.__name__, item)) q.task_done() for i in range(num_threads): t = threading.Thread(target=_thread_run) t.daemon = True t.start() # Fill the Queue with the data to process for item in data: q.put(item) # Start processing the data q.join() return item_list #### # Other functions #### def multikey_sort(items, columns): """Source: https://stackoverflow.com/questions/1143671/python-sorting-list-of-dictionaries-by-multiple-keys """ comparers = [ ((itemgetter(col[1:].strip()), -1) if col.startswith('-') else (itemgetter(col.strip()), 1)) for col in columns ] def cmp(a, b): return (a > b) - (a < b) def comparer(left, right): comparer_iter = ( cmp(fn(left), fn(right)) * mult for fn, mult in comparers ) return next((result for result in comparer_iter if result), 0) return sorted(items, key=cmp_to_key(comparer)) def get_internal_ip(): return socket.gethostbyname(socket.gethostname()) def generate_key(value=None, salt=None, size=8): if value is None: value = random.randint(0, 999999) if salt is None: salt = random.randint(0, 999999) if not isinstance(size, int): # Enforce that size must be an int size = 8 hashids = Hashids(salt=str(salt), min_length=size) return hashids.encode(value) def create_uid(): # Named uid and not uuid because this function does not have to use uuid's uid = uuid.uuid4().hex logger.debug("Created new uid: {uid}".format(uid=uid)) return uid def make_url_safe(string): # Convert special chars to % chars return urllib.parse.quote_plus(string) def sanitize(string): """ Catch and replace invalid path chars [replace, with] """ replace_chars = [ ['\\', '-'], [':', '-'], ['/', '-'], ['?', ''], ['<', ''], ['>', ''], ['`', '`'], ['|', '-'], ['*', '`'], ['"', '\''], ['.', ''], ['&', 'and'] ] for ch in replace_chars: string = string.replace(ch[0], ch[1]) return string def rreplace(s, old, new, occurrence): """ Taken from: http://stackoverflow.com/a/2556252 """ li = s.rsplit(old, occurrence) return new.join(li) def flatten(dict_obj, prev_key='', sep='_'): items = {} for key, value in dict_obj.items(): new_key = prev_key + sep + key if prev_key != '' else key if isinstance(value, dict): items.update(flatten(value, new_key)) else: items[new_key] = value return items def update_dict(d, u): """ Source: https://stackoverflow.com/questions/3232943/update-value-of-a-nested-dictionary-of-varying-depth """ for k, v in u.items(): if isinstance(v, collections.Mapping): r = update_dict(d.get(k, {}), v) d[k] = r else: d[k] = u[k] return d def get_script_name(ext=False): name = os.path.basename(sys.argv[0]) if ext is False: name = name.split('.')[0] return name def chunks_of(max_chunk_size, list_to_chunk): """ Yields the list with a max size of max_chunk_size """ for i in range(0, len(list_to_chunk), max_chunk_size): yield list_to_chunk[i:i + max_chunk_size] def split_into(max_num_chunks, list_to_chunk): """ Yields the list with a max total size of max_num_chunks """ max_chunk_size = math.ceil(len(list_to_chunk) / max_num_chunks) return chunks_of(max_chunk_size, list_to_chunk) #### # File/filesystem related function #### def get_file_ext(file): file_name, file_extension = os.path.splitext(file) return file_extension def norm_path(path): """ :return: Proper path for os with vars expanded out """ # path = os.path.normcase(path) path = os.path.expanduser(path) path = os.path.expandvars(path) path = os.path.normpath(path) return path def create_hashed_path(base_path, name, depth=2): """ Create a directory structure using the hashed filename :return: string of the path to save to not including filename/ext """ if depth > 16: logger.warning("depth cannot be greater then 16, setting to 16") depth = 16 name_hash = hashlib.md5(str(name).encode('utf-8')).hexdigest() if base_path.endswith(os.path.sep): save_path = base_path else: save_path = base_path + os.path.sep for i in range(1, depth + 1): end = i * 2 start = end - 2 save_path += name_hash[start:end] + os.path.sep return {'path': save_path, 'hash': name_hash, } def create_path(path, is_dir=False): """ Check if path exists, if not create it :param path: path or file to create directory for :param is_dir: pass True if we are passing in a directory, default = False :return: os safe path from `path` """ path = norm_path(path) path_check = path if not is_dir: path_check = os.path.dirname(path) does_path_exists = os.path.exists(path_check) if does_path_exists: return path try: os.makedirs(path_check) except OSError: pass return path def dump_json(file_, data, **kwargs): json_args = {'sort_keys': True, 'indent': 4} json_args.update(**kwargs) create_path(file_) if not file_.endswith('.json'): file_ += '.json' with open(file_, 'w') as outfile: json.dump(data, outfile, **json_args) # Lets only do this once price_pattern = re.compile('(?P<low>[\d,.\s]+)(?:\D*(?P<high>[\d,.\s]+))?') def parse_price(price): found_price = {'low': None, 'high': None } price_raw = re.search(price_pattern, price) if price_raw: matched = price_raw.groupdict() found_price['low'] = matched.get('low') found_price['high'] = matched.get('high') for key, value in found_price.items(): if value is not None: value = value.strip() new_value = value.replace(',', '').replace('.', '').replace(' ', '') try: # Check if price has cents if value[-3] in [',', '.']: # Add . for cents back new_value = new_value[:-2] + '.' + new_value[-2:] except IndexError: # Price is 99 or less with no cents pass if new_value != '': found_price[key] = float(new_value) else: found_price[key] = None return found_price def get_image_dimension(url): import requests from PIL import Image # pip install pillow from io import BytesIO size = {'width': None, 'height': None, } try: if url.startswith('//'): url = "http:" + url data = requests.get(url, timeout=15).content im = Image.open(BytesIO(data)) size['width'], size['height'] = im.size except Exception: logger.debug("Error getting image size {url}".format(url=url), exc_info=True) logger.warning("Error getting image size {url}".format(url=url)) return size def crop_image(image_file, output_file=None, height=None, width=None, x=None, y=None): from PIL import Image # pip install pillow if output_file is None or height is None or width is None or x is None or y is None: logger.error("Must pass in all params: output_file, height, width, x, and y as named args") return None if width <= 0 or height <= 0: logger.warning("Width and height must be 1 or greater") raise ValueError image = Image.open(image_file) image.crop((x, y, width + x, height + y)).save(output_file) return output_file #### # Decorators #### def rate_limited(num_calls=1, every=1.0): """ Source: https://github.com/tomasbasham/ratelimit/tree/0ca5a616fa6d184fa180b9ad0b6fd0cf54c46936 Need to make a few changes that included having num_calls be a float Prevent a method from being called if it was previously called before a time widows has elapsed. Keyword Arguments: num_calls (float): Maximum method invocations within a period. Must be greater than 0. every (float): A dampening factor (in seconds). Can be any number greater than 0. Return: function: Decorated function that will forward method invocations if the time window has elapsed. """ frequency = abs(every) / float(num_calls) def decorator(func): """ Extend the behaviour of the following function, forwarding method invocations if the time window hes elapsed. Arguments: func (function): The function to decorate Returns: function: Decorated function """ # To get around issues with function local scope # and reassigning variables, we wrap the time # within a list. When updating the value we're # not reassigning `last_called`, which would not # work, but instead reassigning the value at a # particular index. last_called = [0.0] # Add thread safety lock = threading.RLock() def wrapper(*args, **kargs): """Decorator wrapper function""" with lock: elapsed = time.time() - last_called[0] left_to_wait = frequency - elapsed if left_to_wait > 0: time.sleep(left_to_wait) last_called[0] = time.time() return func(*args, **kargs) return wrapper return decorator def timeit(stat_tracker_func, name): """ Pass in a function and the name of the stat Will time the function that this is a decorator to and send the `name` as well as the value (in seconds) to `stat_tracker_func` `stat_tracker_func` can be used to either print out the data or save it """ def _timeit(func): def wrapper(*args, **kw): start_time = time.time() result = func(*args, **kw) stop_time = time.time() stat_tracker_func(name, stop_time - start_time) return result return wrapper return _timeit #### # Regex #### # Keep re.compile's outside of fn as to only create it once proxy_parts_pattern = re.compile('^(?:(?P<schema>\w+):\/\/)(?:(?P<user>.*):(?P<password>.*)@)?(?P<host>[^:]*)(?::(?P<port>\d+))?$') def get_proxy_parts(proxy): """ Take a proxy url and break it up to its parts """ proxy_parts = {'schema': None, 'user': None, 'password': None, 'host': None, 'port': None, } # Find parts results = re.match(proxy_parts_pattern, proxy) if results: matched = results.groupdict() for key in proxy_parts: proxy_parts[key] = matched.get(key) else: logger.error("Invalid proxy format `{proxy}`".format(proxy=proxy)) if proxy_parts['port'] is None: proxy_parts['port'] = '80' return proxy_parts def remove_html_tag(input_str='', tag=None): """ Returns a string with the html tag and all its contents from a string """ result = input_str if tag is not None: pattern = re.compile('<{tag}[\s\S]+?/{tag}>'.format(tag=tag)) result = re.sub(pattern, '', str(input_str)) return result
xtream1101/cutil
cutil/__init__.py
timeit
python
def timeit(stat_tracker_func, name): def _timeit(func): def wrapper(*args, **kw): start_time = time.time() result = func(*args, **kw) stop_time = time.time() stat_tracker_func(name, stop_time - start_time) return result return wrapper return _timeit
Pass in a function and the name of the stat Will time the function that this is a decorator to and send the `name` as well as the value (in seconds) to `stat_tracker_func` `stat_tracker_func` can be used to either print out the data or save it
train
https://github.com/xtream1101/cutil/blob/2e4d1f00e66154b44d4ccffb9b1db3f37e87f2e8/cutil/__init__.py#L473-L491
null
from cutil.database import Database # noqa: F401 from cutil.config import Config # noqa: F401 from cutil.custom_terminal import CustomTerminal # noqa: F401 from cutil.repeating_timer import RepeatingTimer # noqa: F401 import os import re import sys import uuid import math import time import pytz import json import queue import random import socket import urllib import hashlib import logging import datetime import threading import collections from hashids import Hashids from functools import wraps from operator import itemgetter from functools import cmp_to_key logger = logging.getLogger(__name__) #### # Time related functions #### def get_epoch(): """ :return: time as epoch """ return int(time.time()) def get_datetime(): """ :return: datetime object """ return datetime.datetime.now() def datetime_to_str(timestamp): return timestamp.isoformat() + "+0000" def datetime_to_utc(timestamp): return timestamp.astimezone(pytz.timezone('UTC')) def str_to_date(timestamp, formats=["%Y-%m-%dT%H:%M:%S.%f%z", "%Y-%m-%dT%H:%M:%S%z"]): rdata = None if timestamp is None: return None for time_format in formats: try: rdata = datetime.datetime.strptime(timestamp, time_format) except ValueError: rdata = None else: # If we did not raise an exception break return rdata ### # Threading ### def threads(num_threads, data, callback, *args, **kwargs): q = queue.Queue() item_list = [] def _thread_run(): while True: item = q.get() try: item_list.append(callback(item, *args, **kwargs)) except Exception: logger.exception("Error in _thread_run callback {} with item:\n{}".format(callback.__name__, item)) q.task_done() for i in range(num_threads): t = threading.Thread(target=_thread_run) t.daemon = True t.start() # Fill the Queue with the data to process for item in data: q.put(item) # Start processing the data q.join() return item_list #### # Other functions #### def multikey_sort(items, columns): """Source: https://stackoverflow.com/questions/1143671/python-sorting-list-of-dictionaries-by-multiple-keys """ comparers = [ ((itemgetter(col[1:].strip()), -1) if col.startswith('-') else (itemgetter(col.strip()), 1)) for col in columns ] def cmp(a, b): return (a > b) - (a < b) def comparer(left, right): comparer_iter = ( cmp(fn(left), fn(right)) * mult for fn, mult in comparers ) return next((result for result in comparer_iter if result), 0) return sorted(items, key=cmp_to_key(comparer)) def get_internal_ip(): return socket.gethostbyname(socket.gethostname()) def generate_key(value=None, salt=None, size=8): if value is None: value = random.randint(0, 999999) if salt is None: salt = random.randint(0, 999999) if not isinstance(size, int): # Enforce that size must be an int size = 8 hashids = Hashids(salt=str(salt), min_length=size) return hashids.encode(value) def create_uid(): # Named uid and not uuid because this function does not have to use uuid's uid = uuid.uuid4().hex logger.debug("Created new uid: {uid}".format(uid=uid)) return uid def make_url_safe(string): # Convert special chars to % chars return urllib.parse.quote_plus(string) def sanitize(string): """ Catch and replace invalid path chars [replace, with] """ replace_chars = [ ['\\', '-'], [':', '-'], ['/', '-'], ['?', ''], ['<', ''], ['>', ''], ['`', '`'], ['|', '-'], ['*', '`'], ['"', '\''], ['.', ''], ['&', 'and'] ] for ch in replace_chars: string = string.replace(ch[0], ch[1]) return string def rreplace(s, old, new, occurrence): """ Taken from: http://stackoverflow.com/a/2556252 """ li = s.rsplit(old, occurrence) return new.join(li) def flatten(dict_obj, prev_key='', sep='_'): items = {} for key, value in dict_obj.items(): new_key = prev_key + sep + key if prev_key != '' else key if isinstance(value, dict): items.update(flatten(value, new_key)) else: items[new_key] = value return items def update_dict(d, u): """ Source: https://stackoverflow.com/questions/3232943/update-value-of-a-nested-dictionary-of-varying-depth """ for k, v in u.items(): if isinstance(v, collections.Mapping): r = update_dict(d.get(k, {}), v) d[k] = r else: d[k] = u[k] return d def get_script_name(ext=False): name = os.path.basename(sys.argv[0]) if ext is False: name = name.split('.')[0] return name def chunks_of(max_chunk_size, list_to_chunk): """ Yields the list with a max size of max_chunk_size """ for i in range(0, len(list_to_chunk), max_chunk_size): yield list_to_chunk[i:i + max_chunk_size] def split_into(max_num_chunks, list_to_chunk): """ Yields the list with a max total size of max_num_chunks """ max_chunk_size = math.ceil(len(list_to_chunk) / max_num_chunks) return chunks_of(max_chunk_size, list_to_chunk) #### # File/filesystem related function #### def get_file_ext(file): file_name, file_extension = os.path.splitext(file) return file_extension def norm_path(path): """ :return: Proper path for os with vars expanded out """ # path = os.path.normcase(path) path = os.path.expanduser(path) path = os.path.expandvars(path) path = os.path.normpath(path) return path def create_hashed_path(base_path, name, depth=2): """ Create a directory structure using the hashed filename :return: string of the path to save to not including filename/ext """ if depth > 16: logger.warning("depth cannot be greater then 16, setting to 16") depth = 16 name_hash = hashlib.md5(str(name).encode('utf-8')).hexdigest() if base_path.endswith(os.path.sep): save_path = base_path else: save_path = base_path + os.path.sep for i in range(1, depth + 1): end = i * 2 start = end - 2 save_path += name_hash[start:end] + os.path.sep return {'path': save_path, 'hash': name_hash, } def create_path(path, is_dir=False): """ Check if path exists, if not create it :param path: path or file to create directory for :param is_dir: pass True if we are passing in a directory, default = False :return: os safe path from `path` """ path = norm_path(path) path_check = path if not is_dir: path_check = os.path.dirname(path) does_path_exists = os.path.exists(path_check) if does_path_exists: return path try: os.makedirs(path_check) except OSError: pass return path def dump_json(file_, data, **kwargs): json_args = {'sort_keys': True, 'indent': 4} json_args.update(**kwargs) create_path(file_) if not file_.endswith('.json'): file_ += '.json' with open(file_, 'w') as outfile: json.dump(data, outfile, **json_args) # Lets only do this once price_pattern = re.compile('(?P<low>[\d,.\s]+)(?:\D*(?P<high>[\d,.\s]+))?') def parse_price(price): found_price = {'low': None, 'high': None } price_raw = re.search(price_pattern, price) if price_raw: matched = price_raw.groupdict() found_price['low'] = matched.get('low') found_price['high'] = matched.get('high') for key, value in found_price.items(): if value is not None: value = value.strip() new_value = value.replace(',', '').replace('.', '').replace(' ', '') try: # Check if price has cents if value[-3] in [',', '.']: # Add . for cents back new_value = new_value[:-2] + '.' + new_value[-2:] except IndexError: # Price is 99 or less with no cents pass if new_value != '': found_price[key] = float(new_value) else: found_price[key] = None return found_price def get_image_dimension(url): import requests from PIL import Image # pip install pillow from io import BytesIO size = {'width': None, 'height': None, } try: if url.startswith('//'): url = "http:" + url data = requests.get(url, timeout=15).content im = Image.open(BytesIO(data)) size['width'], size['height'] = im.size except Exception: logger.debug("Error getting image size {url}".format(url=url), exc_info=True) logger.warning("Error getting image size {url}".format(url=url)) return size def crop_image(image_file, output_file=None, height=None, width=None, x=None, y=None): from PIL import Image # pip install pillow if output_file is None or height is None or width is None or x is None or y is None: logger.error("Must pass in all params: output_file, height, width, x, and y as named args") return None if width <= 0 or height <= 0: logger.warning("Width and height must be 1 or greater") raise ValueError image = Image.open(image_file) image.crop((x, y, width + x, height + y)).save(output_file) return output_file #### # Decorators #### def rate_limited(num_calls=1, every=1.0): """ Source: https://github.com/tomasbasham/ratelimit/tree/0ca5a616fa6d184fa180b9ad0b6fd0cf54c46936 Need to make a few changes that included having num_calls be a float Prevent a method from being called if it was previously called before a time widows has elapsed. Keyword Arguments: num_calls (float): Maximum method invocations within a period. Must be greater than 0. every (float): A dampening factor (in seconds). Can be any number greater than 0. Return: function: Decorated function that will forward method invocations if the time window has elapsed. """ frequency = abs(every) / float(num_calls) def decorator(func): """ Extend the behaviour of the following function, forwarding method invocations if the time window hes elapsed. Arguments: func (function): The function to decorate Returns: function: Decorated function """ # To get around issues with function local scope # and reassigning variables, we wrap the time # within a list. When updating the value we're # not reassigning `last_called`, which would not # work, but instead reassigning the value at a # particular index. last_called = [0.0] # Add thread safety lock = threading.RLock() def wrapper(*args, **kargs): """Decorator wrapper function""" with lock: elapsed = time.time() - last_called[0] left_to_wait = frequency - elapsed if left_to_wait > 0: time.sleep(left_to_wait) last_called[0] = time.time() return func(*args, **kargs) return wrapper return decorator def rate_limited_old(max_per_second): """ Source: https://gist.github.com/gregburek/1441055 """ lock = threading.Lock() min_interval = 1.0 / max_per_second def decorate(func): last_time_called = time.perf_counter() @wraps(func) def rate_limited_function(*args, **kwargs): lock.acquire() nonlocal last_time_called try: elapsed = time.perf_counter() - last_time_called left_to_wait = min_interval - elapsed if left_to_wait > 0: time.sleep(left_to_wait) return func(*args, **kwargs) finally: last_time_called = time.perf_counter() lock.release() return rate_limited_function return decorate #### # Regex #### # Keep re.compile's outside of fn as to only create it once proxy_parts_pattern = re.compile('^(?:(?P<schema>\w+):\/\/)(?:(?P<user>.*):(?P<password>.*)@)?(?P<host>[^:]*)(?::(?P<port>\d+))?$') def get_proxy_parts(proxy): """ Take a proxy url and break it up to its parts """ proxy_parts = {'schema': None, 'user': None, 'password': None, 'host': None, 'port': None, } # Find parts results = re.match(proxy_parts_pattern, proxy) if results: matched = results.groupdict() for key in proxy_parts: proxy_parts[key] = matched.get(key) else: logger.error("Invalid proxy format `{proxy}`".format(proxy=proxy)) if proxy_parts['port'] is None: proxy_parts['port'] = '80' return proxy_parts def remove_html_tag(input_str='', tag=None): """ Returns a string with the html tag and all its contents from a string """ result = input_str if tag is not None: pattern = re.compile('<{tag}[\s\S]+?/{tag}>'.format(tag=tag)) result = re.sub(pattern, '', str(input_str)) return result
xtream1101/cutil
cutil/__init__.py
get_proxy_parts
python
def get_proxy_parts(proxy): proxy_parts = {'schema': None, 'user': None, 'password': None, 'host': None, 'port': None, } # Find parts results = re.match(proxy_parts_pattern, proxy) if results: matched = results.groupdict() for key in proxy_parts: proxy_parts[key] = matched.get(key) else: logger.error("Invalid proxy format `{proxy}`".format(proxy=proxy)) if proxy_parts['port'] is None: proxy_parts['port'] = '80' return proxy_parts
Take a proxy url and break it up to its parts
train
https://github.com/xtream1101/cutil/blob/2e4d1f00e66154b44d4ccffb9b1db3f37e87f2e8/cutil/__init__.py#L501-L524
null
from cutil.database import Database # noqa: F401 from cutil.config import Config # noqa: F401 from cutil.custom_terminal import CustomTerminal # noqa: F401 from cutil.repeating_timer import RepeatingTimer # noqa: F401 import os import re import sys import uuid import math import time import pytz import json import queue import random import socket import urllib import hashlib import logging import datetime import threading import collections from hashids import Hashids from functools import wraps from operator import itemgetter from functools import cmp_to_key logger = logging.getLogger(__name__) #### # Time related functions #### def get_epoch(): """ :return: time as epoch """ return int(time.time()) def get_datetime(): """ :return: datetime object """ return datetime.datetime.now() def datetime_to_str(timestamp): return timestamp.isoformat() + "+0000" def datetime_to_utc(timestamp): return timestamp.astimezone(pytz.timezone('UTC')) def str_to_date(timestamp, formats=["%Y-%m-%dT%H:%M:%S.%f%z", "%Y-%m-%dT%H:%M:%S%z"]): rdata = None if timestamp is None: return None for time_format in formats: try: rdata = datetime.datetime.strptime(timestamp, time_format) except ValueError: rdata = None else: # If we did not raise an exception break return rdata ### # Threading ### def threads(num_threads, data, callback, *args, **kwargs): q = queue.Queue() item_list = [] def _thread_run(): while True: item = q.get() try: item_list.append(callback(item, *args, **kwargs)) except Exception: logger.exception("Error in _thread_run callback {} with item:\n{}".format(callback.__name__, item)) q.task_done() for i in range(num_threads): t = threading.Thread(target=_thread_run) t.daemon = True t.start() # Fill the Queue with the data to process for item in data: q.put(item) # Start processing the data q.join() return item_list #### # Other functions #### def multikey_sort(items, columns): """Source: https://stackoverflow.com/questions/1143671/python-sorting-list-of-dictionaries-by-multiple-keys """ comparers = [ ((itemgetter(col[1:].strip()), -1) if col.startswith('-') else (itemgetter(col.strip()), 1)) for col in columns ] def cmp(a, b): return (a > b) - (a < b) def comparer(left, right): comparer_iter = ( cmp(fn(left), fn(right)) * mult for fn, mult in comparers ) return next((result for result in comparer_iter if result), 0) return sorted(items, key=cmp_to_key(comparer)) def get_internal_ip(): return socket.gethostbyname(socket.gethostname()) def generate_key(value=None, salt=None, size=8): if value is None: value = random.randint(0, 999999) if salt is None: salt = random.randint(0, 999999) if not isinstance(size, int): # Enforce that size must be an int size = 8 hashids = Hashids(salt=str(salt), min_length=size) return hashids.encode(value) def create_uid(): # Named uid and not uuid because this function does not have to use uuid's uid = uuid.uuid4().hex logger.debug("Created new uid: {uid}".format(uid=uid)) return uid def make_url_safe(string): # Convert special chars to % chars return urllib.parse.quote_plus(string) def sanitize(string): """ Catch and replace invalid path chars [replace, with] """ replace_chars = [ ['\\', '-'], [':', '-'], ['/', '-'], ['?', ''], ['<', ''], ['>', ''], ['`', '`'], ['|', '-'], ['*', '`'], ['"', '\''], ['.', ''], ['&', 'and'] ] for ch in replace_chars: string = string.replace(ch[0], ch[1]) return string def rreplace(s, old, new, occurrence): """ Taken from: http://stackoverflow.com/a/2556252 """ li = s.rsplit(old, occurrence) return new.join(li) def flatten(dict_obj, prev_key='', sep='_'): items = {} for key, value in dict_obj.items(): new_key = prev_key + sep + key if prev_key != '' else key if isinstance(value, dict): items.update(flatten(value, new_key)) else: items[new_key] = value return items def update_dict(d, u): """ Source: https://stackoverflow.com/questions/3232943/update-value-of-a-nested-dictionary-of-varying-depth """ for k, v in u.items(): if isinstance(v, collections.Mapping): r = update_dict(d.get(k, {}), v) d[k] = r else: d[k] = u[k] return d def get_script_name(ext=False): name = os.path.basename(sys.argv[0]) if ext is False: name = name.split('.')[0] return name def chunks_of(max_chunk_size, list_to_chunk): """ Yields the list with a max size of max_chunk_size """ for i in range(0, len(list_to_chunk), max_chunk_size): yield list_to_chunk[i:i + max_chunk_size] def split_into(max_num_chunks, list_to_chunk): """ Yields the list with a max total size of max_num_chunks """ max_chunk_size = math.ceil(len(list_to_chunk) / max_num_chunks) return chunks_of(max_chunk_size, list_to_chunk) #### # File/filesystem related function #### def get_file_ext(file): file_name, file_extension = os.path.splitext(file) return file_extension def norm_path(path): """ :return: Proper path for os with vars expanded out """ # path = os.path.normcase(path) path = os.path.expanduser(path) path = os.path.expandvars(path) path = os.path.normpath(path) return path def create_hashed_path(base_path, name, depth=2): """ Create a directory structure using the hashed filename :return: string of the path to save to not including filename/ext """ if depth > 16: logger.warning("depth cannot be greater then 16, setting to 16") depth = 16 name_hash = hashlib.md5(str(name).encode('utf-8')).hexdigest() if base_path.endswith(os.path.sep): save_path = base_path else: save_path = base_path + os.path.sep for i in range(1, depth + 1): end = i * 2 start = end - 2 save_path += name_hash[start:end] + os.path.sep return {'path': save_path, 'hash': name_hash, } def create_path(path, is_dir=False): """ Check if path exists, if not create it :param path: path or file to create directory for :param is_dir: pass True if we are passing in a directory, default = False :return: os safe path from `path` """ path = norm_path(path) path_check = path if not is_dir: path_check = os.path.dirname(path) does_path_exists = os.path.exists(path_check) if does_path_exists: return path try: os.makedirs(path_check) except OSError: pass return path def dump_json(file_, data, **kwargs): json_args = {'sort_keys': True, 'indent': 4} json_args.update(**kwargs) create_path(file_) if not file_.endswith('.json'): file_ += '.json' with open(file_, 'w') as outfile: json.dump(data, outfile, **json_args) # Lets only do this once price_pattern = re.compile('(?P<low>[\d,.\s]+)(?:\D*(?P<high>[\d,.\s]+))?') def parse_price(price): found_price = {'low': None, 'high': None } price_raw = re.search(price_pattern, price) if price_raw: matched = price_raw.groupdict() found_price['low'] = matched.get('low') found_price['high'] = matched.get('high') for key, value in found_price.items(): if value is not None: value = value.strip() new_value = value.replace(',', '').replace('.', '').replace(' ', '') try: # Check if price has cents if value[-3] in [',', '.']: # Add . for cents back new_value = new_value[:-2] + '.' + new_value[-2:] except IndexError: # Price is 99 or less with no cents pass if new_value != '': found_price[key] = float(new_value) else: found_price[key] = None return found_price def get_image_dimension(url): import requests from PIL import Image # pip install pillow from io import BytesIO size = {'width': None, 'height': None, } try: if url.startswith('//'): url = "http:" + url data = requests.get(url, timeout=15).content im = Image.open(BytesIO(data)) size['width'], size['height'] = im.size except Exception: logger.debug("Error getting image size {url}".format(url=url), exc_info=True) logger.warning("Error getting image size {url}".format(url=url)) return size def crop_image(image_file, output_file=None, height=None, width=None, x=None, y=None): from PIL import Image # pip install pillow if output_file is None or height is None or width is None or x is None or y is None: logger.error("Must pass in all params: output_file, height, width, x, and y as named args") return None if width <= 0 or height <= 0: logger.warning("Width and height must be 1 or greater") raise ValueError image = Image.open(image_file) image.crop((x, y, width + x, height + y)).save(output_file) return output_file #### # Decorators #### def rate_limited(num_calls=1, every=1.0): """ Source: https://github.com/tomasbasham/ratelimit/tree/0ca5a616fa6d184fa180b9ad0b6fd0cf54c46936 Need to make a few changes that included having num_calls be a float Prevent a method from being called if it was previously called before a time widows has elapsed. Keyword Arguments: num_calls (float): Maximum method invocations within a period. Must be greater than 0. every (float): A dampening factor (in seconds). Can be any number greater than 0. Return: function: Decorated function that will forward method invocations if the time window has elapsed. """ frequency = abs(every) / float(num_calls) def decorator(func): """ Extend the behaviour of the following function, forwarding method invocations if the time window hes elapsed. Arguments: func (function): The function to decorate Returns: function: Decorated function """ # To get around issues with function local scope # and reassigning variables, we wrap the time # within a list. When updating the value we're # not reassigning `last_called`, which would not # work, but instead reassigning the value at a # particular index. last_called = [0.0] # Add thread safety lock = threading.RLock() def wrapper(*args, **kargs): """Decorator wrapper function""" with lock: elapsed = time.time() - last_called[0] left_to_wait = frequency - elapsed if left_to_wait > 0: time.sleep(left_to_wait) last_called[0] = time.time() return func(*args, **kargs) return wrapper return decorator def rate_limited_old(max_per_second): """ Source: https://gist.github.com/gregburek/1441055 """ lock = threading.Lock() min_interval = 1.0 / max_per_second def decorate(func): last_time_called = time.perf_counter() @wraps(func) def rate_limited_function(*args, **kwargs): lock.acquire() nonlocal last_time_called try: elapsed = time.perf_counter() - last_time_called left_to_wait = min_interval - elapsed if left_to_wait > 0: time.sleep(left_to_wait) return func(*args, **kwargs) finally: last_time_called = time.perf_counter() lock.release() return rate_limited_function return decorate def timeit(stat_tracker_func, name): """ Pass in a function and the name of the stat Will time the function that this is a decorator to and send the `name` as well as the value (in seconds) to `stat_tracker_func` `stat_tracker_func` can be used to either print out the data or save it """ def _timeit(func): def wrapper(*args, **kw): start_time = time.time() result = func(*args, **kw) stop_time = time.time() stat_tracker_func(name, stop_time - start_time) return result return wrapper return _timeit #### # Regex #### # Keep re.compile's outside of fn as to only create it once proxy_parts_pattern = re.compile('^(?:(?P<schema>\w+):\/\/)(?:(?P<user>.*):(?P<password>.*)@)?(?P<host>[^:]*)(?::(?P<port>\d+))?$') def remove_html_tag(input_str='', tag=None): """ Returns a string with the html tag and all its contents from a string """ result = input_str if tag is not None: pattern = re.compile('<{tag}[\s\S]+?/{tag}>'.format(tag=tag)) result = re.sub(pattern, '', str(input_str)) return result
xtream1101/cutil
cutil/__init__.py
remove_html_tag
python
def remove_html_tag(input_str='', tag=None): result = input_str if tag is not None: pattern = re.compile('<{tag}[\s\S]+?/{tag}>'.format(tag=tag)) result = re.sub(pattern, '', str(input_str)) return result
Returns a string with the html tag and all its contents from a string
train
https://github.com/xtream1101/cutil/blob/2e4d1f00e66154b44d4ccffb9b1db3f37e87f2e8/cutil/__init__.py#L527-L536
null
from cutil.database import Database # noqa: F401 from cutil.config import Config # noqa: F401 from cutil.custom_terminal import CustomTerminal # noqa: F401 from cutil.repeating_timer import RepeatingTimer # noqa: F401 import os import re import sys import uuid import math import time import pytz import json import queue import random import socket import urllib import hashlib import logging import datetime import threading import collections from hashids import Hashids from functools import wraps from operator import itemgetter from functools import cmp_to_key logger = logging.getLogger(__name__) #### # Time related functions #### def get_epoch(): """ :return: time as epoch """ return int(time.time()) def get_datetime(): """ :return: datetime object """ return datetime.datetime.now() def datetime_to_str(timestamp): return timestamp.isoformat() + "+0000" def datetime_to_utc(timestamp): return timestamp.astimezone(pytz.timezone('UTC')) def str_to_date(timestamp, formats=["%Y-%m-%dT%H:%M:%S.%f%z", "%Y-%m-%dT%H:%M:%S%z"]): rdata = None if timestamp is None: return None for time_format in formats: try: rdata = datetime.datetime.strptime(timestamp, time_format) except ValueError: rdata = None else: # If we did not raise an exception break return rdata ### # Threading ### def threads(num_threads, data, callback, *args, **kwargs): q = queue.Queue() item_list = [] def _thread_run(): while True: item = q.get() try: item_list.append(callback(item, *args, **kwargs)) except Exception: logger.exception("Error in _thread_run callback {} with item:\n{}".format(callback.__name__, item)) q.task_done() for i in range(num_threads): t = threading.Thread(target=_thread_run) t.daemon = True t.start() # Fill the Queue with the data to process for item in data: q.put(item) # Start processing the data q.join() return item_list #### # Other functions #### def multikey_sort(items, columns): """Source: https://stackoverflow.com/questions/1143671/python-sorting-list-of-dictionaries-by-multiple-keys """ comparers = [ ((itemgetter(col[1:].strip()), -1) if col.startswith('-') else (itemgetter(col.strip()), 1)) for col in columns ] def cmp(a, b): return (a > b) - (a < b) def comparer(left, right): comparer_iter = ( cmp(fn(left), fn(right)) * mult for fn, mult in comparers ) return next((result for result in comparer_iter if result), 0) return sorted(items, key=cmp_to_key(comparer)) def get_internal_ip(): return socket.gethostbyname(socket.gethostname()) def generate_key(value=None, salt=None, size=8): if value is None: value = random.randint(0, 999999) if salt is None: salt = random.randint(0, 999999) if not isinstance(size, int): # Enforce that size must be an int size = 8 hashids = Hashids(salt=str(salt), min_length=size) return hashids.encode(value) def create_uid(): # Named uid and not uuid because this function does not have to use uuid's uid = uuid.uuid4().hex logger.debug("Created new uid: {uid}".format(uid=uid)) return uid def make_url_safe(string): # Convert special chars to % chars return urllib.parse.quote_plus(string) def sanitize(string): """ Catch and replace invalid path chars [replace, with] """ replace_chars = [ ['\\', '-'], [':', '-'], ['/', '-'], ['?', ''], ['<', ''], ['>', ''], ['`', '`'], ['|', '-'], ['*', '`'], ['"', '\''], ['.', ''], ['&', 'and'] ] for ch in replace_chars: string = string.replace(ch[0], ch[1]) return string def rreplace(s, old, new, occurrence): """ Taken from: http://stackoverflow.com/a/2556252 """ li = s.rsplit(old, occurrence) return new.join(li) def flatten(dict_obj, prev_key='', sep='_'): items = {} for key, value in dict_obj.items(): new_key = prev_key + sep + key if prev_key != '' else key if isinstance(value, dict): items.update(flatten(value, new_key)) else: items[new_key] = value return items def update_dict(d, u): """ Source: https://stackoverflow.com/questions/3232943/update-value-of-a-nested-dictionary-of-varying-depth """ for k, v in u.items(): if isinstance(v, collections.Mapping): r = update_dict(d.get(k, {}), v) d[k] = r else: d[k] = u[k] return d def get_script_name(ext=False): name = os.path.basename(sys.argv[0]) if ext is False: name = name.split('.')[0] return name def chunks_of(max_chunk_size, list_to_chunk): """ Yields the list with a max size of max_chunk_size """ for i in range(0, len(list_to_chunk), max_chunk_size): yield list_to_chunk[i:i + max_chunk_size] def split_into(max_num_chunks, list_to_chunk): """ Yields the list with a max total size of max_num_chunks """ max_chunk_size = math.ceil(len(list_to_chunk) / max_num_chunks) return chunks_of(max_chunk_size, list_to_chunk) #### # File/filesystem related function #### def get_file_ext(file): file_name, file_extension = os.path.splitext(file) return file_extension def norm_path(path): """ :return: Proper path for os with vars expanded out """ # path = os.path.normcase(path) path = os.path.expanduser(path) path = os.path.expandvars(path) path = os.path.normpath(path) return path def create_hashed_path(base_path, name, depth=2): """ Create a directory structure using the hashed filename :return: string of the path to save to not including filename/ext """ if depth > 16: logger.warning("depth cannot be greater then 16, setting to 16") depth = 16 name_hash = hashlib.md5(str(name).encode('utf-8')).hexdigest() if base_path.endswith(os.path.sep): save_path = base_path else: save_path = base_path + os.path.sep for i in range(1, depth + 1): end = i * 2 start = end - 2 save_path += name_hash[start:end] + os.path.sep return {'path': save_path, 'hash': name_hash, } def create_path(path, is_dir=False): """ Check if path exists, if not create it :param path: path or file to create directory for :param is_dir: pass True if we are passing in a directory, default = False :return: os safe path from `path` """ path = norm_path(path) path_check = path if not is_dir: path_check = os.path.dirname(path) does_path_exists = os.path.exists(path_check) if does_path_exists: return path try: os.makedirs(path_check) except OSError: pass return path def dump_json(file_, data, **kwargs): json_args = {'sort_keys': True, 'indent': 4} json_args.update(**kwargs) create_path(file_) if not file_.endswith('.json'): file_ += '.json' with open(file_, 'w') as outfile: json.dump(data, outfile, **json_args) # Lets only do this once price_pattern = re.compile('(?P<low>[\d,.\s]+)(?:\D*(?P<high>[\d,.\s]+))?') def parse_price(price): found_price = {'low': None, 'high': None } price_raw = re.search(price_pattern, price) if price_raw: matched = price_raw.groupdict() found_price['low'] = matched.get('low') found_price['high'] = matched.get('high') for key, value in found_price.items(): if value is not None: value = value.strip() new_value = value.replace(',', '').replace('.', '').replace(' ', '') try: # Check if price has cents if value[-3] in [',', '.']: # Add . for cents back new_value = new_value[:-2] + '.' + new_value[-2:] except IndexError: # Price is 99 or less with no cents pass if new_value != '': found_price[key] = float(new_value) else: found_price[key] = None return found_price def get_image_dimension(url): import requests from PIL import Image # pip install pillow from io import BytesIO size = {'width': None, 'height': None, } try: if url.startswith('//'): url = "http:" + url data = requests.get(url, timeout=15).content im = Image.open(BytesIO(data)) size['width'], size['height'] = im.size except Exception: logger.debug("Error getting image size {url}".format(url=url), exc_info=True) logger.warning("Error getting image size {url}".format(url=url)) return size def crop_image(image_file, output_file=None, height=None, width=None, x=None, y=None): from PIL import Image # pip install pillow if output_file is None or height is None or width is None or x is None or y is None: logger.error("Must pass in all params: output_file, height, width, x, and y as named args") return None if width <= 0 or height <= 0: logger.warning("Width and height must be 1 or greater") raise ValueError image = Image.open(image_file) image.crop((x, y, width + x, height + y)).save(output_file) return output_file #### # Decorators #### def rate_limited(num_calls=1, every=1.0): """ Source: https://github.com/tomasbasham/ratelimit/tree/0ca5a616fa6d184fa180b9ad0b6fd0cf54c46936 Need to make a few changes that included having num_calls be a float Prevent a method from being called if it was previously called before a time widows has elapsed. Keyword Arguments: num_calls (float): Maximum method invocations within a period. Must be greater than 0. every (float): A dampening factor (in seconds). Can be any number greater than 0. Return: function: Decorated function that will forward method invocations if the time window has elapsed. """ frequency = abs(every) / float(num_calls) def decorator(func): """ Extend the behaviour of the following function, forwarding method invocations if the time window hes elapsed. Arguments: func (function): The function to decorate Returns: function: Decorated function """ # To get around issues with function local scope # and reassigning variables, we wrap the time # within a list. When updating the value we're # not reassigning `last_called`, which would not # work, but instead reassigning the value at a # particular index. last_called = [0.0] # Add thread safety lock = threading.RLock() def wrapper(*args, **kargs): """Decorator wrapper function""" with lock: elapsed = time.time() - last_called[0] left_to_wait = frequency - elapsed if left_to_wait > 0: time.sleep(left_to_wait) last_called[0] = time.time() return func(*args, **kargs) return wrapper return decorator def rate_limited_old(max_per_second): """ Source: https://gist.github.com/gregburek/1441055 """ lock = threading.Lock() min_interval = 1.0 / max_per_second def decorate(func): last_time_called = time.perf_counter() @wraps(func) def rate_limited_function(*args, **kwargs): lock.acquire() nonlocal last_time_called try: elapsed = time.perf_counter() - last_time_called left_to_wait = min_interval - elapsed if left_to_wait > 0: time.sleep(left_to_wait) return func(*args, **kwargs) finally: last_time_called = time.perf_counter() lock.release() return rate_limited_function return decorate def timeit(stat_tracker_func, name): """ Pass in a function and the name of the stat Will time the function that this is a decorator to and send the `name` as well as the value (in seconds) to `stat_tracker_func` `stat_tracker_func` can be used to either print out the data or save it """ def _timeit(func): def wrapper(*args, **kw): start_time = time.time() result = func(*args, **kw) stop_time = time.time() stat_tracker_func(name, stop_time - start_time) return result return wrapper return _timeit #### # Regex #### # Keep re.compile's outside of fn as to only create it once proxy_parts_pattern = re.compile('^(?:(?P<schema>\w+):\/\/)(?:(?P<user>.*):(?P<password>.*)@)?(?P<host>[^:]*)(?::(?P<port>\d+))?$') def get_proxy_parts(proxy): """ Take a proxy url and break it up to its parts """ proxy_parts = {'schema': None, 'user': None, 'password': None, 'host': None, 'port': None, } # Find parts results = re.match(proxy_parts_pattern, proxy) if results: matched = results.groupdict() for key in proxy_parts: proxy_parts[key] = matched.get(key) else: logger.error("Invalid proxy format `{proxy}`".format(proxy=proxy)) if proxy_parts['port'] is None: proxy_parts['port'] = '80' return proxy_parts
stephrdev/django-tapeforms
tapeforms/utils.py
join_css_class
python
def join_css_class(css_class, *additional_css_classes): css_set = set(chain.from_iterable( c.split(' ') for c in [css_class, *additional_css_classes] if c)) return ' '.join(css_set)
Returns the union of one or more CSS classes as a space-separated string. Note that the order will not be preserved.
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/utils.py#L4-L11
null
from itertools import chain
stephrdev/django-tapeforms
tapeforms/fieldsets.py
TapeformFieldset.visible_fields
python
def visible_fields(self): form_visible_fields = self.form.visible_fields() if self.render_fields: fields = self.render_fields else: fields = [field.name for field in form_visible_fields] filtered_fields = [field for field in fields if field not in self.exclude_fields] return [field for field in form_visible_fields if field.name in filtered_fields]
Returns the reduced set of visible fields to output from the form. This method respects the provided ``fields`` configuration _and_ exlcudes all fields from the ``exclude`` configuration. If no ``fields`` where provided when configuring this fieldset, all visible fields minus the excluded fields will be returned. :return: List of bound field instances or empty tuple.
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/fieldsets.py#L76-L97
null
class TapeformFieldset(TapeformLayoutMixin, object): """ Class to render a subset of a form's fields. From a template perspective, a fieldset looks quite similar to a form (and can use the same template tag to render: ``form``. """ def __init__( self, form, fields=None, exclude=None, primary=False, template=None, extra=None ): """ Initializes a fieldset instance to be used like a form in a template. Just like in ModelForm Meta, you have to provide at least a list of fields to render in this fieldset or a list of fields to exclude. If you provide both, exclusions have a higher priority. :param form: The form instance to take fields from. :param fields: A list of visible field names to include in this fieldset. :param exclude: A list of visible fields to _not_ include in this fieldset. :param primary: If the fieldset is `primary`, this fieldset is responsible for rendering the hidden fields and non field errors. :param template: You can provide an alternative layout template to use. :param extra: This argument is carried around with the fieldset and is also available in the template. Useful to pass some special arguments for rendering around (like a fieldset headline. :return: A configured fieldset instance. """ assert fields or exclude is not None, 'Please provide fields or exclude argument.' self.form = form self.render_fields = fields or () self.exclude_fields = exclude or () self.primary_fieldset = primary self.extra = extra or {} if template: self.layout_template = template def __repr__(self): return '<{cls} form={form}, primary={primary}, fields=({fields})/({exclude})>'.format( cls=self.__class__.__name__, form=repr(self.form), primary=self.primary_fieldset, fields=';'.join(self.render_fields), exclude=';'.join(self.exclude_fields), ) def hidden_fields(self): """ Returns the hidden fields of the form for rendering of the fieldset is marked as the primary fieldset. :return: List of bound field instances or empty tuple. """ return self.form.hidden_fields() if self.primary_fieldset else () def non_field_errors(self): """ Returns all non-field errors of the form for rendering of the fieldset is marked as the primary fieldset. :return: ErrorList instance with non field errors or empty ErrorList. """ return self.form.non_field_errors() if self.primary_fieldset else ErrorList()
stephrdev/django-tapeforms
tapeforms/fieldsets.py
TapeformFieldsetsMixin.get_fieldsets
python
def get_fieldsets(self, fieldsets=None): fieldsets = fieldsets or self.fieldsets if not fieldsets: raise StopIteration # Search for primary marker in at least one of the fieldset kwargs. has_primary = any(fieldset.get('primary') for fieldset in fieldsets) for fieldset_kwargs in fieldsets: fieldset_kwargs = copy.deepcopy(fieldset_kwargs) fieldset_kwargs['form'] = self if not has_primary: fieldset_kwargs['primary'] = True has_primary = True yield self.get_fieldset(**fieldset_kwargs)
This method returns a generator which yields fieldset instances. The method uses the optional fieldsets argument to generate fieldsets for. If no fieldsets argument is passed, the class property ``fieldsets`` is used. When generating the fieldsets, the method ensures that at least one fielset will be the primary fieldset which is responsible for rendering the non field errors and hidden fields. :param fieldsets: Alternative set of fieldset kwargs. If passed this set is prevered of the ``fieldsets`` property of the form. :return: generator which yields fieldset instances.
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/fieldsets.py#L132-L163
null
class TapeformFieldsetsMixin: """ Mixin to generate fieldsets based on the `fieldsets` property of a ``TapeformFieldsetsMixin`` enabled form. """ #: Default fieldset class to use when instantiating a fieldset. fieldset_class = TapeformFieldset #: List/tuple of kwargs as `dict`` to generate fieldsets for. fieldsets = None def get_fieldset_class(self, **fieldset_kwargs): """ Returns the fieldset class to use when generating the fieldset using the passed fieldset kwargs. :param fieldset_kwargs: ``dict`` with the fieldset config from ``fieldsets`` :return: Class to use when instantiating the fieldset. """ return self.fieldset_class def get_fieldset(self, **fieldset_kwargs): """ Returns a fieldset instance for the passed ``fieldset_kwargs``. :param fieldset_kwargs: ``dict`` with the fieldset config from ``fieldsets`` :return: Fieldset instance """ cls = self.get_fieldset_class(**fieldset_kwargs) return cls(**fieldset_kwargs)
stephrdev/django-tapeforms
tapeforms/templatetags/tapeforms.py
form
python
def form(context, form, **kwargs): if not isinstance(form, (forms.BaseForm, TapeformFieldset)): raise template.TemplateSyntaxError( 'Provided form should be a `Form` instance, actual type: {0}'.format( form.__class__.__name__)) return render_to_string( form.get_layout_template(kwargs.get('using', None)), form.get_layout_context(), )
The `form` template tag will render a tape-form enabled form using the template provided by `get_layout_template` method of the form using the context generated by `get_layout_context` method of the form. Usage:: {% load tapeforms %} {% form my_form %} You can override the used layout template using the keyword argument `using`:: {% load tapeforms %} {% form my_form using='other_form_layout_template.html' %} :param form: The Django form to render. :return: Rendered form (errors + hidden fields + fields) as HTML.
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/templatetags/tapeforms.py#L11-L39
null
from django import forms, template from django.template.loader import render_to_string from ..fieldsets import TapeformFieldset register = template.Library() @register.simple_tag(takes_context=True) @register.simple_tag(takes_context=True) def formfield(context, bound_field, **kwargs): """ The `formfield` template tag will render a form field of a tape-form enabled form using the template provided by `get_field_template` method of the form together with the context generated by `get_field_context` method of the form. Usage:: {% load tapeforms %} {% formfield my_form.my_field %} You can override the used field template using the keyword argument `using`:: {% load tapeforms %} {% formfield my_form.my_field using='other_field_template.html' %} :param bound_field: The `BoundField` from a Django form to render. :return: Rendered field (label + widget + other stuff) as HTML. """ if not isinstance(bound_field, forms.BoundField): raise template.TemplateSyntaxError( 'Provided field should be a `BoundField` instance, actual type: {0}'.format( bound_field.__class__.__name__)) return render_to_string( bound_field.form.get_field_template(bound_field, kwargs.get('using', None)), bound_field.form.get_field_context(bound_field), )
stephrdev/django-tapeforms
tapeforms/templatetags/tapeforms.py
formfield
python
def formfield(context, bound_field, **kwargs): if not isinstance(bound_field, forms.BoundField): raise template.TemplateSyntaxError( 'Provided field should be a `BoundField` instance, actual type: {0}'.format( bound_field.__class__.__name__)) return render_to_string( bound_field.form.get_field_template(bound_field, kwargs.get('using', None)), bound_field.form.get_field_context(bound_field), )
The `formfield` template tag will render a form field of a tape-form enabled form using the template provided by `get_field_template` method of the form together with the context generated by `get_field_context` method of the form. Usage:: {% load tapeforms %} {% formfield my_form.my_field %} You can override the used field template using the keyword argument `using`:: {% load tapeforms %} {% formfield my_form.my_field using='other_field_template.html' %} :param bound_field: The `BoundField` from a Django form to render. :return: Rendered field (label + widget + other stuff) as HTML.
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/templatetags/tapeforms.py#L43-L71
null
from django import forms, template from django.template.loader import render_to_string from ..fieldsets import TapeformFieldset register = template.Library() @register.simple_tag(takes_context=True) def form(context, form, **kwargs): """ The `form` template tag will render a tape-form enabled form using the template provided by `get_layout_template` method of the form using the context generated by `get_layout_context` method of the form. Usage:: {% load tapeforms %} {% form my_form %} You can override the used layout template using the keyword argument `using`:: {% load tapeforms %} {% form my_form using='other_form_layout_template.html' %} :param form: The Django form to render. :return: Rendered form (errors + hidden fields + fields) as HTML. """ if not isinstance(form, (forms.BaseForm, TapeformFieldset)): raise template.TemplateSyntaxError( 'Provided form should be a `Form` instance, actual type: {0}'.format( form.__class__.__name__)) return render_to_string( form.get_layout_template(kwargs.get('using', None)), form.get_layout_context(), ) @register.simple_tag(takes_context=True)
stephrdev/django-tapeforms
tapeforms/mixins.py
TapeformLayoutMixin.get_layout_template
python
def get_layout_template(self, template_name=None): if template_name: return template_name if self.layout_template: return self.layout_template return defaults.LAYOUT_DEFAULT_TEMPLATE
Returns the layout template to use when rendering the form to HTML. Preference of template selection: 1. Provided method argument `template_name` 2. Form class property `layout_template` 3. Globally defined default template from `defaults.LAYOUT_DEFAULT_TEMPLATE` :param template_name: Optional template to use instead of other configurations. :return: Template name to use when rendering the form.
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/mixins.py#L18-L37
null
class TapeformLayoutMixin: """ Mixin to render a form of fieldset as HTML. """ #: Layout template to use when rendering the form. Optional. layout_template = None def get_layout_context(self): """ Returns the context which is used when rendering the form to HTML. The generated template context will contain the following variables: * form: `Form` instance * errors: `ErrorList` instance with non field errors and hidden field errors * hidden_fields: All hidden fields to render. * visible_fields: All visible fields to render. :return: Template context for form rendering. """ errors = self.non_field_errors() for field in self.hidden_fields(): errors.extend(field.errors) return { 'form': self, 'errors': errors, 'hidden_fields': self.hidden_fields(), 'visible_fields': self.visible_fields(), } def as_tapeform(self): """ Shortcut to render the form as a "tapeform" without including the tapeforms templatetags. Behaves similar to `as_p` and `as_table`. """ return render_to_string( self.get_layout_template(), self.get_layout_context())
stephrdev/django-tapeforms
tapeforms/mixins.py
TapeformLayoutMixin.get_layout_context
python
def get_layout_context(self): errors = self.non_field_errors() for field in self.hidden_fields(): errors.extend(field.errors) return { 'form': self, 'errors': errors, 'hidden_fields': self.hidden_fields(), 'visible_fields': self.visible_fields(), }
Returns the context which is used when rendering the form to HTML. The generated template context will contain the following variables: * form: `Form` instance * errors: `ErrorList` instance with non field errors and hidden field errors * hidden_fields: All hidden fields to render. * visible_fields: All visible fields to render. :return: Template context for form rendering.
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/mixins.py#L39-L61
null
class TapeformLayoutMixin: """ Mixin to render a form of fieldset as HTML. """ #: Layout template to use when rendering the form. Optional. layout_template = None def get_layout_template(self, template_name=None): """ Returns the layout template to use when rendering the form to HTML. Preference of template selection: 1. Provided method argument `template_name` 2. Form class property `layout_template` 3. Globally defined default template from `defaults.LAYOUT_DEFAULT_TEMPLATE` :param template_name: Optional template to use instead of other configurations. :return: Template name to use when rendering the form. """ if template_name: return template_name if self.layout_template: return self.layout_template return defaults.LAYOUT_DEFAULT_TEMPLATE def as_tapeform(self): """ Shortcut to render the form as a "tapeform" without including the tapeforms templatetags. Behaves similar to `as_p` and `as_table`. """ return render_to_string( self.get_layout_template(), self.get_layout_context())
stephrdev/django-tapeforms
tapeforms/mixins.py
TapeformMixin.full_clean
python
def full_clean(self, *args, **kwargs): super().full_clean(*args, **kwargs) for field in self.errors: if field != NON_FIELD_ERRORS: self.apply_widget_invalid_options(field)
The full_clean method is hijacked to apply special treatment to invalid field inputs. For example adding extra options/classes to widgets.
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/mixins.py#L119-L127
[ "def apply_widget_invalid_options(self, field_name):\n \"\"\"\n Applies additional widget options for an invalid field.\n\n This method is called when there is some error on a field to apply\n additional options on its widget. It does the following:\n\n * Sets the aria-invalid property of the widget for accessibility.\n * Adds an invalid CSS class, which is determined by the returned value\n of `get_widget_invalid_css_class` method. If a CSS class is returned,\n it is appended to the current value of the class property of the widget.\n\n :param field_name: A field name of the form.\n \"\"\"\n field = self.fields[field_name]\n class_name = self.get_widget_invalid_css_class(field_name, field)\n\n if class_name:\n field.widget.attrs['class'] = join_css_class(\n field.widget.attrs.get('class', None), class_name)\n\n field.widget.attrs['aria-invalid'] = 'true'\n" ]
class TapeformMixin(TapeformLayoutMixin): """ Mixin to extend the forms capability to render itself as HTML output. (using the template tags provided by `tapeforms`). """ #: Field template to use when rendering a bound form-field. Optional. field_template = None #: A dictionary of form-field names and/or form-field classes to override #: the field template which is used when rendering a certain form-field. #: Optional. field_template_overrides = None #: The CSS class to apply to the form-field container element. field_container_css_class = 'form-field' #: CSS class to append to the rendered field label tag. Optional. field_label_css_class = None #: An additional CSS class to append to the rendered field label tag when #: the field has errors. Optional. field_label_invalid_css_class = None #: A dictionary of form-field names and/or widget classes to override #: the widget template which is used when rendering a certain form-field. #: Optional. widget_template_overrides = None #: CSS class to append to the widget attributes. Optional. widget_css_class = None #: An additional CSS class to append to the widget attributes when the field #: has errors. Optional. widget_invalid_css_class = None def __init__(self, *args, **kwargs): """ The init method is overwritten to apply widget templates and CSS classes. """ super().__init__(*args, **kwargs) for field_name in self.fields: self.apply_widget_options(field_name) self.apply_widget_template(field_name) self.apply_widget_css_class(field_name) def get_field_template(self, bound_field, template_name=None): """ Returns the field template to use when rendering a form field to HTML. Preference of template selection: 1. Provided method argument `template_name` 2. Template from `field_template_overrides` selected by field name 3. Template from `field_template_overrides` selected by field class 4. Form class property `field_template` 5. Globally defined default template from `defaults.LAYOUT_FIELD_TEMPLATE` :param bound_field: `BoundField` instance to select a template for. :param template_name: Optional template to use instead of other configurations. :return: Template name to use when rendering the form field. """ if template_name: return template_name templates = self.field_template_overrides or {} template_name = templates.get(bound_field.name, None) if template_name: return template_name template_name = templates.get(bound_field.field.__class__, None) if template_name: return template_name if self.field_template: return self.field_template return defaults.FIELD_DEFAULT_TEMPLATE def get_field_container_css_class(self, bound_field): """ Returns the container CSS class to use when rendering a field template. By default, returns the Form class property `field_container_css_class`. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string. """ return self.field_container_css_class or None def get_field_label_css_class(self, bound_field): """ Returns the optional label CSS class to use when rendering a field template. By default, returns the Form class property `field_label_css_class`. If the field has errors and the Form class property `field_label_invalid_css_class` is defined, its value is appended to the CSS class. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string or `None` """ class_name = self.field_label_css_class if bound_field.errors and self.field_label_invalid_css_class: class_name = join_css_class( class_name, self.field_label_invalid_css_class) return class_name or None def get_field_context(self, bound_field): """ Returns the context which is used when rendering a form field to HTML. The generated template context will contain the following variables: * form: `Form` instance * field: `BoundField` instance of the field * field_id: Field ID to use in `<label for="..">` * field_name: Name of the form field to render * errors: `ErrorList` instance with errors of the field * required: Boolean flag to signal if the field is required or not * label: The label text of the field * label_css_class: The optional label CSS class, might be `None` * help_text: Optional help text for the form field. Might be `None` * container_css_class: The CSS class for the field container. * widget_class_name: Lowercased version of the widget class name (e.g. 'textinput') * widget_input_type: `input_type` property of the widget instance, falls back to `widget_class_name` if not available. :return: Template context for field rendering. """ widget = bound_field.field.widget widget_class_name = widget.__class__.__name__.lower() # Check if we have an overwritten id in widget attrs, # if not use auto_id of bound field. field_id = widget.attrs.get('id') or bound_field.auto_id if field_id: field_id = widget.id_for_label(field_id) return { 'form': self, 'field': bound_field, 'field_id': field_id, 'field_name': bound_field.name, 'errors': bound_field.errors, 'required': bound_field.field.required, 'label': bound_field.label, 'label_css_class': self.get_field_label_css_class(bound_field), 'help_text': mark_safe(bound_field.help_text) if bound_field.help_text else None, 'container_css_class': self.get_field_container_css_class(bound_field), 'widget_class_name': widget_class_name, 'widget_input_type': getattr(widget, 'input_type', None) or widget_class_name } def apply_widget_options(self, field_name): """ Applies additional widget options like changing the input type of DateInput and TimeInput to "date" / "time" to enable Browser date pickers or other attributes/properties. """ widget = self.fields[field_name].widget if isinstance(widget, forms.DateInput): widget.input_type = 'date' if isinstance(widget, forms.TimeInput): widget.input_type = 'time' if isinstance(widget, forms.SplitDateTimeWidget): widget.widgets[0].input_type = 'date' widget.widgets[1].input_type = 'time' def apply_widget_template(self, field_name): """ Applies widget template overrides if available. The method uses the `get_widget_template` method to determine if the widget template should be exchanged. If a template is available, the template_name property of the widget instance is updated. :param field_name: A field name of the form. """ field = self.fields[field_name] template_name = self.get_widget_template(field_name, field) if template_name: field.widget.template_name = template_name def get_widget_template(self, field_name, field): """ Returns the optional widget template to use when rendering the widget for a form field. Preference of template selection: 1. Template from `widget_template_overrides` selected by field name 2. Template from `widget_template_overrides` selected by widget class By default, returns `None` which means "use Django's default widget template". :param field_name: The field name to select a widget template for. :param field: `Field` instance to return a widget template. :return: Template name to use when rendering the widget or `None` """ templates = self.widget_template_overrides or {} template_name = templates.get(field_name, None) if template_name: return template_name template_name = templates.get(field.widget.__class__, None) if template_name: return template_name return None def apply_widget_css_class(self, field_name): """ Applies CSS classes to widgets if available. The method uses the `get_widget_css_class` method to determine if the widget CSS class should be changed. If a CSS class is returned, it is appended to the current value of the class property of the widget instance. :param field_name: A field name of the form. """ field = self.fields[field_name] class_name = self.get_widget_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) def get_widget_css_class(self, field_name, field): """ Returns the optional widget CSS class to use when rendering the form's field widget. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_css_class or None def apply_widget_invalid_options(self, field_name): """ Applies additional widget options for an invalid field. This method is called when there is some error on a field to apply additional options on its widget. It does the following: * Sets the aria-invalid property of the widget for accessibility. * Adds an invalid CSS class, which is determined by the returned value of `get_widget_invalid_css_class` method. If a CSS class is returned, it is appended to the current value of the class property of the widget. :param field_name: A field name of the form. """ field = self.fields[field_name] class_name = self.get_widget_invalid_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) field.widget.attrs['aria-invalid'] = 'true' def get_widget_invalid_css_class(self, field_name, field): """ Returns the optional widget CSS class to append when rendering the form's field widget in case of error. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_invalid_css_class or None
stephrdev/django-tapeforms
tapeforms/mixins.py
TapeformMixin.get_field_template
python
def get_field_template(self, bound_field, template_name=None): if template_name: return template_name templates = self.field_template_overrides or {} template_name = templates.get(bound_field.name, None) if template_name: return template_name template_name = templates.get(bound_field.field.__class__, None) if template_name: return template_name if self.field_template: return self.field_template return defaults.FIELD_DEFAULT_TEMPLATE
Returns the field template to use when rendering a form field to HTML. Preference of template selection: 1. Provided method argument `template_name` 2. Template from `field_template_overrides` selected by field name 3. Template from `field_template_overrides` selected by field class 4. Form class property `field_template` 5. Globally defined default template from `defaults.LAYOUT_FIELD_TEMPLATE` :param bound_field: `BoundField` instance to select a template for. :param template_name: Optional template to use instead of other configurations. :return: Template name to use when rendering the form field.
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/mixins.py#L129-L161
null
class TapeformMixin(TapeformLayoutMixin): """ Mixin to extend the forms capability to render itself as HTML output. (using the template tags provided by `tapeforms`). """ #: Field template to use when rendering a bound form-field. Optional. field_template = None #: A dictionary of form-field names and/or form-field classes to override #: the field template which is used when rendering a certain form-field. #: Optional. field_template_overrides = None #: The CSS class to apply to the form-field container element. field_container_css_class = 'form-field' #: CSS class to append to the rendered field label tag. Optional. field_label_css_class = None #: An additional CSS class to append to the rendered field label tag when #: the field has errors. Optional. field_label_invalid_css_class = None #: A dictionary of form-field names and/or widget classes to override #: the widget template which is used when rendering a certain form-field. #: Optional. widget_template_overrides = None #: CSS class to append to the widget attributes. Optional. widget_css_class = None #: An additional CSS class to append to the widget attributes when the field #: has errors. Optional. widget_invalid_css_class = None def __init__(self, *args, **kwargs): """ The init method is overwritten to apply widget templates and CSS classes. """ super().__init__(*args, **kwargs) for field_name in self.fields: self.apply_widget_options(field_name) self.apply_widget_template(field_name) self.apply_widget_css_class(field_name) def full_clean(self, *args, **kwargs): """ The full_clean method is hijacked to apply special treatment to invalid field inputs. For example adding extra options/classes to widgets. """ super().full_clean(*args, **kwargs) for field in self.errors: if field != NON_FIELD_ERRORS: self.apply_widget_invalid_options(field) def get_field_container_css_class(self, bound_field): """ Returns the container CSS class to use when rendering a field template. By default, returns the Form class property `field_container_css_class`. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string. """ return self.field_container_css_class or None def get_field_label_css_class(self, bound_field): """ Returns the optional label CSS class to use when rendering a field template. By default, returns the Form class property `field_label_css_class`. If the field has errors and the Form class property `field_label_invalid_css_class` is defined, its value is appended to the CSS class. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string or `None` """ class_name = self.field_label_css_class if bound_field.errors and self.field_label_invalid_css_class: class_name = join_css_class( class_name, self.field_label_invalid_css_class) return class_name or None def get_field_context(self, bound_field): """ Returns the context which is used when rendering a form field to HTML. The generated template context will contain the following variables: * form: `Form` instance * field: `BoundField` instance of the field * field_id: Field ID to use in `<label for="..">` * field_name: Name of the form field to render * errors: `ErrorList` instance with errors of the field * required: Boolean flag to signal if the field is required or not * label: The label text of the field * label_css_class: The optional label CSS class, might be `None` * help_text: Optional help text for the form field. Might be `None` * container_css_class: The CSS class for the field container. * widget_class_name: Lowercased version of the widget class name (e.g. 'textinput') * widget_input_type: `input_type` property of the widget instance, falls back to `widget_class_name` if not available. :return: Template context for field rendering. """ widget = bound_field.field.widget widget_class_name = widget.__class__.__name__.lower() # Check if we have an overwritten id in widget attrs, # if not use auto_id of bound field. field_id = widget.attrs.get('id') or bound_field.auto_id if field_id: field_id = widget.id_for_label(field_id) return { 'form': self, 'field': bound_field, 'field_id': field_id, 'field_name': bound_field.name, 'errors': bound_field.errors, 'required': bound_field.field.required, 'label': bound_field.label, 'label_css_class': self.get_field_label_css_class(bound_field), 'help_text': mark_safe(bound_field.help_text) if bound_field.help_text else None, 'container_css_class': self.get_field_container_css_class(bound_field), 'widget_class_name': widget_class_name, 'widget_input_type': getattr(widget, 'input_type', None) or widget_class_name } def apply_widget_options(self, field_name): """ Applies additional widget options like changing the input type of DateInput and TimeInput to "date" / "time" to enable Browser date pickers or other attributes/properties. """ widget = self.fields[field_name].widget if isinstance(widget, forms.DateInput): widget.input_type = 'date' if isinstance(widget, forms.TimeInput): widget.input_type = 'time' if isinstance(widget, forms.SplitDateTimeWidget): widget.widgets[0].input_type = 'date' widget.widgets[1].input_type = 'time' def apply_widget_template(self, field_name): """ Applies widget template overrides if available. The method uses the `get_widget_template` method to determine if the widget template should be exchanged. If a template is available, the template_name property of the widget instance is updated. :param field_name: A field name of the form. """ field = self.fields[field_name] template_name = self.get_widget_template(field_name, field) if template_name: field.widget.template_name = template_name def get_widget_template(self, field_name, field): """ Returns the optional widget template to use when rendering the widget for a form field. Preference of template selection: 1. Template from `widget_template_overrides` selected by field name 2. Template from `widget_template_overrides` selected by widget class By default, returns `None` which means "use Django's default widget template". :param field_name: The field name to select a widget template for. :param field: `Field` instance to return a widget template. :return: Template name to use when rendering the widget or `None` """ templates = self.widget_template_overrides or {} template_name = templates.get(field_name, None) if template_name: return template_name template_name = templates.get(field.widget.__class__, None) if template_name: return template_name return None def apply_widget_css_class(self, field_name): """ Applies CSS classes to widgets if available. The method uses the `get_widget_css_class` method to determine if the widget CSS class should be changed. If a CSS class is returned, it is appended to the current value of the class property of the widget instance. :param field_name: A field name of the form. """ field = self.fields[field_name] class_name = self.get_widget_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) def get_widget_css_class(self, field_name, field): """ Returns the optional widget CSS class to use when rendering the form's field widget. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_css_class or None def apply_widget_invalid_options(self, field_name): """ Applies additional widget options for an invalid field. This method is called when there is some error on a field to apply additional options on its widget. It does the following: * Sets the aria-invalid property of the widget for accessibility. * Adds an invalid CSS class, which is determined by the returned value of `get_widget_invalid_css_class` method. If a CSS class is returned, it is appended to the current value of the class property of the widget. :param field_name: A field name of the form. """ field = self.fields[field_name] class_name = self.get_widget_invalid_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) field.widget.attrs['aria-invalid'] = 'true' def get_widget_invalid_css_class(self, field_name, field): """ Returns the optional widget CSS class to append when rendering the form's field widget in case of error. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_invalid_css_class or None
stephrdev/django-tapeforms
tapeforms/mixins.py
TapeformMixin.get_field_label_css_class
python
def get_field_label_css_class(self, bound_field): class_name = self.field_label_css_class if bound_field.errors and self.field_label_invalid_css_class: class_name = join_css_class( class_name, self.field_label_invalid_css_class) return class_name or None
Returns the optional label CSS class to use when rendering a field template. By default, returns the Form class property `field_label_css_class`. If the field has errors and the Form class property `field_label_invalid_css_class` is defined, its value is appended to the CSS class. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string or `None`
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/mixins.py#L174-L191
[ "def join_css_class(css_class, *additional_css_classes):\n \"\"\"\n Returns the union of one or more CSS classes as a space-separated string.\n Note that the order will not be preserved.\n \"\"\"\n css_set = set(chain.from_iterable(\n c.split(' ') for c in [css_class, *additional_css_classes] if c))\n return ' '.join(css_set)\n" ]
class TapeformMixin(TapeformLayoutMixin): """ Mixin to extend the forms capability to render itself as HTML output. (using the template tags provided by `tapeforms`). """ #: Field template to use when rendering a bound form-field. Optional. field_template = None #: A dictionary of form-field names and/or form-field classes to override #: the field template which is used when rendering a certain form-field. #: Optional. field_template_overrides = None #: The CSS class to apply to the form-field container element. field_container_css_class = 'form-field' #: CSS class to append to the rendered field label tag. Optional. field_label_css_class = None #: An additional CSS class to append to the rendered field label tag when #: the field has errors. Optional. field_label_invalid_css_class = None #: A dictionary of form-field names and/or widget classes to override #: the widget template which is used when rendering a certain form-field. #: Optional. widget_template_overrides = None #: CSS class to append to the widget attributes. Optional. widget_css_class = None #: An additional CSS class to append to the widget attributes when the field #: has errors. Optional. widget_invalid_css_class = None def __init__(self, *args, **kwargs): """ The init method is overwritten to apply widget templates and CSS classes. """ super().__init__(*args, **kwargs) for field_name in self.fields: self.apply_widget_options(field_name) self.apply_widget_template(field_name) self.apply_widget_css_class(field_name) def full_clean(self, *args, **kwargs): """ The full_clean method is hijacked to apply special treatment to invalid field inputs. For example adding extra options/classes to widgets. """ super().full_clean(*args, **kwargs) for field in self.errors: if field != NON_FIELD_ERRORS: self.apply_widget_invalid_options(field) def get_field_template(self, bound_field, template_name=None): """ Returns the field template to use when rendering a form field to HTML. Preference of template selection: 1. Provided method argument `template_name` 2. Template from `field_template_overrides` selected by field name 3. Template from `field_template_overrides` selected by field class 4. Form class property `field_template` 5. Globally defined default template from `defaults.LAYOUT_FIELD_TEMPLATE` :param bound_field: `BoundField` instance to select a template for. :param template_name: Optional template to use instead of other configurations. :return: Template name to use when rendering the form field. """ if template_name: return template_name templates = self.field_template_overrides or {} template_name = templates.get(bound_field.name, None) if template_name: return template_name template_name = templates.get(bound_field.field.__class__, None) if template_name: return template_name if self.field_template: return self.field_template return defaults.FIELD_DEFAULT_TEMPLATE def get_field_container_css_class(self, bound_field): """ Returns the container CSS class to use when rendering a field template. By default, returns the Form class property `field_container_css_class`. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string. """ return self.field_container_css_class or None def get_field_context(self, bound_field): """ Returns the context which is used when rendering a form field to HTML. The generated template context will contain the following variables: * form: `Form` instance * field: `BoundField` instance of the field * field_id: Field ID to use in `<label for="..">` * field_name: Name of the form field to render * errors: `ErrorList` instance with errors of the field * required: Boolean flag to signal if the field is required or not * label: The label text of the field * label_css_class: The optional label CSS class, might be `None` * help_text: Optional help text for the form field. Might be `None` * container_css_class: The CSS class for the field container. * widget_class_name: Lowercased version of the widget class name (e.g. 'textinput') * widget_input_type: `input_type` property of the widget instance, falls back to `widget_class_name` if not available. :return: Template context for field rendering. """ widget = bound_field.field.widget widget_class_name = widget.__class__.__name__.lower() # Check if we have an overwritten id in widget attrs, # if not use auto_id of bound field. field_id = widget.attrs.get('id') or bound_field.auto_id if field_id: field_id = widget.id_for_label(field_id) return { 'form': self, 'field': bound_field, 'field_id': field_id, 'field_name': bound_field.name, 'errors': bound_field.errors, 'required': bound_field.field.required, 'label': bound_field.label, 'label_css_class': self.get_field_label_css_class(bound_field), 'help_text': mark_safe(bound_field.help_text) if bound_field.help_text else None, 'container_css_class': self.get_field_container_css_class(bound_field), 'widget_class_name': widget_class_name, 'widget_input_type': getattr(widget, 'input_type', None) or widget_class_name } def apply_widget_options(self, field_name): """ Applies additional widget options like changing the input type of DateInput and TimeInput to "date" / "time" to enable Browser date pickers or other attributes/properties. """ widget = self.fields[field_name].widget if isinstance(widget, forms.DateInput): widget.input_type = 'date' if isinstance(widget, forms.TimeInput): widget.input_type = 'time' if isinstance(widget, forms.SplitDateTimeWidget): widget.widgets[0].input_type = 'date' widget.widgets[1].input_type = 'time' def apply_widget_template(self, field_name): """ Applies widget template overrides if available. The method uses the `get_widget_template` method to determine if the widget template should be exchanged. If a template is available, the template_name property of the widget instance is updated. :param field_name: A field name of the form. """ field = self.fields[field_name] template_name = self.get_widget_template(field_name, field) if template_name: field.widget.template_name = template_name def get_widget_template(self, field_name, field): """ Returns the optional widget template to use when rendering the widget for a form field. Preference of template selection: 1. Template from `widget_template_overrides` selected by field name 2. Template from `widget_template_overrides` selected by widget class By default, returns `None` which means "use Django's default widget template". :param field_name: The field name to select a widget template for. :param field: `Field` instance to return a widget template. :return: Template name to use when rendering the widget or `None` """ templates = self.widget_template_overrides or {} template_name = templates.get(field_name, None) if template_name: return template_name template_name = templates.get(field.widget.__class__, None) if template_name: return template_name return None def apply_widget_css_class(self, field_name): """ Applies CSS classes to widgets if available. The method uses the `get_widget_css_class` method to determine if the widget CSS class should be changed. If a CSS class is returned, it is appended to the current value of the class property of the widget instance. :param field_name: A field name of the form. """ field = self.fields[field_name] class_name = self.get_widget_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) def get_widget_css_class(self, field_name, field): """ Returns the optional widget CSS class to use when rendering the form's field widget. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_css_class or None def apply_widget_invalid_options(self, field_name): """ Applies additional widget options for an invalid field. This method is called when there is some error on a field to apply additional options on its widget. It does the following: * Sets the aria-invalid property of the widget for accessibility. * Adds an invalid CSS class, which is determined by the returned value of `get_widget_invalid_css_class` method. If a CSS class is returned, it is appended to the current value of the class property of the widget. :param field_name: A field name of the form. """ field = self.fields[field_name] class_name = self.get_widget_invalid_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) field.widget.attrs['aria-invalid'] = 'true' def get_widget_invalid_css_class(self, field_name, field): """ Returns the optional widget CSS class to append when rendering the form's field widget in case of error. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_invalid_css_class or None
stephrdev/django-tapeforms
tapeforms/mixins.py
TapeformMixin.get_field_context
python
def get_field_context(self, bound_field): widget = bound_field.field.widget widget_class_name = widget.__class__.__name__.lower() # Check if we have an overwritten id in widget attrs, # if not use auto_id of bound field. field_id = widget.attrs.get('id') or bound_field.auto_id if field_id: field_id = widget.id_for_label(field_id) return { 'form': self, 'field': bound_field, 'field_id': field_id, 'field_name': bound_field.name, 'errors': bound_field.errors, 'required': bound_field.field.required, 'label': bound_field.label, 'label_css_class': self.get_field_label_css_class(bound_field), 'help_text': mark_safe(bound_field.help_text) if bound_field.help_text else None, 'container_css_class': self.get_field_container_css_class(bound_field), 'widget_class_name': widget_class_name, 'widget_input_type': getattr(widget, 'input_type', None) or widget_class_name }
Returns the context which is used when rendering a form field to HTML. The generated template context will contain the following variables: * form: `Form` instance * field: `BoundField` instance of the field * field_id: Field ID to use in `<label for="..">` * field_name: Name of the form field to render * errors: `ErrorList` instance with errors of the field * required: Boolean flag to signal if the field is required or not * label: The label text of the field * label_css_class: The optional label CSS class, might be `None` * help_text: Optional help text for the form field. Might be `None` * container_css_class: The CSS class for the field container. * widget_class_name: Lowercased version of the widget class name (e.g. 'textinput') * widget_input_type: `input_type` property of the widget instance, falls back to `widget_class_name` if not available. :return: Template context for field rendering.
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/mixins.py#L193-L237
[ "def get_field_container_css_class(self, bound_field):\n \"\"\"\n Returns the container CSS class to use when rendering a field template.\n\n By default, returns the Form class property `field_container_css_class`.\n\n :param bound_field: `BoundField` instance to return CSS class for.\n :return: A CSS class string.\n \"\"\"\n return self.field_container_css_class or None\n", "def get_field_label_css_class(self, bound_field):\n \"\"\"\n Returns the optional label CSS class to use when rendering a field template.\n\n By default, returns the Form class property `field_label_css_class`. If the\n field has errors and the Form class property `field_label_invalid_css_class`\n is defined, its value is appended to the CSS class.\n\n :param bound_field: `BoundField` instance to return CSS class for.\n :return: A CSS class string or `None`\n \"\"\"\n class_name = self.field_label_css_class\n\n if bound_field.errors and self.field_label_invalid_css_class:\n class_name = join_css_class(\n class_name, self.field_label_invalid_css_class)\n\n return class_name or None\n" ]
class TapeformMixin(TapeformLayoutMixin): """ Mixin to extend the forms capability to render itself as HTML output. (using the template tags provided by `tapeforms`). """ #: Field template to use when rendering a bound form-field. Optional. field_template = None #: A dictionary of form-field names and/or form-field classes to override #: the field template which is used when rendering a certain form-field. #: Optional. field_template_overrides = None #: The CSS class to apply to the form-field container element. field_container_css_class = 'form-field' #: CSS class to append to the rendered field label tag. Optional. field_label_css_class = None #: An additional CSS class to append to the rendered field label tag when #: the field has errors. Optional. field_label_invalid_css_class = None #: A dictionary of form-field names and/or widget classes to override #: the widget template which is used when rendering a certain form-field. #: Optional. widget_template_overrides = None #: CSS class to append to the widget attributes. Optional. widget_css_class = None #: An additional CSS class to append to the widget attributes when the field #: has errors. Optional. widget_invalid_css_class = None def __init__(self, *args, **kwargs): """ The init method is overwritten to apply widget templates and CSS classes. """ super().__init__(*args, **kwargs) for field_name in self.fields: self.apply_widget_options(field_name) self.apply_widget_template(field_name) self.apply_widget_css_class(field_name) def full_clean(self, *args, **kwargs): """ The full_clean method is hijacked to apply special treatment to invalid field inputs. For example adding extra options/classes to widgets. """ super().full_clean(*args, **kwargs) for field in self.errors: if field != NON_FIELD_ERRORS: self.apply_widget_invalid_options(field) def get_field_template(self, bound_field, template_name=None): """ Returns the field template to use when rendering a form field to HTML. Preference of template selection: 1. Provided method argument `template_name` 2. Template from `field_template_overrides` selected by field name 3. Template from `field_template_overrides` selected by field class 4. Form class property `field_template` 5. Globally defined default template from `defaults.LAYOUT_FIELD_TEMPLATE` :param bound_field: `BoundField` instance to select a template for. :param template_name: Optional template to use instead of other configurations. :return: Template name to use when rendering the form field. """ if template_name: return template_name templates = self.field_template_overrides or {} template_name = templates.get(bound_field.name, None) if template_name: return template_name template_name = templates.get(bound_field.field.__class__, None) if template_name: return template_name if self.field_template: return self.field_template return defaults.FIELD_DEFAULT_TEMPLATE def get_field_container_css_class(self, bound_field): """ Returns the container CSS class to use when rendering a field template. By default, returns the Form class property `field_container_css_class`. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string. """ return self.field_container_css_class or None def get_field_label_css_class(self, bound_field): """ Returns the optional label CSS class to use when rendering a field template. By default, returns the Form class property `field_label_css_class`. If the field has errors and the Form class property `field_label_invalid_css_class` is defined, its value is appended to the CSS class. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string or `None` """ class_name = self.field_label_css_class if bound_field.errors and self.field_label_invalid_css_class: class_name = join_css_class( class_name, self.field_label_invalid_css_class) return class_name or None def apply_widget_options(self, field_name): """ Applies additional widget options like changing the input type of DateInput and TimeInput to "date" / "time" to enable Browser date pickers or other attributes/properties. """ widget = self.fields[field_name].widget if isinstance(widget, forms.DateInput): widget.input_type = 'date' if isinstance(widget, forms.TimeInput): widget.input_type = 'time' if isinstance(widget, forms.SplitDateTimeWidget): widget.widgets[0].input_type = 'date' widget.widgets[1].input_type = 'time' def apply_widget_template(self, field_name): """ Applies widget template overrides if available. The method uses the `get_widget_template` method to determine if the widget template should be exchanged. If a template is available, the template_name property of the widget instance is updated. :param field_name: A field name of the form. """ field = self.fields[field_name] template_name = self.get_widget_template(field_name, field) if template_name: field.widget.template_name = template_name def get_widget_template(self, field_name, field): """ Returns the optional widget template to use when rendering the widget for a form field. Preference of template selection: 1. Template from `widget_template_overrides` selected by field name 2. Template from `widget_template_overrides` selected by widget class By default, returns `None` which means "use Django's default widget template". :param field_name: The field name to select a widget template for. :param field: `Field` instance to return a widget template. :return: Template name to use when rendering the widget or `None` """ templates = self.widget_template_overrides or {} template_name = templates.get(field_name, None) if template_name: return template_name template_name = templates.get(field.widget.__class__, None) if template_name: return template_name return None def apply_widget_css_class(self, field_name): """ Applies CSS classes to widgets if available. The method uses the `get_widget_css_class` method to determine if the widget CSS class should be changed. If a CSS class is returned, it is appended to the current value of the class property of the widget instance. :param field_name: A field name of the form. """ field = self.fields[field_name] class_name = self.get_widget_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) def get_widget_css_class(self, field_name, field): """ Returns the optional widget CSS class to use when rendering the form's field widget. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_css_class or None def apply_widget_invalid_options(self, field_name): """ Applies additional widget options for an invalid field. This method is called when there is some error on a field to apply additional options on its widget. It does the following: * Sets the aria-invalid property of the widget for accessibility. * Adds an invalid CSS class, which is determined by the returned value of `get_widget_invalid_css_class` method. If a CSS class is returned, it is appended to the current value of the class property of the widget. :param field_name: A field name of the form. """ field = self.fields[field_name] class_name = self.get_widget_invalid_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) field.widget.attrs['aria-invalid'] = 'true' def get_widget_invalid_css_class(self, field_name, field): """ Returns the optional widget CSS class to append when rendering the form's field widget in case of error. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_invalid_css_class or None
stephrdev/django-tapeforms
tapeforms/mixins.py
TapeformMixin.apply_widget_options
python
def apply_widget_options(self, field_name): widget = self.fields[field_name].widget if isinstance(widget, forms.DateInput): widget.input_type = 'date' if isinstance(widget, forms.TimeInput): widget.input_type = 'time' if isinstance(widget, forms.SplitDateTimeWidget): widget.widgets[0].input_type = 'date' widget.widgets[1].input_type = 'time'
Applies additional widget options like changing the input type of DateInput and TimeInput to "date" / "time" to enable Browser date pickers or other attributes/properties.
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/mixins.py#L239-L255
null
class TapeformMixin(TapeformLayoutMixin): """ Mixin to extend the forms capability to render itself as HTML output. (using the template tags provided by `tapeforms`). """ #: Field template to use when rendering a bound form-field. Optional. field_template = None #: A dictionary of form-field names and/or form-field classes to override #: the field template which is used when rendering a certain form-field. #: Optional. field_template_overrides = None #: The CSS class to apply to the form-field container element. field_container_css_class = 'form-field' #: CSS class to append to the rendered field label tag. Optional. field_label_css_class = None #: An additional CSS class to append to the rendered field label tag when #: the field has errors. Optional. field_label_invalid_css_class = None #: A dictionary of form-field names and/or widget classes to override #: the widget template which is used when rendering a certain form-field. #: Optional. widget_template_overrides = None #: CSS class to append to the widget attributes. Optional. widget_css_class = None #: An additional CSS class to append to the widget attributes when the field #: has errors. Optional. widget_invalid_css_class = None def __init__(self, *args, **kwargs): """ The init method is overwritten to apply widget templates and CSS classes. """ super().__init__(*args, **kwargs) for field_name in self.fields: self.apply_widget_options(field_name) self.apply_widget_template(field_name) self.apply_widget_css_class(field_name) def full_clean(self, *args, **kwargs): """ The full_clean method is hijacked to apply special treatment to invalid field inputs. For example adding extra options/classes to widgets. """ super().full_clean(*args, **kwargs) for field in self.errors: if field != NON_FIELD_ERRORS: self.apply_widget_invalid_options(field) def get_field_template(self, bound_field, template_name=None): """ Returns the field template to use when rendering a form field to HTML. Preference of template selection: 1. Provided method argument `template_name` 2. Template from `field_template_overrides` selected by field name 3. Template from `field_template_overrides` selected by field class 4. Form class property `field_template` 5. Globally defined default template from `defaults.LAYOUT_FIELD_TEMPLATE` :param bound_field: `BoundField` instance to select a template for. :param template_name: Optional template to use instead of other configurations. :return: Template name to use when rendering the form field. """ if template_name: return template_name templates = self.field_template_overrides or {} template_name = templates.get(bound_field.name, None) if template_name: return template_name template_name = templates.get(bound_field.field.__class__, None) if template_name: return template_name if self.field_template: return self.field_template return defaults.FIELD_DEFAULT_TEMPLATE def get_field_container_css_class(self, bound_field): """ Returns the container CSS class to use when rendering a field template. By default, returns the Form class property `field_container_css_class`. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string. """ return self.field_container_css_class or None def get_field_label_css_class(self, bound_field): """ Returns the optional label CSS class to use when rendering a field template. By default, returns the Form class property `field_label_css_class`. If the field has errors and the Form class property `field_label_invalid_css_class` is defined, its value is appended to the CSS class. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string or `None` """ class_name = self.field_label_css_class if bound_field.errors and self.field_label_invalid_css_class: class_name = join_css_class( class_name, self.field_label_invalid_css_class) return class_name or None def get_field_context(self, bound_field): """ Returns the context which is used when rendering a form field to HTML. The generated template context will contain the following variables: * form: `Form` instance * field: `BoundField` instance of the field * field_id: Field ID to use in `<label for="..">` * field_name: Name of the form field to render * errors: `ErrorList` instance with errors of the field * required: Boolean flag to signal if the field is required or not * label: The label text of the field * label_css_class: The optional label CSS class, might be `None` * help_text: Optional help text for the form field. Might be `None` * container_css_class: The CSS class for the field container. * widget_class_name: Lowercased version of the widget class name (e.g. 'textinput') * widget_input_type: `input_type` property of the widget instance, falls back to `widget_class_name` if not available. :return: Template context for field rendering. """ widget = bound_field.field.widget widget_class_name = widget.__class__.__name__.lower() # Check if we have an overwritten id in widget attrs, # if not use auto_id of bound field. field_id = widget.attrs.get('id') or bound_field.auto_id if field_id: field_id = widget.id_for_label(field_id) return { 'form': self, 'field': bound_field, 'field_id': field_id, 'field_name': bound_field.name, 'errors': bound_field.errors, 'required': bound_field.field.required, 'label': bound_field.label, 'label_css_class': self.get_field_label_css_class(bound_field), 'help_text': mark_safe(bound_field.help_text) if bound_field.help_text else None, 'container_css_class': self.get_field_container_css_class(bound_field), 'widget_class_name': widget_class_name, 'widget_input_type': getattr(widget, 'input_type', None) or widget_class_name } def apply_widget_template(self, field_name): """ Applies widget template overrides if available. The method uses the `get_widget_template` method to determine if the widget template should be exchanged. If a template is available, the template_name property of the widget instance is updated. :param field_name: A field name of the form. """ field = self.fields[field_name] template_name = self.get_widget_template(field_name, field) if template_name: field.widget.template_name = template_name def get_widget_template(self, field_name, field): """ Returns the optional widget template to use when rendering the widget for a form field. Preference of template selection: 1. Template from `widget_template_overrides` selected by field name 2. Template from `widget_template_overrides` selected by widget class By default, returns `None` which means "use Django's default widget template". :param field_name: The field name to select a widget template for. :param field: `Field` instance to return a widget template. :return: Template name to use when rendering the widget or `None` """ templates = self.widget_template_overrides or {} template_name = templates.get(field_name, None) if template_name: return template_name template_name = templates.get(field.widget.__class__, None) if template_name: return template_name return None def apply_widget_css_class(self, field_name): """ Applies CSS classes to widgets if available. The method uses the `get_widget_css_class` method to determine if the widget CSS class should be changed. If a CSS class is returned, it is appended to the current value of the class property of the widget instance. :param field_name: A field name of the form. """ field = self.fields[field_name] class_name = self.get_widget_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) def get_widget_css_class(self, field_name, field): """ Returns the optional widget CSS class to use when rendering the form's field widget. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_css_class or None def apply_widget_invalid_options(self, field_name): """ Applies additional widget options for an invalid field. This method is called when there is some error on a field to apply additional options on its widget. It does the following: * Sets the aria-invalid property of the widget for accessibility. * Adds an invalid CSS class, which is determined by the returned value of `get_widget_invalid_css_class` method. If a CSS class is returned, it is appended to the current value of the class property of the widget. :param field_name: A field name of the form. """ field = self.fields[field_name] class_name = self.get_widget_invalid_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) field.widget.attrs['aria-invalid'] = 'true' def get_widget_invalid_css_class(self, field_name, field): """ Returns the optional widget CSS class to append when rendering the form's field widget in case of error. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_invalid_css_class or None
stephrdev/django-tapeforms
tapeforms/mixins.py
TapeformMixin.apply_widget_template
python
def apply_widget_template(self, field_name): field = self.fields[field_name] template_name = self.get_widget_template(field_name, field) if template_name: field.widget.template_name = template_name
Applies widget template overrides if available. The method uses the `get_widget_template` method to determine if the widget template should be exchanged. If a template is available, the template_name property of the widget instance is updated. :param field_name: A field name of the form.
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/mixins.py#L257-L271
[ "def get_widget_template(self, field_name, field):\n \"\"\"\n Returns the optional widget template to use when rendering the widget\n for a form field.\n\n Preference of template selection:\n 1. Template from `widget_template_overrides` selected by field name\n 2. Template from `widget_template_overrides` selected by widget class\n\n By default, returns `None` which means \"use Django's default widget template\".\n\n :param field_name: The field name to select a widget template for.\n :param field: `Field` instance to return a widget template.\n :return: Template name to use when rendering the widget or `None`\n \"\"\"\n templates = self.widget_template_overrides or {}\n\n template_name = templates.get(field_name, None)\n if template_name:\n return template_name\n\n template_name = templates.get(field.widget.__class__, None)\n if template_name:\n return template_name\n\n return None\n" ]
class TapeformMixin(TapeformLayoutMixin): """ Mixin to extend the forms capability to render itself as HTML output. (using the template tags provided by `tapeforms`). """ #: Field template to use when rendering a bound form-field. Optional. field_template = None #: A dictionary of form-field names and/or form-field classes to override #: the field template which is used when rendering a certain form-field. #: Optional. field_template_overrides = None #: The CSS class to apply to the form-field container element. field_container_css_class = 'form-field' #: CSS class to append to the rendered field label tag. Optional. field_label_css_class = None #: An additional CSS class to append to the rendered field label tag when #: the field has errors. Optional. field_label_invalid_css_class = None #: A dictionary of form-field names and/or widget classes to override #: the widget template which is used when rendering a certain form-field. #: Optional. widget_template_overrides = None #: CSS class to append to the widget attributes. Optional. widget_css_class = None #: An additional CSS class to append to the widget attributes when the field #: has errors. Optional. widget_invalid_css_class = None def __init__(self, *args, **kwargs): """ The init method is overwritten to apply widget templates and CSS classes. """ super().__init__(*args, **kwargs) for field_name in self.fields: self.apply_widget_options(field_name) self.apply_widget_template(field_name) self.apply_widget_css_class(field_name) def full_clean(self, *args, **kwargs): """ The full_clean method is hijacked to apply special treatment to invalid field inputs. For example adding extra options/classes to widgets. """ super().full_clean(*args, **kwargs) for field in self.errors: if field != NON_FIELD_ERRORS: self.apply_widget_invalid_options(field) def get_field_template(self, bound_field, template_name=None): """ Returns the field template to use when rendering a form field to HTML. Preference of template selection: 1. Provided method argument `template_name` 2. Template from `field_template_overrides` selected by field name 3. Template from `field_template_overrides` selected by field class 4. Form class property `field_template` 5. Globally defined default template from `defaults.LAYOUT_FIELD_TEMPLATE` :param bound_field: `BoundField` instance to select a template for. :param template_name: Optional template to use instead of other configurations. :return: Template name to use when rendering the form field. """ if template_name: return template_name templates = self.field_template_overrides or {} template_name = templates.get(bound_field.name, None) if template_name: return template_name template_name = templates.get(bound_field.field.__class__, None) if template_name: return template_name if self.field_template: return self.field_template return defaults.FIELD_DEFAULT_TEMPLATE def get_field_container_css_class(self, bound_field): """ Returns the container CSS class to use when rendering a field template. By default, returns the Form class property `field_container_css_class`. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string. """ return self.field_container_css_class or None def get_field_label_css_class(self, bound_field): """ Returns the optional label CSS class to use when rendering a field template. By default, returns the Form class property `field_label_css_class`. If the field has errors and the Form class property `field_label_invalid_css_class` is defined, its value is appended to the CSS class. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string or `None` """ class_name = self.field_label_css_class if bound_field.errors and self.field_label_invalid_css_class: class_name = join_css_class( class_name, self.field_label_invalid_css_class) return class_name or None def get_field_context(self, bound_field): """ Returns the context which is used when rendering a form field to HTML. The generated template context will contain the following variables: * form: `Form` instance * field: `BoundField` instance of the field * field_id: Field ID to use in `<label for="..">` * field_name: Name of the form field to render * errors: `ErrorList` instance with errors of the field * required: Boolean flag to signal if the field is required or not * label: The label text of the field * label_css_class: The optional label CSS class, might be `None` * help_text: Optional help text for the form field. Might be `None` * container_css_class: The CSS class for the field container. * widget_class_name: Lowercased version of the widget class name (e.g. 'textinput') * widget_input_type: `input_type` property of the widget instance, falls back to `widget_class_name` if not available. :return: Template context for field rendering. """ widget = bound_field.field.widget widget_class_name = widget.__class__.__name__.lower() # Check if we have an overwritten id in widget attrs, # if not use auto_id of bound field. field_id = widget.attrs.get('id') or bound_field.auto_id if field_id: field_id = widget.id_for_label(field_id) return { 'form': self, 'field': bound_field, 'field_id': field_id, 'field_name': bound_field.name, 'errors': bound_field.errors, 'required': bound_field.field.required, 'label': bound_field.label, 'label_css_class': self.get_field_label_css_class(bound_field), 'help_text': mark_safe(bound_field.help_text) if bound_field.help_text else None, 'container_css_class': self.get_field_container_css_class(bound_field), 'widget_class_name': widget_class_name, 'widget_input_type': getattr(widget, 'input_type', None) or widget_class_name } def apply_widget_options(self, field_name): """ Applies additional widget options like changing the input type of DateInput and TimeInput to "date" / "time" to enable Browser date pickers or other attributes/properties. """ widget = self.fields[field_name].widget if isinstance(widget, forms.DateInput): widget.input_type = 'date' if isinstance(widget, forms.TimeInput): widget.input_type = 'time' if isinstance(widget, forms.SplitDateTimeWidget): widget.widgets[0].input_type = 'date' widget.widgets[1].input_type = 'time' def get_widget_template(self, field_name, field): """ Returns the optional widget template to use when rendering the widget for a form field. Preference of template selection: 1. Template from `widget_template_overrides` selected by field name 2. Template from `widget_template_overrides` selected by widget class By default, returns `None` which means "use Django's default widget template". :param field_name: The field name to select a widget template for. :param field: `Field` instance to return a widget template. :return: Template name to use when rendering the widget or `None` """ templates = self.widget_template_overrides or {} template_name = templates.get(field_name, None) if template_name: return template_name template_name = templates.get(field.widget.__class__, None) if template_name: return template_name return None def apply_widget_css_class(self, field_name): """ Applies CSS classes to widgets if available. The method uses the `get_widget_css_class` method to determine if the widget CSS class should be changed. If a CSS class is returned, it is appended to the current value of the class property of the widget instance. :param field_name: A field name of the form. """ field = self.fields[field_name] class_name = self.get_widget_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) def get_widget_css_class(self, field_name, field): """ Returns the optional widget CSS class to use when rendering the form's field widget. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_css_class or None def apply_widget_invalid_options(self, field_name): """ Applies additional widget options for an invalid field. This method is called when there is some error on a field to apply additional options on its widget. It does the following: * Sets the aria-invalid property of the widget for accessibility. * Adds an invalid CSS class, which is determined by the returned value of `get_widget_invalid_css_class` method. If a CSS class is returned, it is appended to the current value of the class property of the widget. :param field_name: A field name of the form. """ field = self.fields[field_name] class_name = self.get_widget_invalid_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) field.widget.attrs['aria-invalid'] = 'true' def get_widget_invalid_css_class(self, field_name, field): """ Returns the optional widget CSS class to append when rendering the form's field widget in case of error. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_invalid_css_class or None
stephrdev/django-tapeforms
tapeforms/mixins.py
TapeformMixin.get_widget_template
python
def get_widget_template(self, field_name, field): templates = self.widget_template_overrides or {} template_name = templates.get(field_name, None) if template_name: return template_name template_name = templates.get(field.widget.__class__, None) if template_name: return template_name return None
Returns the optional widget template to use when rendering the widget for a form field. Preference of template selection: 1. Template from `widget_template_overrides` selected by field name 2. Template from `widget_template_overrides` selected by widget class By default, returns `None` which means "use Django's default widget template". :param field_name: The field name to select a widget template for. :param field: `Field` instance to return a widget template. :return: Template name to use when rendering the widget or `None`
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/mixins.py#L273-L298
null
class TapeformMixin(TapeformLayoutMixin): """ Mixin to extend the forms capability to render itself as HTML output. (using the template tags provided by `tapeforms`). """ #: Field template to use when rendering a bound form-field. Optional. field_template = None #: A dictionary of form-field names and/or form-field classes to override #: the field template which is used when rendering a certain form-field. #: Optional. field_template_overrides = None #: The CSS class to apply to the form-field container element. field_container_css_class = 'form-field' #: CSS class to append to the rendered field label tag. Optional. field_label_css_class = None #: An additional CSS class to append to the rendered field label tag when #: the field has errors. Optional. field_label_invalid_css_class = None #: A dictionary of form-field names and/or widget classes to override #: the widget template which is used when rendering a certain form-field. #: Optional. widget_template_overrides = None #: CSS class to append to the widget attributes. Optional. widget_css_class = None #: An additional CSS class to append to the widget attributes when the field #: has errors. Optional. widget_invalid_css_class = None def __init__(self, *args, **kwargs): """ The init method is overwritten to apply widget templates and CSS classes. """ super().__init__(*args, **kwargs) for field_name in self.fields: self.apply_widget_options(field_name) self.apply_widget_template(field_name) self.apply_widget_css_class(field_name) def full_clean(self, *args, **kwargs): """ The full_clean method is hijacked to apply special treatment to invalid field inputs. For example adding extra options/classes to widgets. """ super().full_clean(*args, **kwargs) for field in self.errors: if field != NON_FIELD_ERRORS: self.apply_widget_invalid_options(field) def get_field_template(self, bound_field, template_name=None): """ Returns the field template to use when rendering a form field to HTML. Preference of template selection: 1. Provided method argument `template_name` 2. Template from `field_template_overrides` selected by field name 3. Template from `field_template_overrides` selected by field class 4. Form class property `field_template` 5. Globally defined default template from `defaults.LAYOUT_FIELD_TEMPLATE` :param bound_field: `BoundField` instance to select a template for. :param template_name: Optional template to use instead of other configurations. :return: Template name to use when rendering the form field. """ if template_name: return template_name templates = self.field_template_overrides or {} template_name = templates.get(bound_field.name, None) if template_name: return template_name template_name = templates.get(bound_field.field.__class__, None) if template_name: return template_name if self.field_template: return self.field_template return defaults.FIELD_DEFAULT_TEMPLATE def get_field_container_css_class(self, bound_field): """ Returns the container CSS class to use when rendering a field template. By default, returns the Form class property `field_container_css_class`. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string. """ return self.field_container_css_class or None def get_field_label_css_class(self, bound_field): """ Returns the optional label CSS class to use when rendering a field template. By default, returns the Form class property `field_label_css_class`. If the field has errors and the Form class property `field_label_invalid_css_class` is defined, its value is appended to the CSS class. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string or `None` """ class_name = self.field_label_css_class if bound_field.errors and self.field_label_invalid_css_class: class_name = join_css_class( class_name, self.field_label_invalid_css_class) return class_name or None def get_field_context(self, bound_field): """ Returns the context which is used when rendering a form field to HTML. The generated template context will contain the following variables: * form: `Form` instance * field: `BoundField` instance of the field * field_id: Field ID to use in `<label for="..">` * field_name: Name of the form field to render * errors: `ErrorList` instance with errors of the field * required: Boolean flag to signal if the field is required or not * label: The label text of the field * label_css_class: The optional label CSS class, might be `None` * help_text: Optional help text for the form field. Might be `None` * container_css_class: The CSS class for the field container. * widget_class_name: Lowercased version of the widget class name (e.g. 'textinput') * widget_input_type: `input_type` property of the widget instance, falls back to `widget_class_name` if not available. :return: Template context for field rendering. """ widget = bound_field.field.widget widget_class_name = widget.__class__.__name__.lower() # Check if we have an overwritten id in widget attrs, # if not use auto_id of bound field. field_id = widget.attrs.get('id') or bound_field.auto_id if field_id: field_id = widget.id_for_label(field_id) return { 'form': self, 'field': bound_field, 'field_id': field_id, 'field_name': bound_field.name, 'errors': bound_field.errors, 'required': bound_field.field.required, 'label': bound_field.label, 'label_css_class': self.get_field_label_css_class(bound_field), 'help_text': mark_safe(bound_field.help_text) if bound_field.help_text else None, 'container_css_class': self.get_field_container_css_class(bound_field), 'widget_class_name': widget_class_name, 'widget_input_type': getattr(widget, 'input_type', None) or widget_class_name } def apply_widget_options(self, field_name): """ Applies additional widget options like changing the input type of DateInput and TimeInput to "date" / "time" to enable Browser date pickers or other attributes/properties. """ widget = self.fields[field_name].widget if isinstance(widget, forms.DateInput): widget.input_type = 'date' if isinstance(widget, forms.TimeInput): widget.input_type = 'time' if isinstance(widget, forms.SplitDateTimeWidget): widget.widgets[0].input_type = 'date' widget.widgets[1].input_type = 'time' def apply_widget_template(self, field_name): """ Applies widget template overrides if available. The method uses the `get_widget_template` method to determine if the widget template should be exchanged. If a template is available, the template_name property of the widget instance is updated. :param field_name: A field name of the form. """ field = self.fields[field_name] template_name = self.get_widget_template(field_name, field) if template_name: field.widget.template_name = template_name def apply_widget_css_class(self, field_name): """ Applies CSS classes to widgets if available. The method uses the `get_widget_css_class` method to determine if the widget CSS class should be changed. If a CSS class is returned, it is appended to the current value of the class property of the widget instance. :param field_name: A field name of the form. """ field = self.fields[field_name] class_name = self.get_widget_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) def get_widget_css_class(self, field_name, field): """ Returns the optional widget CSS class to use when rendering the form's field widget. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_css_class or None def apply_widget_invalid_options(self, field_name): """ Applies additional widget options for an invalid field. This method is called when there is some error on a field to apply additional options on its widget. It does the following: * Sets the aria-invalid property of the widget for accessibility. * Adds an invalid CSS class, which is determined by the returned value of `get_widget_invalid_css_class` method. If a CSS class is returned, it is appended to the current value of the class property of the widget. :param field_name: A field name of the form. """ field = self.fields[field_name] class_name = self.get_widget_invalid_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) field.widget.attrs['aria-invalid'] = 'true' def get_widget_invalid_css_class(self, field_name, field): """ Returns the optional widget CSS class to append when rendering the form's field widget in case of error. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_invalid_css_class or None
stephrdev/django-tapeforms
tapeforms/mixins.py
TapeformMixin.apply_widget_css_class
python
def apply_widget_css_class(self, field_name): field = self.fields[field_name] class_name = self.get_widget_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name)
Applies CSS classes to widgets if available. The method uses the `get_widget_css_class` method to determine if the widget CSS class should be changed. If a CSS class is returned, it is appended to the current value of the class property of the widget instance. :param field_name: A field name of the form.
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/mixins.py#L300-L315
[ "def join_css_class(css_class, *additional_css_classes):\n \"\"\"\n Returns the union of one or more CSS classes as a space-separated string.\n Note that the order will not be preserved.\n \"\"\"\n css_set = set(chain.from_iterable(\n c.split(' ') for c in [css_class, *additional_css_classes] if c))\n return ' '.join(css_set)\n", "def get_widget_css_class(self, field_name, field):\n \"\"\"\n Returns the optional widget CSS class to use when rendering the\n form's field widget.\n\n By default, returns `None` which means \"no CSS class / no change\".\n\n :param field_name: The field name of the corresponding field for the widget.\n :param field: `Field` instance to return CSS class for.\n :return: A CSS class string or `None`\n \"\"\"\n return self.widget_css_class or None\n" ]
class TapeformMixin(TapeformLayoutMixin): """ Mixin to extend the forms capability to render itself as HTML output. (using the template tags provided by `tapeforms`). """ #: Field template to use when rendering a bound form-field. Optional. field_template = None #: A dictionary of form-field names and/or form-field classes to override #: the field template which is used when rendering a certain form-field. #: Optional. field_template_overrides = None #: The CSS class to apply to the form-field container element. field_container_css_class = 'form-field' #: CSS class to append to the rendered field label tag. Optional. field_label_css_class = None #: An additional CSS class to append to the rendered field label tag when #: the field has errors. Optional. field_label_invalid_css_class = None #: A dictionary of form-field names and/or widget classes to override #: the widget template which is used when rendering a certain form-field. #: Optional. widget_template_overrides = None #: CSS class to append to the widget attributes. Optional. widget_css_class = None #: An additional CSS class to append to the widget attributes when the field #: has errors. Optional. widget_invalid_css_class = None def __init__(self, *args, **kwargs): """ The init method is overwritten to apply widget templates and CSS classes. """ super().__init__(*args, **kwargs) for field_name in self.fields: self.apply_widget_options(field_name) self.apply_widget_template(field_name) self.apply_widget_css_class(field_name) def full_clean(self, *args, **kwargs): """ The full_clean method is hijacked to apply special treatment to invalid field inputs. For example adding extra options/classes to widgets. """ super().full_clean(*args, **kwargs) for field in self.errors: if field != NON_FIELD_ERRORS: self.apply_widget_invalid_options(field) def get_field_template(self, bound_field, template_name=None): """ Returns the field template to use when rendering a form field to HTML. Preference of template selection: 1. Provided method argument `template_name` 2. Template from `field_template_overrides` selected by field name 3. Template from `field_template_overrides` selected by field class 4. Form class property `field_template` 5. Globally defined default template from `defaults.LAYOUT_FIELD_TEMPLATE` :param bound_field: `BoundField` instance to select a template for. :param template_name: Optional template to use instead of other configurations. :return: Template name to use when rendering the form field. """ if template_name: return template_name templates = self.field_template_overrides or {} template_name = templates.get(bound_field.name, None) if template_name: return template_name template_name = templates.get(bound_field.field.__class__, None) if template_name: return template_name if self.field_template: return self.field_template return defaults.FIELD_DEFAULT_TEMPLATE def get_field_container_css_class(self, bound_field): """ Returns the container CSS class to use when rendering a field template. By default, returns the Form class property `field_container_css_class`. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string. """ return self.field_container_css_class or None def get_field_label_css_class(self, bound_field): """ Returns the optional label CSS class to use when rendering a field template. By default, returns the Form class property `field_label_css_class`. If the field has errors and the Form class property `field_label_invalid_css_class` is defined, its value is appended to the CSS class. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string or `None` """ class_name = self.field_label_css_class if bound_field.errors and self.field_label_invalid_css_class: class_name = join_css_class( class_name, self.field_label_invalid_css_class) return class_name or None def get_field_context(self, bound_field): """ Returns the context which is used when rendering a form field to HTML. The generated template context will contain the following variables: * form: `Form` instance * field: `BoundField` instance of the field * field_id: Field ID to use in `<label for="..">` * field_name: Name of the form field to render * errors: `ErrorList` instance with errors of the field * required: Boolean flag to signal if the field is required or not * label: The label text of the field * label_css_class: The optional label CSS class, might be `None` * help_text: Optional help text for the form field. Might be `None` * container_css_class: The CSS class for the field container. * widget_class_name: Lowercased version of the widget class name (e.g. 'textinput') * widget_input_type: `input_type` property of the widget instance, falls back to `widget_class_name` if not available. :return: Template context for field rendering. """ widget = bound_field.field.widget widget_class_name = widget.__class__.__name__.lower() # Check if we have an overwritten id in widget attrs, # if not use auto_id of bound field. field_id = widget.attrs.get('id') or bound_field.auto_id if field_id: field_id = widget.id_for_label(field_id) return { 'form': self, 'field': bound_field, 'field_id': field_id, 'field_name': bound_field.name, 'errors': bound_field.errors, 'required': bound_field.field.required, 'label': bound_field.label, 'label_css_class': self.get_field_label_css_class(bound_field), 'help_text': mark_safe(bound_field.help_text) if bound_field.help_text else None, 'container_css_class': self.get_field_container_css_class(bound_field), 'widget_class_name': widget_class_name, 'widget_input_type': getattr(widget, 'input_type', None) or widget_class_name } def apply_widget_options(self, field_name): """ Applies additional widget options like changing the input type of DateInput and TimeInput to "date" / "time" to enable Browser date pickers or other attributes/properties. """ widget = self.fields[field_name].widget if isinstance(widget, forms.DateInput): widget.input_type = 'date' if isinstance(widget, forms.TimeInput): widget.input_type = 'time' if isinstance(widget, forms.SplitDateTimeWidget): widget.widgets[0].input_type = 'date' widget.widgets[1].input_type = 'time' def apply_widget_template(self, field_name): """ Applies widget template overrides if available. The method uses the `get_widget_template` method to determine if the widget template should be exchanged. If a template is available, the template_name property of the widget instance is updated. :param field_name: A field name of the form. """ field = self.fields[field_name] template_name = self.get_widget_template(field_name, field) if template_name: field.widget.template_name = template_name def get_widget_template(self, field_name, field): """ Returns the optional widget template to use when rendering the widget for a form field. Preference of template selection: 1. Template from `widget_template_overrides` selected by field name 2. Template from `widget_template_overrides` selected by widget class By default, returns `None` which means "use Django's default widget template". :param field_name: The field name to select a widget template for. :param field: `Field` instance to return a widget template. :return: Template name to use when rendering the widget or `None` """ templates = self.widget_template_overrides or {} template_name = templates.get(field_name, None) if template_name: return template_name template_name = templates.get(field.widget.__class__, None) if template_name: return template_name return None def get_widget_css_class(self, field_name, field): """ Returns the optional widget CSS class to use when rendering the form's field widget. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_css_class or None def apply_widget_invalid_options(self, field_name): """ Applies additional widget options for an invalid field. This method is called when there is some error on a field to apply additional options on its widget. It does the following: * Sets the aria-invalid property of the widget for accessibility. * Adds an invalid CSS class, which is determined by the returned value of `get_widget_invalid_css_class` method. If a CSS class is returned, it is appended to the current value of the class property of the widget. :param field_name: A field name of the form. """ field = self.fields[field_name] class_name = self.get_widget_invalid_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) field.widget.attrs['aria-invalid'] = 'true' def get_widget_invalid_css_class(self, field_name, field): """ Returns the optional widget CSS class to append when rendering the form's field widget in case of error. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_invalid_css_class or None
stephrdev/django-tapeforms
tapeforms/mixins.py
TapeformMixin.apply_widget_invalid_options
python
def apply_widget_invalid_options(self, field_name): field = self.fields[field_name] class_name = self.get_widget_invalid_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) field.widget.attrs['aria-invalid'] = 'true'
Applies additional widget options for an invalid field. This method is called when there is some error on a field to apply additional options on its widget. It does the following: * Sets the aria-invalid property of the widget for accessibility. * Adds an invalid CSS class, which is determined by the returned value of `get_widget_invalid_css_class` method. If a CSS class is returned, it is appended to the current value of the class property of the widget. :param field_name: A field name of the form.
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/mixins.py#L330-L351
[ "def join_css_class(css_class, *additional_css_classes):\n \"\"\"\n Returns the union of one or more CSS classes as a space-separated string.\n Note that the order will not be preserved.\n \"\"\"\n css_set = set(chain.from_iterable(\n c.split(' ') for c in [css_class, *additional_css_classes] if c))\n return ' '.join(css_set)\n", "def get_widget_invalid_css_class(self, field_name, field):\n \"\"\"\n Returns the optional widget CSS class to append when rendering the\n form's field widget in case of error.\n\n By default, returns `None` which means \"no CSS class / no change\".\n\n :param field_name: The field name of the corresponding field for the widget.\n :param field: `Field` instance to return CSS class for.\n :return: A CSS class string or `None`\n \"\"\"\n return self.widget_invalid_css_class or None\n" ]
class TapeformMixin(TapeformLayoutMixin): """ Mixin to extend the forms capability to render itself as HTML output. (using the template tags provided by `tapeforms`). """ #: Field template to use when rendering a bound form-field. Optional. field_template = None #: A dictionary of form-field names and/or form-field classes to override #: the field template which is used when rendering a certain form-field. #: Optional. field_template_overrides = None #: The CSS class to apply to the form-field container element. field_container_css_class = 'form-field' #: CSS class to append to the rendered field label tag. Optional. field_label_css_class = None #: An additional CSS class to append to the rendered field label tag when #: the field has errors. Optional. field_label_invalid_css_class = None #: A dictionary of form-field names and/or widget classes to override #: the widget template which is used when rendering a certain form-field. #: Optional. widget_template_overrides = None #: CSS class to append to the widget attributes. Optional. widget_css_class = None #: An additional CSS class to append to the widget attributes when the field #: has errors. Optional. widget_invalid_css_class = None def __init__(self, *args, **kwargs): """ The init method is overwritten to apply widget templates and CSS classes. """ super().__init__(*args, **kwargs) for field_name in self.fields: self.apply_widget_options(field_name) self.apply_widget_template(field_name) self.apply_widget_css_class(field_name) def full_clean(self, *args, **kwargs): """ The full_clean method is hijacked to apply special treatment to invalid field inputs. For example adding extra options/classes to widgets. """ super().full_clean(*args, **kwargs) for field in self.errors: if field != NON_FIELD_ERRORS: self.apply_widget_invalid_options(field) def get_field_template(self, bound_field, template_name=None): """ Returns the field template to use when rendering a form field to HTML. Preference of template selection: 1. Provided method argument `template_name` 2. Template from `field_template_overrides` selected by field name 3. Template from `field_template_overrides` selected by field class 4. Form class property `field_template` 5. Globally defined default template from `defaults.LAYOUT_FIELD_TEMPLATE` :param bound_field: `BoundField` instance to select a template for. :param template_name: Optional template to use instead of other configurations. :return: Template name to use when rendering the form field. """ if template_name: return template_name templates = self.field_template_overrides or {} template_name = templates.get(bound_field.name, None) if template_name: return template_name template_name = templates.get(bound_field.field.__class__, None) if template_name: return template_name if self.field_template: return self.field_template return defaults.FIELD_DEFAULT_TEMPLATE def get_field_container_css_class(self, bound_field): """ Returns the container CSS class to use when rendering a field template. By default, returns the Form class property `field_container_css_class`. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string. """ return self.field_container_css_class or None def get_field_label_css_class(self, bound_field): """ Returns the optional label CSS class to use when rendering a field template. By default, returns the Form class property `field_label_css_class`. If the field has errors and the Form class property `field_label_invalid_css_class` is defined, its value is appended to the CSS class. :param bound_field: `BoundField` instance to return CSS class for. :return: A CSS class string or `None` """ class_name = self.field_label_css_class if bound_field.errors and self.field_label_invalid_css_class: class_name = join_css_class( class_name, self.field_label_invalid_css_class) return class_name or None def get_field_context(self, bound_field): """ Returns the context which is used when rendering a form field to HTML. The generated template context will contain the following variables: * form: `Form` instance * field: `BoundField` instance of the field * field_id: Field ID to use in `<label for="..">` * field_name: Name of the form field to render * errors: `ErrorList` instance with errors of the field * required: Boolean flag to signal if the field is required or not * label: The label text of the field * label_css_class: The optional label CSS class, might be `None` * help_text: Optional help text for the form field. Might be `None` * container_css_class: The CSS class for the field container. * widget_class_name: Lowercased version of the widget class name (e.g. 'textinput') * widget_input_type: `input_type` property of the widget instance, falls back to `widget_class_name` if not available. :return: Template context for field rendering. """ widget = bound_field.field.widget widget_class_name = widget.__class__.__name__.lower() # Check if we have an overwritten id in widget attrs, # if not use auto_id of bound field. field_id = widget.attrs.get('id') or bound_field.auto_id if field_id: field_id = widget.id_for_label(field_id) return { 'form': self, 'field': bound_field, 'field_id': field_id, 'field_name': bound_field.name, 'errors': bound_field.errors, 'required': bound_field.field.required, 'label': bound_field.label, 'label_css_class': self.get_field_label_css_class(bound_field), 'help_text': mark_safe(bound_field.help_text) if bound_field.help_text else None, 'container_css_class': self.get_field_container_css_class(bound_field), 'widget_class_name': widget_class_name, 'widget_input_type': getattr(widget, 'input_type', None) or widget_class_name } def apply_widget_options(self, field_name): """ Applies additional widget options like changing the input type of DateInput and TimeInput to "date" / "time" to enable Browser date pickers or other attributes/properties. """ widget = self.fields[field_name].widget if isinstance(widget, forms.DateInput): widget.input_type = 'date' if isinstance(widget, forms.TimeInput): widget.input_type = 'time' if isinstance(widget, forms.SplitDateTimeWidget): widget.widgets[0].input_type = 'date' widget.widgets[1].input_type = 'time' def apply_widget_template(self, field_name): """ Applies widget template overrides if available. The method uses the `get_widget_template` method to determine if the widget template should be exchanged. If a template is available, the template_name property of the widget instance is updated. :param field_name: A field name of the form. """ field = self.fields[field_name] template_name = self.get_widget_template(field_name, field) if template_name: field.widget.template_name = template_name def get_widget_template(self, field_name, field): """ Returns the optional widget template to use when rendering the widget for a form field. Preference of template selection: 1. Template from `widget_template_overrides` selected by field name 2. Template from `widget_template_overrides` selected by widget class By default, returns `None` which means "use Django's default widget template". :param field_name: The field name to select a widget template for. :param field: `Field` instance to return a widget template. :return: Template name to use when rendering the widget or `None` """ templates = self.widget_template_overrides or {} template_name = templates.get(field_name, None) if template_name: return template_name template_name = templates.get(field.widget.__class__, None) if template_name: return template_name return None def apply_widget_css_class(self, field_name): """ Applies CSS classes to widgets if available. The method uses the `get_widget_css_class` method to determine if the widget CSS class should be changed. If a CSS class is returned, it is appended to the current value of the class property of the widget instance. :param field_name: A field name of the form. """ field = self.fields[field_name] class_name = self.get_widget_css_class(field_name, field) if class_name: field.widget.attrs['class'] = join_css_class( field.widget.attrs.get('class', None), class_name) def get_widget_css_class(self, field_name, field): """ Returns the optional widget CSS class to use when rendering the form's field widget. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_css_class or None def get_widget_invalid_css_class(self, field_name, field): """ Returns the optional widget CSS class to append when rendering the form's field widget in case of error. By default, returns `None` which means "no CSS class / no change". :param field_name: The field name of the corresponding field for the widget. :param field: `Field` instance to return CSS class for. :return: A CSS class string or `None` """ return self.widget_invalid_css_class or None
stephrdev/django-tapeforms
tapeforms/contrib/foundation.py
FoundationTapeformMixin.get_field_template
python
def get_field_template(self, bound_field, template_name=None): template_name = super().get_field_template(bound_field, template_name) if (template_name == self.field_template and isinstance(bound_field.field.widget, ( forms.RadioSelect, forms.CheckboxSelectMultiple))): return 'tapeforms/fields/foundation_fieldset.html' return template_name
Uses a special field template for widget with multiple inputs. It only applies if no other template than the default one has been defined.
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/contrib/foundation.py#L27-L39
[ "def get_field_template(self, bound_field, template_name=None):\n \"\"\"\n Returns the field template to use when rendering a form field to HTML.\n\n Preference of template selection:\n\n 1. Provided method argument `template_name`\n 2. Template from `field_template_overrides` selected by field name\n 3. Template from `field_template_overrides` selected by field class\n 4. Form class property `field_template`\n 5. Globally defined default template from `defaults.LAYOUT_FIELD_TEMPLATE`\n\n :param bound_field: `BoundField` instance to select a template for.\n :param template_name: Optional template to use instead of other configurations.\n :return: Template name to use when rendering the form field.\n \"\"\"\n if template_name:\n return template_name\n\n templates = self.field_template_overrides or {}\n\n template_name = templates.get(bound_field.name, None)\n if template_name:\n return template_name\n\n template_name = templates.get(bound_field.field.__class__, None)\n if template_name:\n return template_name\n\n if self.field_template:\n return self.field_template\n\n return defaults.FIELD_DEFAULT_TEMPLATE\n" ]
class FoundationTapeformMixin(TapeformMixin): """ Tapeform Mixin to render Foundation compatible forms. (using the template tags provided by `tapeforms`). """ #: Use a special layout template for Foundation compatible forms. layout_template = 'tapeforms/layouts/foundation.html' #: Use a special field template for Foundation compatible forms. field_template = 'tapeforms/fields/foundation.html' #: Use a special class to invalid field's label. field_label_invalid_css_class = 'is-invalid-label' #: Use a special class to invalid field's widget. widget_invalid_css_class = 'is-invalid-input' #: Widgets with multiple inputs require some extra care (don't use ul, etc.) widget_template_overrides = { forms.RadioSelect: 'tapeforms/widgets/foundation_multipleinput.html', forms.CheckboxSelectMultiple: 'tapeforms/widgets/foundation_multipleinput.html' }
stephrdev/django-tapeforms
tapeforms/contrib/bootstrap.py
BootstrapTapeformMixin.get_field_label_css_class
python
def get_field_label_css_class(self, bound_field): # If we render CheckboxInputs, Bootstrap requires a different # field label css class for checkboxes. if isinstance(bound_field.field.widget, forms.CheckboxInput): return 'form-check-label' return super().get_field_label_css_class(bound_field)
Returns 'form-check-label' if widget is CheckboxInput. For all other fields, no css class is added.
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/contrib/bootstrap.py#L43-L53
[ "def get_field_label_css_class(self, bound_field):\n \"\"\"\n Returns the optional label CSS class to use when rendering a field template.\n\n By default, returns the Form class property `field_label_css_class`. If the\n field has errors and the Form class property `field_label_invalid_css_class`\n is defined, its value is appended to the CSS class.\n\n :param bound_field: `BoundField` instance to return CSS class for.\n :return: A CSS class string or `None`\n \"\"\"\n class_name = self.field_label_css_class\n\n if bound_field.errors and self.field_label_invalid_css_class:\n class_name = join_css_class(\n class_name, self.field_label_invalid_css_class)\n\n return class_name or None\n" ]
class BootstrapTapeformMixin(TapeformMixin): """ Tapeform Mixin to render Bootstrap4 compatible forms. (using the template tags provided by `tapeforms`). """ #: Use a special layout template for Bootstrap compatible forms. layout_template = 'tapeforms/layouts/bootstrap.html' #: Use a special field template for Bootstrap compatible forms. field_template = 'tapeforms/fields/bootstrap.html' #: Bootstrap requires that the field has a css class "form-group" applied. field_container_css_class = 'form-group' #: All widgets need a css class "form-control" (expect checkboxes). widget_css_class = 'form-control' #: Use a special class to invalid field's widget. widget_invalid_css_class = 'is-invalid' #: Widgets with multiple inputs require some extra care (don't use ul, etc.) widget_template_overrides = { forms.SelectDateWidget: 'tapeforms/widgets/bootstrap_multiwidget.html', forms.SplitDateTimeWidget: 'tapeforms/widgets/bootstrap_multiwidget.html', forms.RadioSelect: 'tapeforms/widgets/bootstrap_multipleinput.html', forms.CheckboxSelectMultiple: 'tapeforms/widgets/bootstrap_multipleinput.html' } def get_field_container_css_class(self, bound_field): """ Returns 'form-check' if widget is CheckboxInput. For all other fields, return the default value from the form property ("form-group"). """ # If we render CheckboxInputs, Bootstrap requires a different # container class for checkboxes. if isinstance(bound_field.field.widget, forms.CheckboxInput): return 'form-check' return super().get_field_container_css_class(bound_field) def get_widget_css_class(self, field_name, field): """ Returns 'form-check-input' if widget is CheckboxInput or 'form-control-file' if widget is FileInput. For all other fields return the default value from the form property ("form-control"). """ # If we render CheckboxInputs, Bootstrap requires a different # widget css class for checkboxes. if isinstance(field.widget, forms.CheckboxInput): return 'form-check-input' # Idem for fileinput. if isinstance(field.widget, forms.FileInput): return 'form-control-file' return super().get_widget_css_class(field_name, field)
stephrdev/django-tapeforms
tapeforms/contrib/bootstrap.py
BootstrapTapeformMixin.get_widget_css_class
python
def get_widget_css_class(self, field_name, field): # If we render CheckboxInputs, Bootstrap requires a different # widget css class for checkboxes. if isinstance(field.widget, forms.CheckboxInput): return 'form-check-input' # Idem for fileinput. if isinstance(field.widget, forms.FileInput): return 'form-control-file' return super().get_widget_css_class(field_name, field)
Returns 'form-check-input' if widget is CheckboxInput or 'form-control-file' if widget is FileInput. For all other fields return the default value from the form property ("form-control").
train
https://github.com/stephrdev/django-tapeforms/blob/255602de43777141f18afaf30669d7bdd4f7c323/tapeforms/contrib/bootstrap.py#L55-L70
null
class BootstrapTapeformMixin(TapeformMixin): """ Tapeform Mixin to render Bootstrap4 compatible forms. (using the template tags provided by `tapeforms`). """ #: Use a special layout template for Bootstrap compatible forms. layout_template = 'tapeforms/layouts/bootstrap.html' #: Use a special field template for Bootstrap compatible forms. field_template = 'tapeforms/fields/bootstrap.html' #: Bootstrap requires that the field has a css class "form-group" applied. field_container_css_class = 'form-group' #: All widgets need a css class "form-control" (expect checkboxes). widget_css_class = 'form-control' #: Use a special class to invalid field's widget. widget_invalid_css_class = 'is-invalid' #: Widgets with multiple inputs require some extra care (don't use ul, etc.) widget_template_overrides = { forms.SelectDateWidget: 'tapeforms/widgets/bootstrap_multiwidget.html', forms.SplitDateTimeWidget: 'tapeforms/widgets/bootstrap_multiwidget.html', forms.RadioSelect: 'tapeforms/widgets/bootstrap_multipleinput.html', forms.CheckboxSelectMultiple: 'tapeforms/widgets/bootstrap_multipleinput.html' } def get_field_container_css_class(self, bound_field): """ Returns 'form-check' if widget is CheckboxInput. For all other fields, return the default value from the form property ("form-group"). """ # If we render CheckboxInputs, Bootstrap requires a different # container class for checkboxes. if isinstance(bound_field.field.widget, forms.CheckboxInput): return 'form-check' return super().get_field_container_css_class(bound_field) def get_field_label_css_class(self, bound_field): """ Returns 'form-check-label' if widget is CheckboxInput. For all other fields, no css class is added. """ # If we render CheckboxInputs, Bootstrap requires a different # field label css class for checkboxes. if isinstance(bound_field.field.widget, forms.CheckboxInput): return 'form-check-label' return super().get_field_label_css_class(bound_field)
ten10solutions/Geist
geist/backends/replay.py
geist_replay
python
def geist_replay(wrapped, instance, args, kwargs): path_parts = [] file_parts = [] if hasattr(wrapped, '__module__'): module = wrapped.__module__ module_file = sys.modules[module].__file__ root, _file = os.path.split(module_file) path_parts.append(root) _file, _ = os.path.splitext(_file) file_parts.append(_file) if hasattr(wrapped, '__objclass__'): file_parts.append(wrapped.__objclass__.__name__) elif hasattr(wrapped, '__self__'): file_parts.append(wrapped.__self__.__class__.__name__) file_parts.append(wrapped.__name__ + '.log') path_parts.append('_'.join(file_parts)) filename = os.path.join(*path_parts) if is_in_record_mode(): platform_backend = get_platform_backend() backend = RecordingBackend( source_backend=platform_backend, recording_filename=filename ) else: backend = PlaybackBackend( recording_filename=filename ) gui = GUI(backend) return wrapped(gui, *args, **kwargs)
Wraps a test of other function and injects a Geist GUI which will enable replay (set environment variable GEIST_REPLAY_MODE to 'record' to active record mode.
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/backends/replay.py#L51-L84
[ "def get_platform_backend(**kwargs):\n if sys.platform.startswith('win'):\n import geist.backends.windows\n return geist.backends.windows.GeistWindowsBackend(**kwargs)\n else:\n import geist.backends.xvfb\n return geist.backends.xvfb.GeistXvfbBackend(**kwargs)\n", "def is_in_record_mode():\n return os.environ.get(_MODE_ENV_VAR_NAME, '') == _RECORD_MODE_ENV_VAR_VALUE\n" ]
from __future__ import division, absolute_import, print_function import os import sys import json import base64 import StringIO import wrapt from PIL import Image from . import get_platform_backend from ..core import GUI from ..finders import Location, LocationList from ._common import BackendActionBuilder import numpy import logging logger = logging.getLogger(__name__) """These backends provide support for writing tests of code using Geist. The intention is for you to write unit tests which can be run in record mode on the system under test which will produce an output file which can then run the test in playback mode. Example: User the @geist_replay decorator on your test. To record set the GEIST_REPLAY_MODE=record environment variable Windows: set GEIST_REPLAY_MODE=record Linux: export GEIST_REPLAY_MODE=record To replay don't declare or set the GEIST_REPLAY_MODE to any other value. """ _MODE_ENV_VAR_NAME = 'GEIST_REPLAY_MODE' _RECORD_MODE_ENV_VAR_VALUE = 'record' def is_in_record_mode(): return os.environ.get(_MODE_ENV_VAR_NAME, '') == _RECORD_MODE_ENV_VAR_VALUE @wrapt.decorator class _ActionsTransaction(object): def __init__(self, backend): self._actions_builder = BackendActionBuilder(backend) def __enter__(self): return self._actions_builder def __exit__(self, *args): self._actions_builder.execute() return False _FORMAT_VERSION = [0, 1] class PlaybackBackend(object): def __init__(self, recording_filename='backend_recording.log'): self._record_file = open(recording_filename, 'rb') version = json.loads(next(self._record_file)) assert version['version'] == _FORMAT_VERSION def capture_locations(self): logger.debug('replay func: "capture_locations" called') try: data = json.loads(next(self._record_file)) except StopIteration: raise AssertionError('capture_locations end of replay') if data['func'] != 'capture_locations': raise AssertionError( "function mismatch recorded %r != 'capture_locations'" % (data['func'],) ) b64_locations = data['locations'] location_list = LocationList() for b64_location in b64_locations: base64_png = b64_location['base64_png'] string_file = StringIO.StringIO( base64.b64decode(base64_png) ) x, y, w, h = ( b64_location['x'], b64_location['y'], b64_location['w'], b64_location['h'] ) image = numpy.array(Image.open(string_file)) location_list.append(Location(x, y, w, h, image=image)) return location_list def actions_transaction(self): return _ActionsTransaction(self) def _json_type_coercian(self, data): return json.loads(json.dumps(data)) def __getattr__(self, name): def func(*args, **kwargs): logger.debug('replay func: %r called with args: %r, kwargs %r', name, args, kwargs) try: data = json.loads(next(self._record_file)) except StopIteration: raise AssertionError('%s end of replay' % (name,)) func_name = name recd_name = data['func'] if func_name != recd_name: raise AssertionError( 'function mismatch recorded %r != %r' % (recd_name, func_name) ) recd_args = data['args'] recd_kwargs = data['kwargs'] if self._json_type_coercian(args) != recd_args: raise AssertionError( 'args mismatch recorded %r != %r' % (recd_args, args) ) if self._json_type_coercian(kwargs) != recd_kwargs: raise AssertionError( 'kwargs mismatch recorded %r != %r' % (recd_kwargs, kwargs) ) return func class RecordingBackend(object): def __init__( self, source_backend=None, recording_filename='backend_recording.log', **kwargs ): if source_backend is None: raise ValueError('source_backend is required for %r' % (self)) self._source_backend = source_backend self._record_file = open(recording_filename, 'wb') json.dump({'version': _FORMAT_VERSION}, self._record_file) self._record_file.write('\n') self._record_file.flush() def _write_action(self, funcname, *args, **kwargs): json.dump( {'func': funcname, 'args': args, 'kwargs': kwargs}, self._record_file ) self._record_file.write('\n') self._record_file.flush() def _write_capture_locations(self, locations): b64_locations = [] for location in locations: string_file = StringIO.StringIO() Image.fromarray(location.image).save(string_file, 'png') b64_png = base64.b64encode(string_file.getvalue()) b64_locations.append({ 'base64_png': b64_png, 'x': location.x, 'y': location.y, 'w': location.w, 'h': location.h }) json.dump( {'func': 'capture_locations', 'locations': b64_locations}, self._record_file ) self._record_file.write('\n') self._record_file.flush() def capture_locations(self): locations = list(self._source_backend.capture_locations()) self._write_capture_locations(locations) return locations def actions_transaction(self): return _ActionsTransaction(self) def __getattr__(self, name): if not name.startswith('_'): def callable_action(*args, **kwargs): getattr(self._source_backend, name)(*args, **kwargs) self._write_action(name, *args, **kwargs) return callable_action
ten10solutions/Geist
geist/backends/xvfb.py
GeistXvfbBackend._find_display
python
def _find_display(self): self.display_num = 2 while os.path.isdir(XVFB_PATH % (self.display_num,)): self.display_num += 1
Find a usable display, which doesn't have an existing Xvfb file
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/backends/xvfb.py#L85-L91
null
class GeistXvfbBackend(GeistXBase): _FB_OFFSET = 3232 def __init__(self, **kwargs): self.display_num = kwargs.get('display_num', None) width = kwargs.get('width', 1280) height = kwargs.get('height', 1024) if self.display_num is None: self._find_display() display = ":%d" % (self.display_num, ) self._display_dir = XVFB_PATH % (self.display_num,) os.makedirs(self._display_dir, 0700) dev_null = open('/dev/null', 'w') self._xvfb_proc = subprocess.Popen( [ "Xvfb", display, "-screen", "0", "%dx%dx24" % (width, height), "-fbdir", self._display_dir, ], stdout=dev_null, stderr=subprocess.STDOUT ) fb_filepath = '%s/Xvfb_screen0' % (self._display_dir,) start_t = time.time() while not os.path.exists(fb_filepath): if time.time() - start_t > 10: raise Exception('Xvfb mmap file did not appear') time.sleep(1) GeistXBase.__init__(self, display=display) self._xwd_reader = XwdToNumpyReader(fb_filepath) logger.info("Started Xvfb with file in %s", self._display_dir) def capture_locations(self): image = self._xwd_reader.get_image() h, w = image.shape[:2] return LocationList([Location(0, 0, w, h, image=image)]) def close(self): GeistXBase.close(self) if hasattr(self, '_xvfb_proc'): self._xvfb_proc.kill() self._xvfb_proc.wait() shutil.rmtree(self._display_dir, ignore_errors=False) del self._xvfb_proc
ten10solutions/Geist
geist/vision.py
pad_bin_image_to_shape
python
def pad_bin_image_to_shape(image, shape): """ Padd image to size :shape: with zeros """ h, w = shape ih, iw = image.shape assert ih <= h assert iw <= w if iw < w: result = numpy.hstack((image, numpy.zeros((ih, w - iw), bool))) else: result = image if ih < h: result = numpy.vstack((result, numpy.zeros((h - ih, w), bool))) return result
Padd image to size :shape: with zeros
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/vision.py#L13-L27
null
from __future__ import division from numpy.fft import irfft2, rfft2 import numpy import itertools import operator def subimage(rect, image): x, y, w, h = rect return image[y:y + h, x:x + w] def pad_bin_image_to_shape(image, shape): """ Padd image to size :shape: with zeros """ h, w = shape ih, iw = image.shape assert ih <= h assert iw <= w if iw < w: result = numpy.hstack((image, numpy.zeros((ih, w - iw), bool))) else: result = image if ih < h: result = numpy.vstack((result, numpy.zeros((h - ih, w), bool))) return result def sum_2d_images(images): it = iter(images) result = numpy.copy(next(it)) for image in it: result += image return result OVERLAP_TABLE = { 16: (16, 8, 4, 2, 1), 15: (15, 5, 3, 1), 12: (12, 6, 4, 3, 2, 1), 10: (10, 5, 2, 1), 9: (9, 3, 1), 8: (8, 4, 2, 1), 6: (6, 3, 2, 1), 5: (5, 1), 4: (4, 2, 1), 3: (3, 1), 2: (2, 1) } # A number near float max which we don't want to get near to keep precision # If you get false matches consider reducing this number. ACCURACY_LIMIT = 2 ** (64 - 23) def best_convolution(bin_template, bin_image, tollerance=0.5, overlap_table=OVERLAP_TABLE): """ Selects and applies the best convolution method to find template in image. Returns a list of matches in (width, height, x offset, y offset) format (where the x and y offsets are from the top left corner). As the images are binary images, we can utilise the extra bit space in the float64's by cutting the image into tiles and stacking them into variable grayscale values. This allows converting a sparse binary image into a dense(r) grayscale one. """ template_sum = numpy.count_nonzero(bin_template) th, tw = bin_template.shape ih, iw = bin_image.shape if template_sum == 0 or th == 0 or tw == 0: # If we don't have a template return [] if th > ih or tw > iw: # If the template is bigger than the image return [] # How many cells can we split the image into? max_vert_cells = ih // th max_hor_cells = iw // th # Try to work out how many times we can stack the image usable_factors = {n: factors for n, factors in overlap_table.iteritems() if ((template_sum + 1) ** (n)) < ACCURACY_LIMIT} overlap_options = [(factor, n // factor) for n, factors in usable_factors.iteritems() for factor in factors if (factor <= max_vert_cells and n // factor <= max_hor_cells)] if not overlap_options: # We can't stack the image return convolution(bin_template, bin_image, tollerance=tollerance) best_overlap = min(overlap_options, key=lambda x: ((ih // x[0] + th) * (iw // x[1] + tw))) return overlapped_convolution(bin_template, bin_image, tollerance=tollerance, splits=best_overlap) def convolution(bin_template, bin_image, tollerance=0.5): expected = numpy.count_nonzero(bin_template) ih, iw = bin_image.shape th, tw = bin_template.shape # Padd image to even dimensions if ih % 2 or iw % 2: if ih % 2: ih += 1 if iw % 2: iw += 1 bin_image = pad_bin_image_to_shape(bin_image, (ih, iw)) if expected == 0: return [] # Calculate the convolution of the FFT's of the image & template convolution_freqs = rfft2(bin_image) * rfft2(bin_template[::-1, ::-1], bin_image.shape) # Reverse the FFT to find the result image convolution_image = irfft2(convolution_freqs) # At this point, the maximum point in convolution_image should be the # bottom right (why?) of the area of greatest match # The areas in the result image within expected +- tollerance are where we # saw matches found_bitmap = ((convolution_image > (expected - tollerance)) & (convolution_image < (expected + tollerance))) match_points = numpy.transpose(numpy.nonzero(found_bitmap)) # bottom right # Find the top left point from the template (remember match_point is # inside the template (hence -1) return [((fx - (tw - 1)), (fy - (th - 1))) for (fy, fx) in match_points] def overlapped_convolution(bin_template, bin_image, tollerance=0.5, splits=(4, 2)): """ As each of these images are hold only binary values, and RFFT2 works on float64 greyscale values, we can make the convolution more efficient by breaking the image up into :splits: sectons. Each one of these sections then has its greyscale value adjusted and then stacked. We then apply the convolution to this 'stack' of images, and adjust the resultant position matches. """ th, tw = bin_template.shape ih, iw = bin_image.shape hs, ws = splits h = ih // hs w = iw // ws count = numpy.count_nonzero(bin_template) assert count > 0 assert h >= th assert w >= tw yoffset = [(i * h, ((i + 1) * h) + (th - 1)) for i in range(hs)] xoffset = [(i * w, ((i + 1) * w) + (tw - 1)) for i in range(ws)] # image_stacks is Origin (x,y), array, z (height in stack) image_stacks = [((x1, y1), bin_image[y1:y2, x1:x2], float((count + 1) ** (num))) for num, (x1, x2, y1, y2) in enumerate((x1, x2, y1, y2) for (x1, x2) in xoffset for (y1, y2) in yoffset)] pad_h = max(i.shape[0] for _, i, _ in image_stacks) pad_w = max(i.shape[1] for _, i, _ in image_stacks) # rfft metrics must be an even size - why ... maths? pad_w += pad_w % 2 pad_h += pad_h % 2 overlapped_image = sum_2d_images(pad_bin_image_to_shape(i, (pad_h, pad_w)) * num for _, i, num in image_stacks) #print "Overlap splits %r, Image Size (%d,%d), #Overlapped Size (%d,%d)" % (splits,iw,ih,pad_w,pad_h) # Calculate the convolution of the FFT's of the overlapped image & template convolution_freqs = (rfft2(overlapped_image) * rfft2(bin_template[::-1, ::-1], overlapped_image.shape)) # Reverse the FFT to find the result overlapped image convolution_image = irfft2(convolution_freqs) # At this point, the maximum point in convolution_image should be the # bottom right (why?) of the area of greatest match results = set() for (x, y), _, num in image_stacks[::-1]: test = convolution_image / num filtered = ((test >= (count - tollerance)) & (test <= (count + tollerance))) match_points = numpy.transpose(numpy.nonzero(filtered)) # bottom right for (fy, fx) in match_points: if fx < (tw - 1) or fy < (th - 1): continue results.add((x + fx - (tw - 1), y + fy - (th - 1))) convolution_image %= num return list(results) def get_possible_convolution_regions(bin_template, bin_image, tollerance=0.5, rescale=10): result = [] h, w = bin_template.shape[:2] rects = set((x, y, w, h) for x, y in convolution_r3m_targets( bin_template, bin_image, tollerance)) ih, iw = bin_image.shape[:2] # areas of interest aoi = numpy.zeros((ih // rescale + 1, iw // rescale + 1), bool) for x, y, w, h in rects: aoi[y // rescale:(y + h) // rescale + 1, x // rescale:(x + w) // rescale + 1] = True bp = binary_partition_image(aoi, min_w=w // rescale, min_h=h // rescale) if bp: bp = prune_unbeneficial_partitions(aoi, bp) return result def binary_partition_to_rects(bp, image, template_w, template_h, xoffset=0, yoffset=0): h, w = image.shape[2:] if bp is None: return [(xoffset, yoffset, w, h)] pos, axis, (p1, p2) = bp if axis == 0: new_xoffset, new_yoffset = xoffset, yoffset + pos i1, i2 = image[pos:], image[:pos] else: new_xoffset, new_yoffset = xoffset + pos, yoffset i1, i2 = image[:, pos:], image[:, :pos] def prune_unbeneficial_partitions(image, bp): pos, axis, (p1, p2) = bp if axis == 0: i1, i2 = image[pos:], image[:pos] else: i1, i2 = image[:, pos:], image[:, :pos] if p1 is None: p1_result = numpy.count_nonzero(i1) == 0 else: p1_result = prune_unbeneficial_partitions(i1, p1) if p2 is None: p2_result = numpy.count_nonzero(i2) == 0 else: p2_result = prune_unbeneficial_partitions(i2, p2) if p1_result or p2_result: return [pos, axis, [None if p1_result in [True, False] else p1_result, None if p2_result in [True, False] else p2_result]] else: return None def get_partition_scores(image, min_w=1, min_h=1): """Return list of best to worst binary splits along the x and y axis. """ h, w = image.shape[:2] if w == 0 or h == 0: return [] area = h * w cnz = numpy.count_nonzero total = cnz(image) if total == 0 or area == total: return [] if h < min_h * 2: y_c = [] else: y_c = [(-abs((count / ((h - y) * w)) - ((total - count) / (y * w))), y, 0) for count, y in ((cnz(image[y:]), y) for y in range(min_h, image.shape[0] - min_h))] if w < min_w * 2: x_c = [] else: x_c = [(-abs((count / (h * (w - x))) - ((total - count) / (h * x))), x, 1) for count, x in ((cnz(image[:, x:]), x) for x in range(min_w, image.shape[1] - min_w))] return sorted(x_c + y_c) def get_best_partition(image, min_w=1, min_h=1): partitions = get_partition_scores(image, min_w=min_w, min_h=min_h) if partitions: return partitions[0][-2:] else: return None def binary_partition_image(image, min_w=1, min_h=1, depth=0, max_depth=-1): """Return a bsp of [pos, axis, [before_node, after_node]] nodes where leaf nodes == None. If max_depth < 0 this function will continue until all leaf nodes have been found, if it is >= 0 leaf nodes will be created at that depth. min_w and min_h are the minimum width or height of a partition. """ if max_depth >= 0 and depth >= max_depth: return None partition = get_best_partition(image, min_w=min_w, min_h=min_h) if partition is None: return None pos, axis = partition if axis == 0: p1 = binary_partition_image( image[pos:], min_w, min_h, depth + 1, max_depth) p2 = binary_partition_image( image[:pos], min_w, min_h, depth + 1, max_depth) elif axis == 1: p1 = binary_partition_image( image[:, pos:], min_w, min_h, depth + 1, max_depth) p2 = binary_partition_image( image[:, :pos], min_w, min_h, depth + 1, max_depth) return [pos, axis, [p1, p2]] def draw_binary_partition(image, subdiv, res_image=None, counter=None): if res_image is None: res_image = numpy.zeros(image.shape) if counter is None: counter = itertools.count(15) pos, axis, (p1, p2) = subdiv if axis == 0: s1, s2 = res_image[pos:], res_image[:pos] i1, i2 = image[pos:], image[:pos] else: s1, s2 = res_image[:, pos:], res_image[:, :pos] i1, i2 = image[:, pos:], image[:, :pos] if p1 is None: if numpy.count_nonzero(i1): s1[:] = next(counter) else: draw_binary_partition(i1, p1, s1, counter) if p2 is None: if numpy.count_nonzero(i2): s2[:] = next(counter) else: draw_binary_partition(i2, p2, s2, counter) return res_image def rescale2avg(image): sub1 = image[:-1:2, :-1:2] sub2 = image[1::2, 1::2] res = numpy.zeros(sub1.shape, numpy.uint32) res += sub1 res += sub2 res /= 2 return res.astype(numpy.uint8) def rescale2max(image): sub1 = image[:-1:2, :-1:2] sub2 = image[1::2, 1::2] if len(image.shape) == 3: max_map = grey_scale(sub1) > grey_scale(sub2) else: max_map = sub1 > sub2 inv_max_map = max_map == False res = numpy.zeros(sub1.shape, numpy.uint8) res[max_map] = sub1[max_map] res[inv_max_map] = sub2[inv_max_map] return res def rescale3avg(image): sub1 = image[:-2:3, :-2:3] sub2 = image[1:-1:3, 1:-1:3] sub3 = image[2::3, 2::3] res = numpy.zeros(sub1.shape, numpy.uint32) res += sub1 res += sub2 res += sub3 res /= 3 return res.astype(numpy.uint8) def rescale3max(image): sub1 = image[:-2:3, :-2:3] sub2 = image[1:-1:3, 1:-1:3] sub3 = image[2::3, 2::3] if len(image.shape) == 3: grey1 = grey_scale(sub1) grey2 = grey_scale(sub2) grey3 = grey_scale(sub3) else: grey1, grey2, grey3 = sub1, sub2, sub3 max_map_1 = (grey1 > grey2) & (grey1 > grey3) max_map_2 = (grey2 > grey1) & (grey2 > grey3) max_map_3 = (max_map_1 | max_map_2) == False res = numpy.zeros(sub1.shape, numpy.uint8) res[max_map_1] = sub1[max_map_1] res[max_map_2] = sub2[max_map_2] res[max_map_3] = sub3[max_map_3] return res def or_reduce_rescale3max_offset(image): return reduce(operator.or_, (rescale3max(image[y1:y2, x1:x2]) for (y1, y2), (x1, x2) in itertools.product( *[[(i, -(3 - i)) for i in range(3)]] * 2) ) ) def numpy_or_all(images): it = iter(images) result = numpy.copy(next(it)) for image in it: result |= image return result def grey_scale(image): """Converts RGB image to Greyscale :param image: input image :type image: 3 channel 3d :class:`numpy.ndarray` :rtype: :class:`numpy.ndarray` """ return image.astype(numpy.int32).sum(axis=2) // 3 def find_edges(image): base = image[:, :, numpy.newaxis].astype(numpy.int16) c = base[1:-1, 1:-1] e = base[1:-1, :-2] w = base[1:-1, 2:] s = base[:-2, 1:-1] n = base[2:, 1:-1] diffs = numpy.concatenate([c-e, c-w, c-s, c-n], axis=2).max(axis=2) diffs[diffs < 0] = 0 dh, dw = diffs.shape col = numpy.zeros((dh, 1), numpy.uint8) row = numpy.zeros((1, dw+2), numpy.uint8) return numpy.vstack( (row, numpy.hstack((col, diffs.astype(numpy.uint8), col)), row) ) def find_threshold_near_density(img, density, low=0, high=255): """Find a threshold where the fraction of pixels above the threshold is closest to density where density is (count of pixels above threshold / count of pixels). The highest threshold closest to the desired density will be returned. Use low and high to exclude undesirable thresholds. :param img: target image :type img: 2d :class:`numpy.ndarray` :param density: target density :type density: float between 0.0 and 1.0 :param low: min threshold to test :type low: ubyte :param migh: max threshold to test :type low: ubyte :rtype: ubyte """ size = numpy.size(img) densities = [] last_t = None while True: t = ((high - low) // 2) + low if t == last_t: densities.sort(key=lambda x: (abs(x[0] - density), 256 - x[1])) return densities[0][1] else: last_t = t d = numpy.count_nonzero(img > t) / size densities.append((d, t)) if d < density: high = t elif d >= density: # search away from low low = t def filter_greys_using_image(image, target): """Filter out any values in target not in image :param image: image containing values to appear in filtered image :param target: the image to filter :rtype: 2d :class:`numpy.ndarray` containing only value in image and with the same dimensions as target """ maskbase = numpy.array(range(256), dtype=numpy.uint8) mask = numpy.where(numpy.in1d(maskbase, numpy.unique(image)), maskbase, 0) return mask[target] def correlation_coefficient_normed(template, image): h, w = template.shape[:2] H, W = image.shape[:2] template_size = template.size template_distance = template - (template.sum() / template_size) corr_num = numpy.zeros((H, W), numpy.float64) corr_denum = numpy.zeros((H, W), numpy.float64) for y in xrange(H): for x in xrange(W): image_in_template_area = image[y:y + h, x:x + w] image_distance_of_template_area = ( image_in_template_area - ( image_in_template_area.sum() / template_size ) ) I_H, I_W = image_distance_of_template_area.shape[:2] sum_of_template_by_image_distance_at_xy = ( image_distance_of_template_area * template_distance[:I_H, :I_W] ).sum() corr_num[y, x] = sum_of_template_by_image_distance_at_xy corr_denum[y, x] = numpy.sqrt( (template_distance ** 2).sum() * (image_distance_of_template_area ** 2).sum() ) print y return corr_num / corr_denum
ten10solutions/Geist
geist/vision.py
best_convolution
python
def best_convolution(bin_template, bin_image, tollerance=0.5, overlap_table=OVERLAP_TABLE): """ Selects and applies the best convolution method to find template in image. Returns a list of matches in (width, height, x offset, y offset) format (where the x and y offsets are from the top left corner). As the images are binary images, we can utilise the extra bit space in the float64's by cutting the image into tiles and stacking them into variable grayscale values. This allows converting a sparse binary image into a dense(r) grayscale one. """ template_sum = numpy.count_nonzero(bin_template) th, tw = bin_template.shape ih, iw = bin_image.shape if template_sum == 0 or th == 0 or tw == 0: # If we don't have a template return [] if th > ih or tw > iw: # If the template is bigger than the image return [] # How many cells can we split the image into? max_vert_cells = ih // th max_hor_cells = iw // th # Try to work out how many times we can stack the image usable_factors = {n: factors for n, factors in overlap_table.iteritems() if ((template_sum + 1) ** (n)) < ACCURACY_LIMIT} overlap_options = [(factor, n // factor) for n, factors in usable_factors.iteritems() for factor in factors if (factor <= max_vert_cells and n // factor <= max_hor_cells)] if not overlap_options: # We can't stack the image return convolution(bin_template, bin_image, tollerance=tollerance) best_overlap = min(overlap_options, key=lambda x: ((ih // x[0] + th) * (iw // x[1] + tw))) return overlapped_convolution(bin_template, bin_image, tollerance=tollerance, splits=best_overlap)
Selects and applies the best convolution method to find template in image. Returns a list of matches in (width, height, x offset, y offset) format (where the x and y offsets are from the top left corner). As the images are binary images, we can utilise the extra bit space in the float64's by cutting the image into tiles and stacking them into variable grayscale values. This allows converting a sparse binary image into a dense(r) grayscale one.
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/vision.py#L57-L100
null
from __future__ import division from numpy.fft import irfft2, rfft2 import numpy import itertools import operator def subimage(rect, image): x, y, w, h = rect return image[y:y + h, x:x + w] def pad_bin_image_to_shape(image, shape): """ Padd image to size :shape: with zeros """ h, w = shape ih, iw = image.shape assert ih <= h assert iw <= w if iw < w: result = numpy.hstack((image, numpy.zeros((ih, w - iw), bool))) else: result = image if ih < h: result = numpy.vstack((result, numpy.zeros((h - ih, w), bool))) return result def sum_2d_images(images): it = iter(images) result = numpy.copy(next(it)) for image in it: result += image return result OVERLAP_TABLE = { 16: (16, 8, 4, 2, 1), 15: (15, 5, 3, 1), 12: (12, 6, 4, 3, 2, 1), 10: (10, 5, 2, 1), 9: (9, 3, 1), 8: (8, 4, 2, 1), 6: (6, 3, 2, 1), 5: (5, 1), 4: (4, 2, 1), 3: (3, 1), 2: (2, 1) } # A number near float max which we don't want to get near to keep precision # If you get false matches consider reducing this number. ACCURACY_LIMIT = 2 ** (64 - 23) def best_convolution(bin_template, bin_image, tollerance=0.5, overlap_table=OVERLAP_TABLE): """ Selects and applies the best convolution method to find template in image. Returns a list of matches in (width, height, x offset, y offset) format (where the x and y offsets are from the top left corner). As the images are binary images, we can utilise the extra bit space in the float64's by cutting the image into tiles and stacking them into variable grayscale values. This allows converting a sparse binary image into a dense(r) grayscale one. """ template_sum = numpy.count_nonzero(bin_template) th, tw = bin_template.shape ih, iw = bin_image.shape if template_sum == 0 or th == 0 or tw == 0: # If we don't have a template return [] if th > ih or tw > iw: # If the template is bigger than the image return [] # How many cells can we split the image into? max_vert_cells = ih // th max_hor_cells = iw // th # Try to work out how many times we can stack the image usable_factors = {n: factors for n, factors in overlap_table.iteritems() if ((template_sum + 1) ** (n)) < ACCURACY_LIMIT} overlap_options = [(factor, n // factor) for n, factors in usable_factors.iteritems() for factor in factors if (factor <= max_vert_cells and n // factor <= max_hor_cells)] if not overlap_options: # We can't stack the image return convolution(bin_template, bin_image, tollerance=tollerance) best_overlap = min(overlap_options, key=lambda x: ((ih // x[0] + th) * (iw // x[1] + tw))) return overlapped_convolution(bin_template, bin_image, tollerance=tollerance, splits=best_overlap) def convolution(bin_template, bin_image, tollerance=0.5): expected = numpy.count_nonzero(bin_template) ih, iw = bin_image.shape th, tw = bin_template.shape # Padd image to even dimensions if ih % 2 or iw % 2: if ih % 2: ih += 1 if iw % 2: iw += 1 bin_image = pad_bin_image_to_shape(bin_image, (ih, iw)) if expected == 0: return [] # Calculate the convolution of the FFT's of the image & template convolution_freqs = rfft2(bin_image) * rfft2(bin_template[::-1, ::-1], bin_image.shape) # Reverse the FFT to find the result image convolution_image = irfft2(convolution_freqs) # At this point, the maximum point in convolution_image should be the # bottom right (why?) of the area of greatest match # The areas in the result image within expected +- tollerance are where we # saw matches found_bitmap = ((convolution_image > (expected - tollerance)) & (convolution_image < (expected + tollerance))) match_points = numpy.transpose(numpy.nonzero(found_bitmap)) # bottom right # Find the top left point from the template (remember match_point is # inside the template (hence -1) return [((fx - (tw - 1)), (fy - (th - 1))) for (fy, fx) in match_points] def overlapped_convolution(bin_template, bin_image, tollerance=0.5, splits=(4, 2)): """ As each of these images are hold only binary values, and RFFT2 works on float64 greyscale values, we can make the convolution more efficient by breaking the image up into :splits: sectons. Each one of these sections then has its greyscale value adjusted and then stacked. We then apply the convolution to this 'stack' of images, and adjust the resultant position matches. """ th, tw = bin_template.shape ih, iw = bin_image.shape hs, ws = splits h = ih // hs w = iw // ws count = numpy.count_nonzero(bin_template) assert count > 0 assert h >= th assert w >= tw yoffset = [(i * h, ((i + 1) * h) + (th - 1)) for i in range(hs)] xoffset = [(i * w, ((i + 1) * w) + (tw - 1)) for i in range(ws)] # image_stacks is Origin (x,y), array, z (height in stack) image_stacks = [((x1, y1), bin_image[y1:y2, x1:x2], float((count + 1) ** (num))) for num, (x1, x2, y1, y2) in enumerate((x1, x2, y1, y2) for (x1, x2) in xoffset for (y1, y2) in yoffset)] pad_h = max(i.shape[0] for _, i, _ in image_stacks) pad_w = max(i.shape[1] for _, i, _ in image_stacks) # rfft metrics must be an even size - why ... maths? pad_w += pad_w % 2 pad_h += pad_h % 2 overlapped_image = sum_2d_images(pad_bin_image_to_shape(i, (pad_h, pad_w)) * num for _, i, num in image_stacks) #print "Overlap splits %r, Image Size (%d,%d), #Overlapped Size (%d,%d)" % (splits,iw,ih,pad_w,pad_h) # Calculate the convolution of the FFT's of the overlapped image & template convolution_freqs = (rfft2(overlapped_image) * rfft2(bin_template[::-1, ::-1], overlapped_image.shape)) # Reverse the FFT to find the result overlapped image convolution_image = irfft2(convolution_freqs) # At this point, the maximum point in convolution_image should be the # bottom right (why?) of the area of greatest match results = set() for (x, y), _, num in image_stacks[::-1]: test = convolution_image / num filtered = ((test >= (count - tollerance)) & (test <= (count + tollerance))) match_points = numpy.transpose(numpy.nonzero(filtered)) # bottom right for (fy, fx) in match_points: if fx < (tw - 1) or fy < (th - 1): continue results.add((x + fx - (tw - 1), y + fy - (th - 1))) convolution_image %= num return list(results) def get_possible_convolution_regions(bin_template, bin_image, tollerance=0.5, rescale=10): result = [] h, w = bin_template.shape[:2] rects = set((x, y, w, h) for x, y in convolution_r3m_targets( bin_template, bin_image, tollerance)) ih, iw = bin_image.shape[:2] # areas of interest aoi = numpy.zeros((ih // rescale + 1, iw // rescale + 1), bool) for x, y, w, h in rects: aoi[y // rescale:(y + h) // rescale + 1, x // rescale:(x + w) // rescale + 1] = True bp = binary_partition_image(aoi, min_w=w // rescale, min_h=h // rescale) if bp: bp = prune_unbeneficial_partitions(aoi, bp) return result def binary_partition_to_rects(bp, image, template_w, template_h, xoffset=0, yoffset=0): h, w = image.shape[2:] if bp is None: return [(xoffset, yoffset, w, h)] pos, axis, (p1, p2) = bp if axis == 0: new_xoffset, new_yoffset = xoffset, yoffset + pos i1, i2 = image[pos:], image[:pos] else: new_xoffset, new_yoffset = xoffset + pos, yoffset i1, i2 = image[:, pos:], image[:, :pos] def prune_unbeneficial_partitions(image, bp): pos, axis, (p1, p2) = bp if axis == 0: i1, i2 = image[pos:], image[:pos] else: i1, i2 = image[:, pos:], image[:, :pos] if p1 is None: p1_result = numpy.count_nonzero(i1) == 0 else: p1_result = prune_unbeneficial_partitions(i1, p1) if p2 is None: p2_result = numpy.count_nonzero(i2) == 0 else: p2_result = prune_unbeneficial_partitions(i2, p2) if p1_result or p2_result: return [pos, axis, [None if p1_result in [True, False] else p1_result, None if p2_result in [True, False] else p2_result]] else: return None def get_partition_scores(image, min_w=1, min_h=1): """Return list of best to worst binary splits along the x and y axis. """ h, w = image.shape[:2] if w == 0 or h == 0: return [] area = h * w cnz = numpy.count_nonzero total = cnz(image) if total == 0 or area == total: return [] if h < min_h * 2: y_c = [] else: y_c = [(-abs((count / ((h - y) * w)) - ((total - count) / (y * w))), y, 0) for count, y in ((cnz(image[y:]), y) for y in range(min_h, image.shape[0] - min_h))] if w < min_w * 2: x_c = [] else: x_c = [(-abs((count / (h * (w - x))) - ((total - count) / (h * x))), x, 1) for count, x in ((cnz(image[:, x:]), x) for x in range(min_w, image.shape[1] - min_w))] return sorted(x_c + y_c) def get_best_partition(image, min_w=1, min_h=1): partitions = get_partition_scores(image, min_w=min_w, min_h=min_h) if partitions: return partitions[0][-2:] else: return None def binary_partition_image(image, min_w=1, min_h=1, depth=0, max_depth=-1): """Return a bsp of [pos, axis, [before_node, after_node]] nodes where leaf nodes == None. If max_depth < 0 this function will continue until all leaf nodes have been found, if it is >= 0 leaf nodes will be created at that depth. min_w and min_h are the minimum width or height of a partition. """ if max_depth >= 0 and depth >= max_depth: return None partition = get_best_partition(image, min_w=min_w, min_h=min_h) if partition is None: return None pos, axis = partition if axis == 0: p1 = binary_partition_image( image[pos:], min_w, min_h, depth + 1, max_depth) p2 = binary_partition_image( image[:pos], min_w, min_h, depth + 1, max_depth) elif axis == 1: p1 = binary_partition_image( image[:, pos:], min_w, min_h, depth + 1, max_depth) p2 = binary_partition_image( image[:, :pos], min_w, min_h, depth + 1, max_depth) return [pos, axis, [p1, p2]] def draw_binary_partition(image, subdiv, res_image=None, counter=None): if res_image is None: res_image = numpy.zeros(image.shape) if counter is None: counter = itertools.count(15) pos, axis, (p1, p2) = subdiv if axis == 0: s1, s2 = res_image[pos:], res_image[:pos] i1, i2 = image[pos:], image[:pos] else: s1, s2 = res_image[:, pos:], res_image[:, :pos] i1, i2 = image[:, pos:], image[:, :pos] if p1 is None: if numpy.count_nonzero(i1): s1[:] = next(counter) else: draw_binary_partition(i1, p1, s1, counter) if p2 is None: if numpy.count_nonzero(i2): s2[:] = next(counter) else: draw_binary_partition(i2, p2, s2, counter) return res_image def rescale2avg(image): sub1 = image[:-1:2, :-1:2] sub2 = image[1::2, 1::2] res = numpy.zeros(sub1.shape, numpy.uint32) res += sub1 res += sub2 res /= 2 return res.astype(numpy.uint8) def rescale2max(image): sub1 = image[:-1:2, :-1:2] sub2 = image[1::2, 1::2] if len(image.shape) == 3: max_map = grey_scale(sub1) > grey_scale(sub2) else: max_map = sub1 > sub2 inv_max_map = max_map == False res = numpy.zeros(sub1.shape, numpy.uint8) res[max_map] = sub1[max_map] res[inv_max_map] = sub2[inv_max_map] return res def rescale3avg(image): sub1 = image[:-2:3, :-2:3] sub2 = image[1:-1:3, 1:-1:3] sub3 = image[2::3, 2::3] res = numpy.zeros(sub1.shape, numpy.uint32) res += sub1 res += sub2 res += sub3 res /= 3 return res.astype(numpy.uint8) def rescale3max(image): sub1 = image[:-2:3, :-2:3] sub2 = image[1:-1:3, 1:-1:3] sub3 = image[2::3, 2::3] if len(image.shape) == 3: grey1 = grey_scale(sub1) grey2 = grey_scale(sub2) grey3 = grey_scale(sub3) else: grey1, grey2, grey3 = sub1, sub2, sub3 max_map_1 = (grey1 > grey2) & (grey1 > grey3) max_map_2 = (grey2 > grey1) & (grey2 > grey3) max_map_3 = (max_map_1 | max_map_2) == False res = numpy.zeros(sub1.shape, numpy.uint8) res[max_map_1] = sub1[max_map_1] res[max_map_2] = sub2[max_map_2] res[max_map_3] = sub3[max_map_3] return res def or_reduce_rescale3max_offset(image): return reduce(operator.or_, (rescale3max(image[y1:y2, x1:x2]) for (y1, y2), (x1, x2) in itertools.product( *[[(i, -(3 - i)) for i in range(3)]] * 2) ) ) def numpy_or_all(images): it = iter(images) result = numpy.copy(next(it)) for image in it: result |= image return result def grey_scale(image): """Converts RGB image to Greyscale :param image: input image :type image: 3 channel 3d :class:`numpy.ndarray` :rtype: :class:`numpy.ndarray` """ return image.astype(numpy.int32).sum(axis=2) // 3 def find_edges(image): base = image[:, :, numpy.newaxis].astype(numpy.int16) c = base[1:-1, 1:-1] e = base[1:-1, :-2] w = base[1:-1, 2:] s = base[:-2, 1:-1] n = base[2:, 1:-1] diffs = numpy.concatenate([c-e, c-w, c-s, c-n], axis=2).max(axis=2) diffs[diffs < 0] = 0 dh, dw = diffs.shape col = numpy.zeros((dh, 1), numpy.uint8) row = numpy.zeros((1, dw+2), numpy.uint8) return numpy.vstack( (row, numpy.hstack((col, diffs.astype(numpy.uint8), col)), row) ) def find_threshold_near_density(img, density, low=0, high=255): """Find a threshold where the fraction of pixels above the threshold is closest to density where density is (count of pixels above threshold / count of pixels). The highest threshold closest to the desired density will be returned. Use low and high to exclude undesirable thresholds. :param img: target image :type img: 2d :class:`numpy.ndarray` :param density: target density :type density: float between 0.0 and 1.0 :param low: min threshold to test :type low: ubyte :param migh: max threshold to test :type low: ubyte :rtype: ubyte """ size = numpy.size(img) densities = [] last_t = None while True: t = ((high - low) // 2) + low if t == last_t: densities.sort(key=lambda x: (abs(x[0] - density), 256 - x[1])) return densities[0][1] else: last_t = t d = numpy.count_nonzero(img > t) / size densities.append((d, t)) if d < density: high = t elif d >= density: # search away from low low = t def filter_greys_using_image(image, target): """Filter out any values in target not in image :param image: image containing values to appear in filtered image :param target: the image to filter :rtype: 2d :class:`numpy.ndarray` containing only value in image and with the same dimensions as target """ maskbase = numpy.array(range(256), dtype=numpy.uint8) mask = numpy.where(numpy.in1d(maskbase, numpy.unique(image)), maskbase, 0) return mask[target] def correlation_coefficient_normed(template, image): h, w = template.shape[:2] H, W = image.shape[:2] template_size = template.size template_distance = template - (template.sum() / template_size) corr_num = numpy.zeros((H, W), numpy.float64) corr_denum = numpy.zeros((H, W), numpy.float64) for y in xrange(H): for x in xrange(W): image_in_template_area = image[y:y + h, x:x + w] image_distance_of_template_area = ( image_in_template_area - ( image_in_template_area.sum() / template_size ) ) I_H, I_W = image_distance_of_template_area.shape[:2] sum_of_template_by_image_distance_at_xy = ( image_distance_of_template_area * template_distance[:I_H, :I_W] ).sum() corr_num[y, x] = sum_of_template_by_image_distance_at_xy corr_denum[y, x] = numpy.sqrt( (template_distance ** 2).sum() * (image_distance_of_template_area ** 2).sum() ) print y return corr_num / corr_denum
ten10solutions/Geist
geist/vision.py
overlapped_convolution
python
def overlapped_convolution(bin_template, bin_image, tollerance=0.5, splits=(4, 2)): """ As each of these images are hold only binary values, and RFFT2 works on float64 greyscale values, we can make the convolution more efficient by breaking the image up into :splits: sectons. Each one of these sections then has its greyscale value adjusted and then stacked. We then apply the convolution to this 'stack' of images, and adjust the resultant position matches. """ th, tw = bin_template.shape ih, iw = bin_image.shape hs, ws = splits h = ih // hs w = iw // ws count = numpy.count_nonzero(bin_template) assert count > 0 assert h >= th assert w >= tw yoffset = [(i * h, ((i + 1) * h) + (th - 1)) for i in range(hs)] xoffset = [(i * w, ((i + 1) * w) + (tw - 1)) for i in range(ws)] # image_stacks is Origin (x,y), array, z (height in stack) image_stacks = [((x1, y1), bin_image[y1:y2, x1:x2], float((count + 1) ** (num))) for num, (x1, x2, y1, y2) in enumerate((x1, x2, y1, y2) for (x1, x2) in xoffset for (y1, y2) in yoffset)] pad_h = max(i.shape[0] for _, i, _ in image_stacks) pad_w = max(i.shape[1] for _, i, _ in image_stacks) # rfft metrics must be an even size - why ... maths? pad_w += pad_w % 2 pad_h += pad_h % 2 overlapped_image = sum_2d_images(pad_bin_image_to_shape(i, (pad_h, pad_w)) * num for _, i, num in image_stacks) #print "Overlap splits %r, Image Size (%d,%d), #Overlapped Size (%d,%d)" % (splits,iw,ih,pad_w,pad_h) # Calculate the convolution of the FFT's of the overlapped image & template convolution_freqs = (rfft2(overlapped_image) * rfft2(bin_template[::-1, ::-1], overlapped_image.shape)) # Reverse the FFT to find the result overlapped image convolution_image = irfft2(convolution_freqs) # At this point, the maximum point in convolution_image should be the # bottom right (why?) of the area of greatest match results = set() for (x, y), _, num in image_stacks[::-1]: test = convolution_image / num filtered = ((test >= (count - tollerance)) & (test <= (count + tollerance))) match_points = numpy.transpose(numpy.nonzero(filtered)) # bottom right for (fy, fx) in match_points: if fx < (tw - 1) or fy < (th - 1): continue results.add((x + fx - (tw - 1), y + fy - (th - 1))) convolution_image %= num return list(results)
As each of these images are hold only binary values, and RFFT2 works on float64 greyscale values, we can make the convolution more efficient by breaking the image up into :splits: sectons. Each one of these sections then has its greyscale value adjusted and then stacked. We then apply the convolution to this 'stack' of images, and adjust the resultant position matches.
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/vision.py#L138-L201
null
from __future__ import division from numpy.fft import irfft2, rfft2 import numpy import itertools import operator def subimage(rect, image): x, y, w, h = rect return image[y:y + h, x:x + w] def pad_bin_image_to_shape(image, shape): """ Padd image to size :shape: with zeros """ h, w = shape ih, iw = image.shape assert ih <= h assert iw <= w if iw < w: result = numpy.hstack((image, numpy.zeros((ih, w - iw), bool))) else: result = image if ih < h: result = numpy.vstack((result, numpy.zeros((h - ih, w), bool))) return result def sum_2d_images(images): it = iter(images) result = numpy.copy(next(it)) for image in it: result += image return result OVERLAP_TABLE = { 16: (16, 8, 4, 2, 1), 15: (15, 5, 3, 1), 12: (12, 6, 4, 3, 2, 1), 10: (10, 5, 2, 1), 9: (9, 3, 1), 8: (8, 4, 2, 1), 6: (6, 3, 2, 1), 5: (5, 1), 4: (4, 2, 1), 3: (3, 1), 2: (2, 1) } # A number near float max which we don't want to get near to keep precision # If you get false matches consider reducing this number. ACCURACY_LIMIT = 2 ** (64 - 23) def best_convolution(bin_template, bin_image, tollerance=0.5, overlap_table=OVERLAP_TABLE): """ Selects and applies the best convolution method to find template in image. Returns a list of matches in (width, height, x offset, y offset) format (where the x and y offsets are from the top left corner). As the images are binary images, we can utilise the extra bit space in the float64's by cutting the image into tiles and stacking them into variable grayscale values. This allows converting a sparse binary image into a dense(r) grayscale one. """ template_sum = numpy.count_nonzero(bin_template) th, tw = bin_template.shape ih, iw = bin_image.shape if template_sum == 0 or th == 0 or tw == 0: # If we don't have a template return [] if th > ih or tw > iw: # If the template is bigger than the image return [] # How many cells can we split the image into? max_vert_cells = ih // th max_hor_cells = iw // th # Try to work out how many times we can stack the image usable_factors = {n: factors for n, factors in overlap_table.iteritems() if ((template_sum + 1) ** (n)) < ACCURACY_LIMIT} overlap_options = [(factor, n // factor) for n, factors in usable_factors.iteritems() for factor in factors if (factor <= max_vert_cells and n // factor <= max_hor_cells)] if not overlap_options: # We can't stack the image return convolution(bin_template, bin_image, tollerance=tollerance) best_overlap = min(overlap_options, key=lambda x: ((ih // x[0] + th) * (iw // x[1] + tw))) return overlapped_convolution(bin_template, bin_image, tollerance=tollerance, splits=best_overlap) def convolution(bin_template, bin_image, tollerance=0.5): expected = numpy.count_nonzero(bin_template) ih, iw = bin_image.shape th, tw = bin_template.shape # Padd image to even dimensions if ih % 2 or iw % 2: if ih % 2: ih += 1 if iw % 2: iw += 1 bin_image = pad_bin_image_to_shape(bin_image, (ih, iw)) if expected == 0: return [] # Calculate the convolution of the FFT's of the image & template convolution_freqs = rfft2(bin_image) * rfft2(bin_template[::-1, ::-1], bin_image.shape) # Reverse the FFT to find the result image convolution_image = irfft2(convolution_freqs) # At this point, the maximum point in convolution_image should be the # bottom right (why?) of the area of greatest match # The areas in the result image within expected +- tollerance are where we # saw matches found_bitmap = ((convolution_image > (expected - tollerance)) & (convolution_image < (expected + tollerance))) match_points = numpy.transpose(numpy.nonzero(found_bitmap)) # bottom right # Find the top left point from the template (remember match_point is # inside the template (hence -1) return [((fx - (tw - 1)), (fy - (th - 1))) for (fy, fx) in match_points] def overlapped_convolution(bin_template, bin_image, tollerance=0.5, splits=(4, 2)): """ As each of these images are hold only binary values, and RFFT2 works on float64 greyscale values, we can make the convolution more efficient by breaking the image up into :splits: sectons. Each one of these sections then has its greyscale value adjusted and then stacked. We then apply the convolution to this 'stack' of images, and adjust the resultant position matches. """ th, tw = bin_template.shape ih, iw = bin_image.shape hs, ws = splits h = ih // hs w = iw // ws count = numpy.count_nonzero(bin_template) assert count > 0 assert h >= th assert w >= tw yoffset = [(i * h, ((i + 1) * h) + (th - 1)) for i in range(hs)] xoffset = [(i * w, ((i + 1) * w) + (tw - 1)) for i in range(ws)] # image_stacks is Origin (x,y), array, z (height in stack) image_stacks = [((x1, y1), bin_image[y1:y2, x1:x2], float((count + 1) ** (num))) for num, (x1, x2, y1, y2) in enumerate((x1, x2, y1, y2) for (x1, x2) in xoffset for (y1, y2) in yoffset)] pad_h = max(i.shape[0] for _, i, _ in image_stacks) pad_w = max(i.shape[1] for _, i, _ in image_stacks) # rfft metrics must be an even size - why ... maths? pad_w += pad_w % 2 pad_h += pad_h % 2 overlapped_image = sum_2d_images(pad_bin_image_to_shape(i, (pad_h, pad_w)) * num for _, i, num in image_stacks) #print "Overlap splits %r, Image Size (%d,%d), #Overlapped Size (%d,%d)" % (splits,iw,ih,pad_w,pad_h) # Calculate the convolution of the FFT's of the overlapped image & template convolution_freqs = (rfft2(overlapped_image) * rfft2(bin_template[::-1, ::-1], overlapped_image.shape)) # Reverse the FFT to find the result overlapped image convolution_image = irfft2(convolution_freqs) # At this point, the maximum point in convolution_image should be the # bottom right (why?) of the area of greatest match results = set() for (x, y), _, num in image_stacks[::-1]: test = convolution_image / num filtered = ((test >= (count - tollerance)) & (test <= (count + tollerance))) match_points = numpy.transpose(numpy.nonzero(filtered)) # bottom right for (fy, fx) in match_points: if fx < (tw - 1) or fy < (th - 1): continue results.add((x + fx - (tw - 1), y + fy - (th - 1))) convolution_image %= num return list(results) def get_possible_convolution_regions(bin_template, bin_image, tollerance=0.5, rescale=10): result = [] h, w = bin_template.shape[:2] rects = set((x, y, w, h) for x, y in convolution_r3m_targets( bin_template, bin_image, tollerance)) ih, iw = bin_image.shape[:2] # areas of interest aoi = numpy.zeros((ih // rescale + 1, iw // rescale + 1), bool) for x, y, w, h in rects: aoi[y // rescale:(y + h) // rescale + 1, x // rescale:(x + w) // rescale + 1] = True bp = binary_partition_image(aoi, min_w=w // rescale, min_h=h // rescale) if bp: bp = prune_unbeneficial_partitions(aoi, bp) return result def binary_partition_to_rects(bp, image, template_w, template_h, xoffset=0, yoffset=0): h, w = image.shape[2:] if bp is None: return [(xoffset, yoffset, w, h)] pos, axis, (p1, p2) = bp if axis == 0: new_xoffset, new_yoffset = xoffset, yoffset + pos i1, i2 = image[pos:], image[:pos] else: new_xoffset, new_yoffset = xoffset + pos, yoffset i1, i2 = image[:, pos:], image[:, :pos] def prune_unbeneficial_partitions(image, bp): pos, axis, (p1, p2) = bp if axis == 0: i1, i2 = image[pos:], image[:pos] else: i1, i2 = image[:, pos:], image[:, :pos] if p1 is None: p1_result = numpy.count_nonzero(i1) == 0 else: p1_result = prune_unbeneficial_partitions(i1, p1) if p2 is None: p2_result = numpy.count_nonzero(i2) == 0 else: p2_result = prune_unbeneficial_partitions(i2, p2) if p1_result or p2_result: return [pos, axis, [None if p1_result in [True, False] else p1_result, None if p2_result in [True, False] else p2_result]] else: return None def get_partition_scores(image, min_w=1, min_h=1): """Return list of best to worst binary splits along the x and y axis. """ h, w = image.shape[:2] if w == 0 or h == 0: return [] area = h * w cnz = numpy.count_nonzero total = cnz(image) if total == 0 or area == total: return [] if h < min_h * 2: y_c = [] else: y_c = [(-abs((count / ((h - y) * w)) - ((total - count) / (y * w))), y, 0) for count, y in ((cnz(image[y:]), y) for y in range(min_h, image.shape[0] - min_h))] if w < min_w * 2: x_c = [] else: x_c = [(-abs((count / (h * (w - x))) - ((total - count) / (h * x))), x, 1) for count, x in ((cnz(image[:, x:]), x) for x in range(min_w, image.shape[1] - min_w))] return sorted(x_c + y_c) def get_best_partition(image, min_w=1, min_h=1): partitions = get_partition_scores(image, min_w=min_w, min_h=min_h) if partitions: return partitions[0][-2:] else: return None def binary_partition_image(image, min_w=1, min_h=1, depth=0, max_depth=-1): """Return a bsp of [pos, axis, [before_node, after_node]] nodes where leaf nodes == None. If max_depth < 0 this function will continue until all leaf nodes have been found, if it is >= 0 leaf nodes will be created at that depth. min_w and min_h are the minimum width or height of a partition. """ if max_depth >= 0 and depth >= max_depth: return None partition = get_best_partition(image, min_w=min_w, min_h=min_h) if partition is None: return None pos, axis = partition if axis == 0: p1 = binary_partition_image( image[pos:], min_w, min_h, depth + 1, max_depth) p2 = binary_partition_image( image[:pos], min_w, min_h, depth + 1, max_depth) elif axis == 1: p1 = binary_partition_image( image[:, pos:], min_w, min_h, depth + 1, max_depth) p2 = binary_partition_image( image[:, :pos], min_w, min_h, depth + 1, max_depth) return [pos, axis, [p1, p2]] def draw_binary_partition(image, subdiv, res_image=None, counter=None): if res_image is None: res_image = numpy.zeros(image.shape) if counter is None: counter = itertools.count(15) pos, axis, (p1, p2) = subdiv if axis == 0: s1, s2 = res_image[pos:], res_image[:pos] i1, i2 = image[pos:], image[:pos] else: s1, s2 = res_image[:, pos:], res_image[:, :pos] i1, i2 = image[:, pos:], image[:, :pos] if p1 is None: if numpy.count_nonzero(i1): s1[:] = next(counter) else: draw_binary_partition(i1, p1, s1, counter) if p2 is None: if numpy.count_nonzero(i2): s2[:] = next(counter) else: draw_binary_partition(i2, p2, s2, counter) return res_image def rescale2avg(image): sub1 = image[:-1:2, :-1:2] sub2 = image[1::2, 1::2] res = numpy.zeros(sub1.shape, numpy.uint32) res += sub1 res += sub2 res /= 2 return res.astype(numpy.uint8) def rescale2max(image): sub1 = image[:-1:2, :-1:2] sub2 = image[1::2, 1::2] if len(image.shape) == 3: max_map = grey_scale(sub1) > grey_scale(sub2) else: max_map = sub1 > sub2 inv_max_map = max_map == False res = numpy.zeros(sub1.shape, numpy.uint8) res[max_map] = sub1[max_map] res[inv_max_map] = sub2[inv_max_map] return res def rescale3avg(image): sub1 = image[:-2:3, :-2:3] sub2 = image[1:-1:3, 1:-1:3] sub3 = image[2::3, 2::3] res = numpy.zeros(sub1.shape, numpy.uint32) res += sub1 res += sub2 res += sub3 res /= 3 return res.astype(numpy.uint8) def rescale3max(image): sub1 = image[:-2:3, :-2:3] sub2 = image[1:-1:3, 1:-1:3] sub3 = image[2::3, 2::3] if len(image.shape) == 3: grey1 = grey_scale(sub1) grey2 = grey_scale(sub2) grey3 = grey_scale(sub3) else: grey1, grey2, grey3 = sub1, sub2, sub3 max_map_1 = (grey1 > grey2) & (grey1 > grey3) max_map_2 = (grey2 > grey1) & (grey2 > grey3) max_map_3 = (max_map_1 | max_map_2) == False res = numpy.zeros(sub1.shape, numpy.uint8) res[max_map_1] = sub1[max_map_1] res[max_map_2] = sub2[max_map_2] res[max_map_3] = sub3[max_map_3] return res def or_reduce_rescale3max_offset(image): return reduce(operator.or_, (rescale3max(image[y1:y2, x1:x2]) for (y1, y2), (x1, x2) in itertools.product( *[[(i, -(3 - i)) for i in range(3)]] * 2) ) ) def numpy_or_all(images): it = iter(images) result = numpy.copy(next(it)) for image in it: result |= image return result def grey_scale(image): """Converts RGB image to Greyscale :param image: input image :type image: 3 channel 3d :class:`numpy.ndarray` :rtype: :class:`numpy.ndarray` """ return image.astype(numpy.int32).sum(axis=2) // 3 def find_edges(image): base = image[:, :, numpy.newaxis].astype(numpy.int16) c = base[1:-1, 1:-1] e = base[1:-1, :-2] w = base[1:-1, 2:] s = base[:-2, 1:-1] n = base[2:, 1:-1] diffs = numpy.concatenate([c-e, c-w, c-s, c-n], axis=2).max(axis=2) diffs[diffs < 0] = 0 dh, dw = diffs.shape col = numpy.zeros((dh, 1), numpy.uint8) row = numpy.zeros((1, dw+2), numpy.uint8) return numpy.vstack( (row, numpy.hstack((col, diffs.astype(numpy.uint8), col)), row) ) def find_threshold_near_density(img, density, low=0, high=255): """Find a threshold where the fraction of pixels above the threshold is closest to density where density is (count of pixels above threshold / count of pixels). The highest threshold closest to the desired density will be returned. Use low and high to exclude undesirable thresholds. :param img: target image :type img: 2d :class:`numpy.ndarray` :param density: target density :type density: float between 0.0 and 1.0 :param low: min threshold to test :type low: ubyte :param migh: max threshold to test :type low: ubyte :rtype: ubyte """ size = numpy.size(img) densities = [] last_t = None while True: t = ((high - low) // 2) + low if t == last_t: densities.sort(key=lambda x: (abs(x[0] - density), 256 - x[1])) return densities[0][1] else: last_t = t d = numpy.count_nonzero(img > t) / size densities.append((d, t)) if d < density: high = t elif d >= density: # search away from low low = t def filter_greys_using_image(image, target): """Filter out any values in target not in image :param image: image containing values to appear in filtered image :param target: the image to filter :rtype: 2d :class:`numpy.ndarray` containing only value in image and with the same dimensions as target """ maskbase = numpy.array(range(256), dtype=numpy.uint8) mask = numpy.where(numpy.in1d(maskbase, numpy.unique(image)), maskbase, 0) return mask[target] def correlation_coefficient_normed(template, image): h, w = template.shape[:2] H, W = image.shape[:2] template_size = template.size template_distance = template - (template.sum() / template_size) corr_num = numpy.zeros((H, W), numpy.float64) corr_denum = numpy.zeros((H, W), numpy.float64) for y in xrange(H): for x in xrange(W): image_in_template_area = image[y:y + h, x:x + w] image_distance_of_template_area = ( image_in_template_area - ( image_in_template_area.sum() / template_size ) ) I_H, I_W = image_distance_of_template_area.shape[:2] sum_of_template_by_image_distance_at_xy = ( image_distance_of_template_area * template_distance[:I_H, :I_W] ).sum() corr_num[y, x] = sum_of_template_by_image_distance_at_xy corr_denum[y, x] = numpy.sqrt( (template_distance ** 2).sum() * (image_distance_of_template_area ** 2).sum() ) print y return corr_num / corr_denum
ten10solutions/Geist
geist/vision.py
get_partition_scores
python
def get_partition_scores(image, min_w=1, min_h=1): """Return list of best to worst binary splits along the x and y axis. """ h, w = image.shape[:2] if w == 0 or h == 0: return [] area = h * w cnz = numpy.count_nonzero total = cnz(image) if total == 0 or area == total: return [] if h < min_h * 2: y_c = [] else: y_c = [(-abs((count / ((h - y) * w)) - ((total - count) / (y * w))), y, 0) for count, y in ((cnz(image[y:]), y) for y in range(min_h, image.shape[0] - min_h))] if w < min_w * 2: x_c = [] else: x_c = [(-abs((count / (h * (w - x))) - ((total - count) / (h * x))), x, 1) for count, x in ((cnz(image[:, x:]), x) for x in range(min_w, image.shape[1] - min_w))] return sorted(x_c + y_c)
Return list of best to worst binary splits along the x and y axis.
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/vision.py#L259-L285
null
from __future__ import division from numpy.fft import irfft2, rfft2 import numpy import itertools import operator def subimage(rect, image): x, y, w, h = rect return image[y:y + h, x:x + w] def pad_bin_image_to_shape(image, shape): """ Padd image to size :shape: with zeros """ h, w = shape ih, iw = image.shape assert ih <= h assert iw <= w if iw < w: result = numpy.hstack((image, numpy.zeros((ih, w - iw), bool))) else: result = image if ih < h: result = numpy.vstack((result, numpy.zeros((h - ih, w), bool))) return result def sum_2d_images(images): it = iter(images) result = numpy.copy(next(it)) for image in it: result += image return result OVERLAP_TABLE = { 16: (16, 8, 4, 2, 1), 15: (15, 5, 3, 1), 12: (12, 6, 4, 3, 2, 1), 10: (10, 5, 2, 1), 9: (9, 3, 1), 8: (8, 4, 2, 1), 6: (6, 3, 2, 1), 5: (5, 1), 4: (4, 2, 1), 3: (3, 1), 2: (2, 1) } # A number near float max which we don't want to get near to keep precision # If you get false matches consider reducing this number. ACCURACY_LIMIT = 2 ** (64 - 23) def best_convolution(bin_template, bin_image, tollerance=0.5, overlap_table=OVERLAP_TABLE): """ Selects and applies the best convolution method to find template in image. Returns a list of matches in (width, height, x offset, y offset) format (where the x and y offsets are from the top left corner). As the images are binary images, we can utilise the extra bit space in the float64's by cutting the image into tiles and stacking them into variable grayscale values. This allows converting a sparse binary image into a dense(r) grayscale one. """ template_sum = numpy.count_nonzero(bin_template) th, tw = bin_template.shape ih, iw = bin_image.shape if template_sum == 0 or th == 0 or tw == 0: # If we don't have a template return [] if th > ih or tw > iw: # If the template is bigger than the image return [] # How many cells can we split the image into? max_vert_cells = ih // th max_hor_cells = iw // th # Try to work out how many times we can stack the image usable_factors = {n: factors for n, factors in overlap_table.iteritems() if ((template_sum + 1) ** (n)) < ACCURACY_LIMIT} overlap_options = [(factor, n // factor) for n, factors in usable_factors.iteritems() for factor in factors if (factor <= max_vert_cells and n // factor <= max_hor_cells)] if not overlap_options: # We can't stack the image return convolution(bin_template, bin_image, tollerance=tollerance) best_overlap = min(overlap_options, key=lambda x: ((ih // x[0] + th) * (iw // x[1] + tw))) return overlapped_convolution(bin_template, bin_image, tollerance=tollerance, splits=best_overlap) def convolution(bin_template, bin_image, tollerance=0.5): expected = numpy.count_nonzero(bin_template) ih, iw = bin_image.shape th, tw = bin_template.shape # Padd image to even dimensions if ih % 2 or iw % 2: if ih % 2: ih += 1 if iw % 2: iw += 1 bin_image = pad_bin_image_to_shape(bin_image, (ih, iw)) if expected == 0: return [] # Calculate the convolution of the FFT's of the image & template convolution_freqs = rfft2(bin_image) * rfft2(bin_template[::-1, ::-1], bin_image.shape) # Reverse the FFT to find the result image convolution_image = irfft2(convolution_freqs) # At this point, the maximum point in convolution_image should be the # bottom right (why?) of the area of greatest match # The areas in the result image within expected +- tollerance are where we # saw matches found_bitmap = ((convolution_image > (expected - tollerance)) & (convolution_image < (expected + tollerance))) match_points = numpy.transpose(numpy.nonzero(found_bitmap)) # bottom right # Find the top left point from the template (remember match_point is # inside the template (hence -1) return [((fx - (tw - 1)), (fy - (th - 1))) for (fy, fx) in match_points] def overlapped_convolution(bin_template, bin_image, tollerance=0.5, splits=(4, 2)): """ As each of these images are hold only binary values, and RFFT2 works on float64 greyscale values, we can make the convolution more efficient by breaking the image up into :splits: sectons. Each one of these sections then has its greyscale value adjusted and then stacked. We then apply the convolution to this 'stack' of images, and adjust the resultant position matches. """ th, tw = bin_template.shape ih, iw = bin_image.shape hs, ws = splits h = ih // hs w = iw // ws count = numpy.count_nonzero(bin_template) assert count > 0 assert h >= th assert w >= tw yoffset = [(i * h, ((i + 1) * h) + (th - 1)) for i in range(hs)] xoffset = [(i * w, ((i + 1) * w) + (tw - 1)) for i in range(ws)] # image_stacks is Origin (x,y), array, z (height in stack) image_stacks = [((x1, y1), bin_image[y1:y2, x1:x2], float((count + 1) ** (num))) for num, (x1, x2, y1, y2) in enumerate((x1, x2, y1, y2) for (x1, x2) in xoffset for (y1, y2) in yoffset)] pad_h = max(i.shape[0] for _, i, _ in image_stacks) pad_w = max(i.shape[1] for _, i, _ in image_stacks) # rfft metrics must be an even size - why ... maths? pad_w += pad_w % 2 pad_h += pad_h % 2 overlapped_image = sum_2d_images(pad_bin_image_to_shape(i, (pad_h, pad_w)) * num for _, i, num in image_stacks) #print "Overlap splits %r, Image Size (%d,%d), #Overlapped Size (%d,%d)" % (splits,iw,ih,pad_w,pad_h) # Calculate the convolution of the FFT's of the overlapped image & template convolution_freqs = (rfft2(overlapped_image) * rfft2(bin_template[::-1, ::-1], overlapped_image.shape)) # Reverse the FFT to find the result overlapped image convolution_image = irfft2(convolution_freqs) # At this point, the maximum point in convolution_image should be the # bottom right (why?) of the area of greatest match results = set() for (x, y), _, num in image_stacks[::-1]: test = convolution_image / num filtered = ((test >= (count - tollerance)) & (test <= (count + tollerance))) match_points = numpy.transpose(numpy.nonzero(filtered)) # bottom right for (fy, fx) in match_points: if fx < (tw - 1) or fy < (th - 1): continue results.add((x + fx - (tw - 1), y + fy - (th - 1))) convolution_image %= num return list(results) def get_possible_convolution_regions(bin_template, bin_image, tollerance=0.5, rescale=10): result = [] h, w = bin_template.shape[:2] rects = set((x, y, w, h) for x, y in convolution_r3m_targets( bin_template, bin_image, tollerance)) ih, iw = bin_image.shape[:2] # areas of interest aoi = numpy.zeros((ih // rescale + 1, iw // rescale + 1), bool) for x, y, w, h in rects: aoi[y // rescale:(y + h) // rescale + 1, x // rescale:(x + w) // rescale + 1] = True bp = binary_partition_image(aoi, min_w=w // rescale, min_h=h // rescale) if bp: bp = prune_unbeneficial_partitions(aoi, bp) return result def binary_partition_to_rects(bp, image, template_w, template_h, xoffset=0, yoffset=0): h, w = image.shape[2:] if bp is None: return [(xoffset, yoffset, w, h)] pos, axis, (p1, p2) = bp if axis == 0: new_xoffset, new_yoffset = xoffset, yoffset + pos i1, i2 = image[pos:], image[:pos] else: new_xoffset, new_yoffset = xoffset + pos, yoffset i1, i2 = image[:, pos:], image[:, :pos] def prune_unbeneficial_partitions(image, bp): pos, axis, (p1, p2) = bp if axis == 0: i1, i2 = image[pos:], image[:pos] else: i1, i2 = image[:, pos:], image[:, :pos] if p1 is None: p1_result = numpy.count_nonzero(i1) == 0 else: p1_result = prune_unbeneficial_partitions(i1, p1) if p2 is None: p2_result = numpy.count_nonzero(i2) == 0 else: p2_result = prune_unbeneficial_partitions(i2, p2) if p1_result or p2_result: return [pos, axis, [None if p1_result in [True, False] else p1_result, None if p2_result in [True, False] else p2_result]] else: return None def get_partition_scores(image, min_w=1, min_h=1): """Return list of best to worst binary splits along the x and y axis. """ h, w = image.shape[:2] if w == 0 or h == 0: return [] area = h * w cnz = numpy.count_nonzero total = cnz(image) if total == 0 or area == total: return [] if h < min_h * 2: y_c = [] else: y_c = [(-abs((count / ((h - y) * w)) - ((total - count) / (y * w))), y, 0) for count, y in ((cnz(image[y:]), y) for y in range(min_h, image.shape[0] - min_h))] if w < min_w * 2: x_c = [] else: x_c = [(-abs((count / (h * (w - x))) - ((total - count) / (h * x))), x, 1) for count, x in ((cnz(image[:, x:]), x) for x in range(min_w, image.shape[1] - min_w))] return sorted(x_c + y_c) def get_best_partition(image, min_w=1, min_h=1): partitions = get_partition_scores(image, min_w=min_w, min_h=min_h) if partitions: return partitions[0][-2:] else: return None def binary_partition_image(image, min_w=1, min_h=1, depth=0, max_depth=-1): """Return a bsp of [pos, axis, [before_node, after_node]] nodes where leaf nodes == None. If max_depth < 0 this function will continue until all leaf nodes have been found, if it is >= 0 leaf nodes will be created at that depth. min_w and min_h are the minimum width or height of a partition. """ if max_depth >= 0 and depth >= max_depth: return None partition = get_best_partition(image, min_w=min_w, min_h=min_h) if partition is None: return None pos, axis = partition if axis == 0: p1 = binary_partition_image( image[pos:], min_w, min_h, depth + 1, max_depth) p2 = binary_partition_image( image[:pos], min_w, min_h, depth + 1, max_depth) elif axis == 1: p1 = binary_partition_image( image[:, pos:], min_w, min_h, depth + 1, max_depth) p2 = binary_partition_image( image[:, :pos], min_w, min_h, depth + 1, max_depth) return [pos, axis, [p1, p2]] def draw_binary_partition(image, subdiv, res_image=None, counter=None): if res_image is None: res_image = numpy.zeros(image.shape) if counter is None: counter = itertools.count(15) pos, axis, (p1, p2) = subdiv if axis == 0: s1, s2 = res_image[pos:], res_image[:pos] i1, i2 = image[pos:], image[:pos] else: s1, s2 = res_image[:, pos:], res_image[:, :pos] i1, i2 = image[:, pos:], image[:, :pos] if p1 is None: if numpy.count_nonzero(i1): s1[:] = next(counter) else: draw_binary_partition(i1, p1, s1, counter) if p2 is None: if numpy.count_nonzero(i2): s2[:] = next(counter) else: draw_binary_partition(i2, p2, s2, counter) return res_image def rescale2avg(image): sub1 = image[:-1:2, :-1:2] sub2 = image[1::2, 1::2] res = numpy.zeros(sub1.shape, numpy.uint32) res += sub1 res += sub2 res /= 2 return res.astype(numpy.uint8) def rescale2max(image): sub1 = image[:-1:2, :-1:2] sub2 = image[1::2, 1::2] if len(image.shape) == 3: max_map = grey_scale(sub1) > grey_scale(sub2) else: max_map = sub1 > sub2 inv_max_map = max_map == False res = numpy.zeros(sub1.shape, numpy.uint8) res[max_map] = sub1[max_map] res[inv_max_map] = sub2[inv_max_map] return res def rescale3avg(image): sub1 = image[:-2:3, :-2:3] sub2 = image[1:-1:3, 1:-1:3] sub3 = image[2::3, 2::3] res = numpy.zeros(sub1.shape, numpy.uint32) res += sub1 res += sub2 res += sub3 res /= 3 return res.astype(numpy.uint8) def rescale3max(image): sub1 = image[:-2:3, :-2:3] sub2 = image[1:-1:3, 1:-1:3] sub3 = image[2::3, 2::3] if len(image.shape) == 3: grey1 = grey_scale(sub1) grey2 = grey_scale(sub2) grey3 = grey_scale(sub3) else: grey1, grey2, grey3 = sub1, sub2, sub3 max_map_1 = (grey1 > grey2) & (grey1 > grey3) max_map_2 = (grey2 > grey1) & (grey2 > grey3) max_map_3 = (max_map_1 | max_map_2) == False res = numpy.zeros(sub1.shape, numpy.uint8) res[max_map_1] = sub1[max_map_1] res[max_map_2] = sub2[max_map_2] res[max_map_3] = sub3[max_map_3] return res def or_reduce_rescale3max_offset(image): return reduce(operator.or_, (rescale3max(image[y1:y2, x1:x2]) for (y1, y2), (x1, x2) in itertools.product( *[[(i, -(3 - i)) for i in range(3)]] * 2) ) ) def numpy_or_all(images): it = iter(images) result = numpy.copy(next(it)) for image in it: result |= image return result def grey_scale(image): """Converts RGB image to Greyscale :param image: input image :type image: 3 channel 3d :class:`numpy.ndarray` :rtype: :class:`numpy.ndarray` """ return image.astype(numpy.int32).sum(axis=2) // 3 def find_edges(image): base = image[:, :, numpy.newaxis].astype(numpy.int16) c = base[1:-1, 1:-1] e = base[1:-1, :-2] w = base[1:-1, 2:] s = base[:-2, 1:-1] n = base[2:, 1:-1] diffs = numpy.concatenate([c-e, c-w, c-s, c-n], axis=2).max(axis=2) diffs[diffs < 0] = 0 dh, dw = diffs.shape col = numpy.zeros((dh, 1), numpy.uint8) row = numpy.zeros((1, dw+2), numpy.uint8) return numpy.vstack( (row, numpy.hstack((col, diffs.astype(numpy.uint8), col)), row) ) def find_threshold_near_density(img, density, low=0, high=255): """Find a threshold where the fraction of pixels above the threshold is closest to density where density is (count of pixels above threshold / count of pixels). The highest threshold closest to the desired density will be returned. Use low and high to exclude undesirable thresholds. :param img: target image :type img: 2d :class:`numpy.ndarray` :param density: target density :type density: float between 0.0 and 1.0 :param low: min threshold to test :type low: ubyte :param migh: max threshold to test :type low: ubyte :rtype: ubyte """ size = numpy.size(img) densities = [] last_t = None while True: t = ((high - low) // 2) + low if t == last_t: densities.sort(key=lambda x: (abs(x[0] - density), 256 - x[1])) return densities[0][1] else: last_t = t d = numpy.count_nonzero(img > t) / size densities.append((d, t)) if d < density: high = t elif d >= density: # search away from low low = t def filter_greys_using_image(image, target): """Filter out any values in target not in image :param image: image containing values to appear in filtered image :param target: the image to filter :rtype: 2d :class:`numpy.ndarray` containing only value in image and with the same dimensions as target """ maskbase = numpy.array(range(256), dtype=numpy.uint8) mask = numpy.where(numpy.in1d(maskbase, numpy.unique(image)), maskbase, 0) return mask[target] def correlation_coefficient_normed(template, image): h, w = template.shape[:2] H, W = image.shape[:2] template_size = template.size template_distance = template - (template.sum() / template_size) corr_num = numpy.zeros((H, W), numpy.float64) corr_denum = numpy.zeros((H, W), numpy.float64) for y in xrange(H): for x in xrange(W): image_in_template_area = image[y:y + h, x:x + w] image_distance_of_template_area = ( image_in_template_area - ( image_in_template_area.sum() / template_size ) ) I_H, I_W = image_distance_of_template_area.shape[:2] sum_of_template_by_image_distance_at_xy = ( image_distance_of_template_area * template_distance[:I_H, :I_W] ).sum() corr_num[y, x] = sum_of_template_by_image_distance_at_xy corr_denum[y, x] = numpy.sqrt( (template_distance ** 2).sum() * (image_distance_of_template_area ** 2).sum() ) print y return corr_num / corr_denum
ten10solutions/Geist
geist/vision.py
binary_partition_image
python
def binary_partition_image(image, min_w=1, min_h=1, depth=0, max_depth=-1): """Return a bsp of [pos, axis, [before_node, after_node]] nodes where leaf nodes == None. If max_depth < 0 this function will continue until all leaf nodes have been found, if it is >= 0 leaf nodes will be created at that depth. min_w and min_h are the minimum width or height of a partition. """ if max_depth >= 0 and depth >= max_depth: return None partition = get_best_partition(image, min_w=min_w, min_h=min_h) if partition is None: return None pos, axis = partition if axis == 0: p1 = binary_partition_image( image[pos:], min_w, min_h, depth + 1, max_depth) p2 = binary_partition_image( image[:pos], min_w, min_h, depth + 1, max_depth) elif axis == 1: p1 = binary_partition_image( image[:, pos:], min_w, min_h, depth + 1, max_depth) p2 = binary_partition_image( image[:, :pos], min_w, min_h, depth + 1, max_depth) return [pos, axis, [p1, p2]]
Return a bsp of [pos, axis, [before_node, after_node]] nodes where leaf nodes == None. If max_depth < 0 this function will continue until all leaf nodes have been found, if it is >= 0 leaf nodes will be created at that depth. min_w and min_h are the minimum width or height of a partition.
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/vision.py#L296-L322
[ "def binary_partition_image(image, min_w=1, min_h=1, depth=0, max_depth=-1):\n \"\"\"Return a bsp of [pos, axis, [before_node, after_node]] nodes where leaf\n nodes == None.\n\n If max_depth < 0 this function will continue until all leaf nodes have\n been found, if it is >= 0 leaf nodes will be created at that depth.\n\n min_w and min_h are the minimum width or height of a partition.\n\n \"\"\"\n if max_depth >= 0 and depth >= max_depth:\n return None\n partition = get_best_partition(image, min_w=min_w, min_h=min_h)\n if partition is None:\n return None\n pos, axis = partition\n if axis == 0:\n p1 = binary_partition_image(\n image[pos:], min_w, min_h, depth + 1, max_depth)\n p2 = binary_partition_image(\n image[:pos], min_w, min_h, depth + 1, max_depth)\n elif axis == 1:\n p1 = binary_partition_image(\n image[:, pos:], min_w, min_h, depth + 1, max_depth)\n p2 = binary_partition_image(\n image[:, :pos], min_w, min_h, depth + 1, max_depth)\n return [pos, axis, [p1, p2]]\n", "def get_best_partition(image, min_w=1, min_h=1):\n partitions = get_partition_scores(image, min_w=min_w, min_h=min_h)\n if partitions:\n return partitions[0][-2:]\n else:\n return None\n" ]
from __future__ import division from numpy.fft import irfft2, rfft2 import numpy import itertools import operator def subimage(rect, image): x, y, w, h = rect return image[y:y + h, x:x + w] def pad_bin_image_to_shape(image, shape): """ Padd image to size :shape: with zeros """ h, w = shape ih, iw = image.shape assert ih <= h assert iw <= w if iw < w: result = numpy.hstack((image, numpy.zeros((ih, w - iw), bool))) else: result = image if ih < h: result = numpy.vstack((result, numpy.zeros((h - ih, w), bool))) return result def sum_2d_images(images): it = iter(images) result = numpy.copy(next(it)) for image in it: result += image return result OVERLAP_TABLE = { 16: (16, 8, 4, 2, 1), 15: (15, 5, 3, 1), 12: (12, 6, 4, 3, 2, 1), 10: (10, 5, 2, 1), 9: (9, 3, 1), 8: (8, 4, 2, 1), 6: (6, 3, 2, 1), 5: (5, 1), 4: (4, 2, 1), 3: (3, 1), 2: (2, 1) } # A number near float max which we don't want to get near to keep precision # If you get false matches consider reducing this number. ACCURACY_LIMIT = 2 ** (64 - 23) def best_convolution(bin_template, bin_image, tollerance=0.5, overlap_table=OVERLAP_TABLE): """ Selects and applies the best convolution method to find template in image. Returns a list of matches in (width, height, x offset, y offset) format (where the x and y offsets are from the top left corner). As the images are binary images, we can utilise the extra bit space in the float64's by cutting the image into tiles and stacking them into variable grayscale values. This allows converting a sparse binary image into a dense(r) grayscale one. """ template_sum = numpy.count_nonzero(bin_template) th, tw = bin_template.shape ih, iw = bin_image.shape if template_sum == 0 or th == 0 or tw == 0: # If we don't have a template return [] if th > ih or tw > iw: # If the template is bigger than the image return [] # How many cells can we split the image into? max_vert_cells = ih // th max_hor_cells = iw // th # Try to work out how many times we can stack the image usable_factors = {n: factors for n, factors in overlap_table.iteritems() if ((template_sum + 1) ** (n)) < ACCURACY_LIMIT} overlap_options = [(factor, n // factor) for n, factors in usable_factors.iteritems() for factor in factors if (factor <= max_vert_cells and n // factor <= max_hor_cells)] if not overlap_options: # We can't stack the image return convolution(bin_template, bin_image, tollerance=tollerance) best_overlap = min(overlap_options, key=lambda x: ((ih // x[0] + th) * (iw // x[1] + tw))) return overlapped_convolution(bin_template, bin_image, tollerance=tollerance, splits=best_overlap) def convolution(bin_template, bin_image, tollerance=0.5): expected = numpy.count_nonzero(bin_template) ih, iw = bin_image.shape th, tw = bin_template.shape # Padd image to even dimensions if ih % 2 or iw % 2: if ih % 2: ih += 1 if iw % 2: iw += 1 bin_image = pad_bin_image_to_shape(bin_image, (ih, iw)) if expected == 0: return [] # Calculate the convolution of the FFT's of the image & template convolution_freqs = rfft2(bin_image) * rfft2(bin_template[::-1, ::-1], bin_image.shape) # Reverse the FFT to find the result image convolution_image = irfft2(convolution_freqs) # At this point, the maximum point in convolution_image should be the # bottom right (why?) of the area of greatest match # The areas in the result image within expected +- tollerance are where we # saw matches found_bitmap = ((convolution_image > (expected - tollerance)) & (convolution_image < (expected + tollerance))) match_points = numpy.transpose(numpy.nonzero(found_bitmap)) # bottom right # Find the top left point from the template (remember match_point is # inside the template (hence -1) return [((fx - (tw - 1)), (fy - (th - 1))) for (fy, fx) in match_points] def overlapped_convolution(bin_template, bin_image, tollerance=0.5, splits=(4, 2)): """ As each of these images are hold only binary values, and RFFT2 works on float64 greyscale values, we can make the convolution more efficient by breaking the image up into :splits: sectons. Each one of these sections then has its greyscale value adjusted and then stacked. We then apply the convolution to this 'stack' of images, and adjust the resultant position matches. """ th, tw = bin_template.shape ih, iw = bin_image.shape hs, ws = splits h = ih // hs w = iw // ws count = numpy.count_nonzero(bin_template) assert count > 0 assert h >= th assert w >= tw yoffset = [(i * h, ((i + 1) * h) + (th - 1)) for i in range(hs)] xoffset = [(i * w, ((i + 1) * w) + (tw - 1)) for i in range(ws)] # image_stacks is Origin (x,y), array, z (height in stack) image_stacks = [((x1, y1), bin_image[y1:y2, x1:x2], float((count + 1) ** (num))) for num, (x1, x2, y1, y2) in enumerate((x1, x2, y1, y2) for (x1, x2) in xoffset for (y1, y2) in yoffset)] pad_h = max(i.shape[0] for _, i, _ in image_stacks) pad_w = max(i.shape[1] for _, i, _ in image_stacks) # rfft metrics must be an even size - why ... maths? pad_w += pad_w % 2 pad_h += pad_h % 2 overlapped_image = sum_2d_images(pad_bin_image_to_shape(i, (pad_h, pad_w)) * num for _, i, num in image_stacks) #print "Overlap splits %r, Image Size (%d,%d), #Overlapped Size (%d,%d)" % (splits,iw,ih,pad_w,pad_h) # Calculate the convolution of the FFT's of the overlapped image & template convolution_freqs = (rfft2(overlapped_image) * rfft2(bin_template[::-1, ::-1], overlapped_image.shape)) # Reverse the FFT to find the result overlapped image convolution_image = irfft2(convolution_freqs) # At this point, the maximum point in convolution_image should be the # bottom right (why?) of the area of greatest match results = set() for (x, y), _, num in image_stacks[::-1]: test = convolution_image / num filtered = ((test >= (count - tollerance)) & (test <= (count + tollerance))) match_points = numpy.transpose(numpy.nonzero(filtered)) # bottom right for (fy, fx) in match_points: if fx < (tw - 1) or fy < (th - 1): continue results.add((x + fx - (tw - 1), y + fy - (th - 1))) convolution_image %= num return list(results) def get_possible_convolution_regions(bin_template, bin_image, tollerance=0.5, rescale=10): result = [] h, w = bin_template.shape[:2] rects = set((x, y, w, h) for x, y in convolution_r3m_targets( bin_template, bin_image, tollerance)) ih, iw = bin_image.shape[:2] # areas of interest aoi = numpy.zeros((ih // rescale + 1, iw // rescale + 1), bool) for x, y, w, h in rects: aoi[y // rescale:(y + h) // rescale + 1, x // rescale:(x + w) // rescale + 1] = True bp = binary_partition_image(aoi, min_w=w // rescale, min_h=h // rescale) if bp: bp = prune_unbeneficial_partitions(aoi, bp) return result def binary_partition_to_rects(bp, image, template_w, template_h, xoffset=0, yoffset=0): h, w = image.shape[2:] if bp is None: return [(xoffset, yoffset, w, h)] pos, axis, (p1, p2) = bp if axis == 0: new_xoffset, new_yoffset = xoffset, yoffset + pos i1, i2 = image[pos:], image[:pos] else: new_xoffset, new_yoffset = xoffset + pos, yoffset i1, i2 = image[:, pos:], image[:, :pos] def prune_unbeneficial_partitions(image, bp): pos, axis, (p1, p2) = bp if axis == 0: i1, i2 = image[pos:], image[:pos] else: i1, i2 = image[:, pos:], image[:, :pos] if p1 is None: p1_result = numpy.count_nonzero(i1) == 0 else: p1_result = prune_unbeneficial_partitions(i1, p1) if p2 is None: p2_result = numpy.count_nonzero(i2) == 0 else: p2_result = prune_unbeneficial_partitions(i2, p2) if p1_result or p2_result: return [pos, axis, [None if p1_result in [True, False] else p1_result, None if p2_result in [True, False] else p2_result]] else: return None def get_partition_scores(image, min_w=1, min_h=1): """Return list of best to worst binary splits along the x and y axis. """ h, w = image.shape[:2] if w == 0 or h == 0: return [] area = h * w cnz = numpy.count_nonzero total = cnz(image) if total == 0 or area == total: return [] if h < min_h * 2: y_c = [] else: y_c = [(-abs((count / ((h - y) * w)) - ((total - count) / (y * w))), y, 0) for count, y in ((cnz(image[y:]), y) for y in range(min_h, image.shape[0] - min_h))] if w < min_w * 2: x_c = [] else: x_c = [(-abs((count / (h * (w - x))) - ((total - count) / (h * x))), x, 1) for count, x in ((cnz(image[:, x:]), x) for x in range(min_w, image.shape[1] - min_w))] return sorted(x_c + y_c) def get_best_partition(image, min_w=1, min_h=1): partitions = get_partition_scores(image, min_w=min_w, min_h=min_h) if partitions: return partitions[0][-2:] else: return None def binary_partition_image(image, min_w=1, min_h=1, depth=0, max_depth=-1): """Return a bsp of [pos, axis, [before_node, after_node]] nodes where leaf nodes == None. If max_depth < 0 this function will continue until all leaf nodes have been found, if it is >= 0 leaf nodes will be created at that depth. min_w and min_h are the minimum width or height of a partition. """ if max_depth >= 0 and depth >= max_depth: return None partition = get_best_partition(image, min_w=min_w, min_h=min_h) if partition is None: return None pos, axis = partition if axis == 0: p1 = binary_partition_image( image[pos:], min_w, min_h, depth + 1, max_depth) p2 = binary_partition_image( image[:pos], min_w, min_h, depth + 1, max_depth) elif axis == 1: p1 = binary_partition_image( image[:, pos:], min_w, min_h, depth + 1, max_depth) p2 = binary_partition_image( image[:, :pos], min_w, min_h, depth + 1, max_depth) return [pos, axis, [p1, p2]] def draw_binary_partition(image, subdiv, res_image=None, counter=None): if res_image is None: res_image = numpy.zeros(image.shape) if counter is None: counter = itertools.count(15) pos, axis, (p1, p2) = subdiv if axis == 0: s1, s2 = res_image[pos:], res_image[:pos] i1, i2 = image[pos:], image[:pos] else: s1, s2 = res_image[:, pos:], res_image[:, :pos] i1, i2 = image[:, pos:], image[:, :pos] if p1 is None: if numpy.count_nonzero(i1): s1[:] = next(counter) else: draw_binary_partition(i1, p1, s1, counter) if p2 is None: if numpy.count_nonzero(i2): s2[:] = next(counter) else: draw_binary_partition(i2, p2, s2, counter) return res_image def rescale2avg(image): sub1 = image[:-1:2, :-1:2] sub2 = image[1::2, 1::2] res = numpy.zeros(sub1.shape, numpy.uint32) res += sub1 res += sub2 res /= 2 return res.astype(numpy.uint8) def rescale2max(image): sub1 = image[:-1:2, :-1:2] sub2 = image[1::2, 1::2] if len(image.shape) == 3: max_map = grey_scale(sub1) > grey_scale(sub2) else: max_map = sub1 > sub2 inv_max_map = max_map == False res = numpy.zeros(sub1.shape, numpy.uint8) res[max_map] = sub1[max_map] res[inv_max_map] = sub2[inv_max_map] return res def rescale3avg(image): sub1 = image[:-2:3, :-2:3] sub2 = image[1:-1:3, 1:-1:3] sub3 = image[2::3, 2::3] res = numpy.zeros(sub1.shape, numpy.uint32) res += sub1 res += sub2 res += sub3 res /= 3 return res.astype(numpy.uint8) def rescale3max(image): sub1 = image[:-2:3, :-2:3] sub2 = image[1:-1:3, 1:-1:3] sub3 = image[2::3, 2::3] if len(image.shape) == 3: grey1 = grey_scale(sub1) grey2 = grey_scale(sub2) grey3 = grey_scale(sub3) else: grey1, grey2, grey3 = sub1, sub2, sub3 max_map_1 = (grey1 > grey2) & (grey1 > grey3) max_map_2 = (grey2 > grey1) & (grey2 > grey3) max_map_3 = (max_map_1 | max_map_2) == False res = numpy.zeros(sub1.shape, numpy.uint8) res[max_map_1] = sub1[max_map_1] res[max_map_2] = sub2[max_map_2] res[max_map_3] = sub3[max_map_3] return res def or_reduce_rescale3max_offset(image): return reduce(operator.or_, (rescale3max(image[y1:y2, x1:x2]) for (y1, y2), (x1, x2) in itertools.product( *[[(i, -(3 - i)) for i in range(3)]] * 2) ) ) def numpy_or_all(images): it = iter(images) result = numpy.copy(next(it)) for image in it: result |= image return result def grey_scale(image): """Converts RGB image to Greyscale :param image: input image :type image: 3 channel 3d :class:`numpy.ndarray` :rtype: :class:`numpy.ndarray` """ return image.astype(numpy.int32).sum(axis=2) // 3 def find_edges(image): base = image[:, :, numpy.newaxis].astype(numpy.int16) c = base[1:-1, 1:-1] e = base[1:-1, :-2] w = base[1:-1, 2:] s = base[:-2, 1:-1] n = base[2:, 1:-1] diffs = numpy.concatenate([c-e, c-w, c-s, c-n], axis=2).max(axis=2) diffs[diffs < 0] = 0 dh, dw = diffs.shape col = numpy.zeros((dh, 1), numpy.uint8) row = numpy.zeros((1, dw+2), numpy.uint8) return numpy.vstack( (row, numpy.hstack((col, diffs.astype(numpy.uint8), col)), row) ) def find_threshold_near_density(img, density, low=0, high=255): """Find a threshold where the fraction of pixels above the threshold is closest to density where density is (count of pixels above threshold / count of pixels). The highest threshold closest to the desired density will be returned. Use low and high to exclude undesirable thresholds. :param img: target image :type img: 2d :class:`numpy.ndarray` :param density: target density :type density: float between 0.0 and 1.0 :param low: min threshold to test :type low: ubyte :param migh: max threshold to test :type low: ubyte :rtype: ubyte """ size = numpy.size(img) densities = [] last_t = None while True: t = ((high - low) // 2) + low if t == last_t: densities.sort(key=lambda x: (abs(x[0] - density), 256 - x[1])) return densities[0][1] else: last_t = t d = numpy.count_nonzero(img > t) / size densities.append((d, t)) if d < density: high = t elif d >= density: # search away from low low = t def filter_greys_using_image(image, target): """Filter out any values in target not in image :param image: image containing values to appear in filtered image :param target: the image to filter :rtype: 2d :class:`numpy.ndarray` containing only value in image and with the same dimensions as target """ maskbase = numpy.array(range(256), dtype=numpy.uint8) mask = numpy.where(numpy.in1d(maskbase, numpy.unique(image)), maskbase, 0) return mask[target] def correlation_coefficient_normed(template, image): h, w = template.shape[:2] H, W = image.shape[:2] template_size = template.size template_distance = template - (template.sum() / template_size) corr_num = numpy.zeros((H, W), numpy.float64) corr_denum = numpy.zeros((H, W), numpy.float64) for y in xrange(H): for x in xrange(W): image_in_template_area = image[y:y + h, x:x + w] image_distance_of_template_area = ( image_in_template_area - ( image_in_template_area.sum() / template_size ) ) I_H, I_W = image_distance_of_template_area.shape[:2] sum_of_template_by_image_distance_at_xy = ( image_distance_of_template_area * template_distance[:I_H, :I_W] ).sum() corr_num[y, x] = sum_of_template_by_image_distance_at_xy corr_denum[y, x] = numpy.sqrt( (template_distance ** 2).sum() * (image_distance_of_template_area ** 2).sum() ) print y return corr_num / corr_denum
ten10solutions/Geist
geist/vision.py
find_threshold_near_density
python
def find_threshold_near_density(img, density, low=0, high=255): """Find a threshold where the fraction of pixels above the threshold is closest to density where density is (count of pixels above threshold / count of pixels). The highest threshold closest to the desired density will be returned. Use low and high to exclude undesirable thresholds. :param img: target image :type img: 2d :class:`numpy.ndarray` :param density: target density :type density: float between 0.0 and 1.0 :param low: min threshold to test :type low: ubyte :param migh: max threshold to test :type low: ubyte :rtype: ubyte """ size = numpy.size(img) densities = [] last_t = None while True: t = ((high - low) // 2) + low if t == last_t: densities.sort(key=lambda x: (abs(x[0] - density), 256 - x[1])) return densities[0][1] else: last_t = t d = numpy.count_nonzero(img > t) / size densities.append((d, t)) if d < density: high = t elif d >= density: # search away from low low = t
Find a threshold where the fraction of pixels above the threshold is closest to density where density is (count of pixels above threshold / count of pixels). The highest threshold closest to the desired density will be returned. Use low and high to exclude undesirable thresholds. :param img: target image :type img: 2d :class:`numpy.ndarray` :param density: target density :type density: float between 0.0 and 1.0 :param low: min threshold to test :type low: ubyte :param migh: max threshold to test :type low: ubyte :rtype: ubyte
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/vision.py#L450-L485
null
from __future__ import division from numpy.fft import irfft2, rfft2 import numpy import itertools import operator def subimage(rect, image): x, y, w, h = rect return image[y:y + h, x:x + w] def pad_bin_image_to_shape(image, shape): """ Padd image to size :shape: with zeros """ h, w = shape ih, iw = image.shape assert ih <= h assert iw <= w if iw < w: result = numpy.hstack((image, numpy.zeros((ih, w - iw), bool))) else: result = image if ih < h: result = numpy.vstack((result, numpy.zeros((h - ih, w), bool))) return result def sum_2d_images(images): it = iter(images) result = numpy.copy(next(it)) for image in it: result += image return result OVERLAP_TABLE = { 16: (16, 8, 4, 2, 1), 15: (15, 5, 3, 1), 12: (12, 6, 4, 3, 2, 1), 10: (10, 5, 2, 1), 9: (9, 3, 1), 8: (8, 4, 2, 1), 6: (6, 3, 2, 1), 5: (5, 1), 4: (4, 2, 1), 3: (3, 1), 2: (2, 1) } # A number near float max which we don't want to get near to keep precision # If you get false matches consider reducing this number. ACCURACY_LIMIT = 2 ** (64 - 23) def best_convolution(bin_template, bin_image, tollerance=0.5, overlap_table=OVERLAP_TABLE): """ Selects and applies the best convolution method to find template in image. Returns a list of matches in (width, height, x offset, y offset) format (where the x and y offsets are from the top left corner). As the images are binary images, we can utilise the extra bit space in the float64's by cutting the image into tiles and stacking them into variable grayscale values. This allows converting a sparse binary image into a dense(r) grayscale one. """ template_sum = numpy.count_nonzero(bin_template) th, tw = bin_template.shape ih, iw = bin_image.shape if template_sum == 0 or th == 0 or tw == 0: # If we don't have a template return [] if th > ih or tw > iw: # If the template is bigger than the image return [] # How many cells can we split the image into? max_vert_cells = ih // th max_hor_cells = iw // th # Try to work out how many times we can stack the image usable_factors = {n: factors for n, factors in overlap_table.iteritems() if ((template_sum + 1) ** (n)) < ACCURACY_LIMIT} overlap_options = [(factor, n // factor) for n, factors in usable_factors.iteritems() for factor in factors if (factor <= max_vert_cells and n // factor <= max_hor_cells)] if not overlap_options: # We can't stack the image return convolution(bin_template, bin_image, tollerance=tollerance) best_overlap = min(overlap_options, key=lambda x: ((ih // x[0] + th) * (iw // x[1] + tw))) return overlapped_convolution(bin_template, bin_image, tollerance=tollerance, splits=best_overlap) def convolution(bin_template, bin_image, tollerance=0.5): expected = numpy.count_nonzero(bin_template) ih, iw = bin_image.shape th, tw = bin_template.shape # Padd image to even dimensions if ih % 2 or iw % 2: if ih % 2: ih += 1 if iw % 2: iw += 1 bin_image = pad_bin_image_to_shape(bin_image, (ih, iw)) if expected == 0: return [] # Calculate the convolution of the FFT's of the image & template convolution_freqs = rfft2(bin_image) * rfft2(bin_template[::-1, ::-1], bin_image.shape) # Reverse the FFT to find the result image convolution_image = irfft2(convolution_freqs) # At this point, the maximum point in convolution_image should be the # bottom right (why?) of the area of greatest match # The areas in the result image within expected +- tollerance are where we # saw matches found_bitmap = ((convolution_image > (expected - tollerance)) & (convolution_image < (expected + tollerance))) match_points = numpy.transpose(numpy.nonzero(found_bitmap)) # bottom right # Find the top left point from the template (remember match_point is # inside the template (hence -1) return [((fx - (tw - 1)), (fy - (th - 1))) for (fy, fx) in match_points] def overlapped_convolution(bin_template, bin_image, tollerance=0.5, splits=(4, 2)): """ As each of these images are hold only binary values, and RFFT2 works on float64 greyscale values, we can make the convolution more efficient by breaking the image up into :splits: sectons. Each one of these sections then has its greyscale value adjusted and then stacked. We then apply the convolution to this 'stack' of images, and adjust the resultant position matches. """ th, tw = bin_template.shape ih, iw = bin_image.shape hs, ws = splits h = ih // hs w = iw // ws count = numpy.count_nonzero(bin_template) assert count > 0 assert h >= th assert w >= tw yoffset = [(i * h, ((i + 1) * h) + (th - 1)) for i in range(hs)] xoffset = [(i * w, ((i + 1) * w) + (tw - 1)) for i in range(ws)] # image_stacks is Origin (x,y), array, z (height in stack) image_stacks = [((x1, y1), bin_image[y1:y2, x1:x2], float((count + 1) ** (num))) for num, (x1, x2, y1, y2) in enumerate((x1, x2, y1, y2) for (x1, x2) in xoffset for (y1, y2) in yoffset)] pad_h = max(i.shape[0] for _, i, _ in image_stacks) pad_w = max(i.shape[1] for _, i, _ in image_stacks) # rfft metrics must be an even size - why ... maths? pad_w += pad_w % 2 pad_h += pad_h % 2 overlapped_image = sum_2d_images(pad_bin_image_to_shape(i, (pad_h, pad_w)) * num for _, i, num in image_stacks) #print "Overlap splits %r, Image Size (%d,%d), #Overlapped Size (%d,%d)" % (splits,iw,ih,pad_w,pad_h) # Calculate the convolution of the FFT's of the overlapped image & template convolution_freqs = (rfft2(overlapped_image) * rfft2(bin_template[::-1, ::-1], overlapped_image.shape)) # Reverse the FFT to find the result overlapped image convolution_image = irfft2(convolution_freqs) # At this point, the maximum point in convolution_image should be the # bottom right (why?) of the area of greatest match results = set() for (x, y), _, num in image_stacks[::-1]: test = convolution_image / num filtered = ((test >= (count - tollerance)) & (test <= (count + tollerance))) match_points = numpy.transpose(numpy.nonzero(filtered)) # bottom right for (fy, fx) in match_points: if fx < (tw - 1) or fy < (th - 1): continue results.add((x + fx - (tw - 1), y + fy - (th - 1))) convolution_image %= num return list(results) def get_possible_convolution_regions(bin_template, bin_image, tollerance=0.5, rescale=10): result = [] h, w = bin_template.shape[:2] rects = set((x, y, w, h) for x, y in convolution_r3m_targets( bin_template, bin_image, tollerance)) ih, iw = bin_image.shape[:2] # areas of interest aoi = numpy.zeros((ih // rescale + 1, iw // rescale + 1), bool) for x, y, w, h in rects: aoi[y // rescale:(y + h) // rescale + 1, x // rescale:(x + w) // rescale + 1] = True bp = binary_partition_image(aoi, min_w=w // rescale, min_h=h // rescale) if bp: bp = prune_unbeneficial_partitions(aoi, bp) return result def binary_partition_to_rects(bp, image, template_w, template_h, xoffset=0, yoffset=0): h, w = image.shape[2:] if bp is None: return [(xoffset, yoffset, w, h)] pos, axis, (p1, p2) = bp if axis == 0: new_xoffset, new_yoffset = xoffset, yoffset + pos i1, i2 = image[pos:], image[:pos] else: new_xoffset, new_yoffset = xoffset + pos, yoffset i1, i2 = image[:, pos:], image[:, :pos] def prune_unbeneficial_partitions(image, bp): pos, axis, (p1, p2) = bp if axis == 0: i1, i2 = image[pos:], image[:pos] else: i1, i2 = image[:, pos:], image[:, :pos] if p1 is None: p1_result = numpy.count_nonzero(i1) == 0 else: p1_result = prune_unbeneficial_partitions(i1, p1) if p2 is None: p2_result = numpy.count_nonzero(i2) == 0 else: p2_result = prune_unbeneficial_partitions(i2, p2) if p1_result or p2_result: return [pos, axis, [None if p1_result in [True, False] else p1_result, None if p2_result in [True, False] else p2_result]] else: return None def get_partition_scores(image, min_w=1, min_h=1): """Return list of best to worst binary splits along the x and y axis. """ h, w = image.shape[:2] if w == 0 or h == 0: return [] area = h * w cnz = numpy.count_nonzero total = cnz(image) if total == 0 or area == total: return [] if h < min_h * 2: y_c = [] else: y_c = [(-abs((count / ((h - y) * w)) - ((total - count) / (y * w))), y, 0) for count, y in ((cnz(image[y:]), y) for y in range(min_h, image.shape[0] - min_h))] if w < min_w * 2: x_c = [] else: x_c = [(-abs((count / (h * (w - x))) - ((total - count) / (h * x))), x, 1) for count, x in ((cnz(image[:, x:]), x) for x in range(min_w, image.shape[1] - min_w))] return sorted(x_c + y_c) def get_best_partition(image, min_w=1, min_h=1): partitions = get_partition_scores(image, min_w=min_w, min_h=min_h) if partitions: return partitions[0][-2:] else: return None def binary_partition_image(image, min_w=1, min_h=1, depth=0, max_depth=-1): """Return a bsp of [pos, axis, [before_node, after_node]] nodes where leaf nodes == None. If max_depth < 0 this function will continue until all leaf nodes have been found, if it is >= 0 leaf nodes will be created at that depth. min_w and min_h are the minimum width or height of a partition. """ if max_depth >= 0 and depth >= max_depth: return None partition = get_best_partition(image, min_w=min_w, min_h=min_h) if partition is None: return None pos, axis = partition if axis == 0: p1 = binary_partition_image( image[pos:], min_w, min_h, depth + 1, max_depth) p2 = binary_partition_image( image[:pos], min_w, min_h, depth + 1, max_depth) elif axis == 1: p1 = binary_partition_image( image[:, pos:], min_w, min_h, depth + 1, max_depth) p2 = binary_partition_image( image[:, :pos], min_w, min_h, depth + 1, max_depth) return [pos, axis, [p1, p2]] def draw_binary_partition(image, subdiv, res_image=None, counter=None): if res_image is None: res_image = numpy.zeros(image.shape) if counter is None: counter = itertools.count(15) pos, axis, (p1, p2) = subdiv if axis == 0: s1, s2 = res_image[pos:], res_image[:pos] i1, i2 = image[pos:], image[:pos] else: s1, s2 = res_image[:, pos:], res_image[:, :pos] i1, i2 = image[:, pos:], image[:, :pos] if p1 is None: if numpy.count_nonzero(i1): s1[:] = next(counter) else: draw_binary_partition(i1, p1, s1, counter) if p2 is None: if numpy.count_nonzero(i2): s2[:] = next(counter) else: draw_binary_partition(i2, p2, s2, counter) return res_image def rescale2avg(image): sub1 = image[:-1:2, :-1:2] sub2 = image[1::2, 1::2] res = numpy.zeros(sub1.shape, numpy.uint32) res += sub1 res += sub2 res /= 2 return res.astype(numpy.uint8) def rescale2max(image): sub1 = image[:-1:2, :-1:2] sub2 = image[1::2, 1::2] if len(image.shape) == 3: max_map = grey_scale(sub1) > grey_scale(sub2) else: max_map = sub1 > sub2 inv_max_map = max_map == False res = numpy.zeros(sub1.shape, numpy.uint8) res[max_map] = sub1[max_map] res[inv_max_map] = sub2[inv_max_map] return res def rescale3avg(image): sub1 = image[:-2:3, :-2:3] sub2 = image[1:-1:3, 1:-1:3] sub3 = image[2::3, 2::3] res = numpy.zeros(sub1.shape, numpy.uint32) res += sub1 res += sub2 res += sub3 res /= 3 return res.astype(numpy.uint8) def rescale3max(image): sub1 = image[:-2:3, :-2:3] sub2 = image[1:-1:3, 1:-1:3] sub3 = image[2::3, 2::3] if len(image.shape) == 3: grey1 = grey_scale(sub1) grey2 = grey_scale(sub2) grey3 = grey_scale(sub3) else: grey1, grey2, grey3 = sub1, sub2, sub3 max_map_1 = (grey1 > grey2) & (grey1 > grey3) max_map_2 = (grey2 > grey1) & (grey2 > grey3) max_map_3 = (max_map_1 | max_map_2) == False res = numpy.zeros(sub1.shape, numpy.uint8) res[max_map_1] = sub1[max_map_1] res[max_map_2] = sub2[max_map_2] res[max_map_3] = sub3[max_map_3] return res def or_reduce_rescale3max_offset(image): return reduce(operator.or_, (rescale3max(image[y1:y2, x1:x2]) for (y1, y2), (x1, x2) in itertools.product( *[[(i, -(3 - i)) for i in range(3)]] * 2) ) ) def numpy_or_all(images): it = iter(images) result = numpy.copy(next(it)) for image in it: result |= image return result def grey_scale(image): """Converts RGB image to Greyscale :param image: input image :type image: 3 channel 3d :class:`numpy.ndarray` :rtype: :class:`numpy.ndarray` """ return image.astype(numpy.int32).sum(axis=2) // 3 def find_edges(image): base = image[:, :, numpy.newaxis].astype(numpy.int16) c = base[1:-1, 1:-1] e = base[1:-1, :-2] w = base[1:-1, 2:] s = base[:-2, 1:-1] n = base[2:, 1:-1] diffs = numpy.concatenate([c-e, c-w, c-s, c-n], axis=2).max(axis=2) diffs[diffs < 0] = 0 dh, dw = diffs.shape col = numpy.zeros((dh, 1), numpy.uint8) row = numpy.zeros((1, dw+2), numpy.uint8) return numpy.vstack( (row, numpy.hstack((col, diffs.astype(numpy.uint8), col)), row) ) def find_threshold_near_density(img, density, low=0, high=255): """Find a threshold where the fraction of pixels above the threshold is closest to density where density is (count of pixels above threshold / count of pixels). The highest threshold closest to the desired density will be returned. Use low and high to exclude undesirable thresholds. :param img: target image :type img: 2d :class:`numpy.ndarray` :param density: target density :type density: float between 0.0 and 1.0 :param low: min threshold to test :type low: ubyte :param migh: max threshold to test :type low: ubyte :rtype: ubyte """ size = numpy.size(img) densities = [] last_t = None while True: t = ((high - low) // 2) + low if t == last_t: densities.sort(key=lambda x: (abs(x[0] - density), 256 - x[1])) return densities[0][1] else: last_t = t d = numpy.count_nonzero(img > t) / size densities.append((d, t)) if d < density: high = t elif d >= density: # search away from low low = t def filter_greys_using_image(image, target): """Filter out any values in target not in image :param image: image containing values to appear in filtered image :param target: the image to filter :rtype: 2d :class:`numpy.ndarray` containing only value in image and with the same dimensions as target """ maskbase = numpy.array(range(256), dtype=numpy.uint8) mask = numpy.where(numpy.in1d(maskbase, numpy.unique(image)), maskbase, 0) return mask[target] def correlation_coefficient_normed(template, image): h, w = template.shape[:2] H, W = image.shape[:2] template_size = template.size template_distance = template - (template.sum() / template_size) corr_num = numpy.zeros((H, W), numpy.float64) corr_denum = numpy.zeros((H, W), numpy.float64) for y in xrange(H): for x in xrange(W): image_in_template_area = image[y:y + h, x:x + w] image_distance_of_template_area = ( image_in_template_area - ( image_in_template_area.sum() / template_size ) ) I_H, I_W = image_distance_of_template_area.shape[:2] sum_of_template_by_image_distance_at_xy = ( image_distance_of_template_area * template_distance[:I_H, :I_W] ).sum() corr_num[y, x] = sum_of_template_by_image_distance_at_xy corr_denum[y, x] = numpy.sqrt( (template_distance ** 2).sum() * (image_distance_of_template_area ** 2).sum() ) print y return corr_num / corr_denum
ten10solutions/Geist
geist/vision.py
filter_greys_using_image
python
def filter_greys_using_image(image, target): """Filter out any values in target not in image :param image: image containing values to appear in filtered image :param target: the image to filter :rtype: 2d :class:`numpy.ndarray` containing only value in image and with the same dimensions as target """ maskbase = numpy.array(range(256), dtype=numpy.uint8) mask = numpy.where(numpy.in1d(maskbase, numpy.unique(image)), maskbase, 0) return mask[target]
Filter out any values in target not in image :param image: image containing values to appear in filtered image :param target: the image to filter :rtype: 2d :class:`numpy.ndarray` containing only value in image and with the same dimensions as target
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/vision.py#L488-L499
null
from __future__ import division from numpy.fft import irfft2, rfft2 import numpy import itertools import operator def subimage(rect, image): x, y, w, h = rect return image[y:y + h, x:x + w] def pad_bin_image_to_shape(image, shape): """ Padd image to size :shape: with zeros """ h, w = shape ih, iw = image.shape assert ih <= h assert iw <= w if iw < w: result = numpy.hstack((image, numpy.zeros((ih, w - iw), bool))) else: result = image if ih < h: result = numpy.vstack((result, numpy.zeros((h - ih, w), bool))) return result def sum_2d_images(images): it = iter(images) result = numpy.copy(next(it)) for image in it: result += image return result OVERLAP_TABLE = { 16: (16, 8, 4, 2, 1), 15: (15, 5, 3, 1), 12: (12, 6, 4, 3, 2, 1), 10: (10, 5, 2, 1), 9: (9, 3, 1), 8: (8, 4, 2, 1), 6: (6, 3, 2, 1), 5: (5, 1), 4: (4, 2, 1), 3: (3, 1), 2: (2, 1) } # A number near float max which we don't want to get near to keep precision # If you get false matches consider reducing this number. ACCURACY_LIMIT = 2 ** (64 - 23) def best_convolution(bin_template, bin_image, tollerance=0.5, overlap_table=OVERLAP_TABLE): """ Selects and applies the best convolution method to find template in image. Returns a list of matches in (width, height, x offset, y offset) format (where the x and y offsets are from the top left corner). As the images are binary images, we can utilise the extra bit space in the float64's by cutting the image into tiles and stacking them into variable grayscale values. This allows converting a sparse binary image into a dense(r) grayscale one. """ template_sum = numpy.count_nonzero(bin_template) th, tw = bin_template.shape ih, iw = bin_image.shape if template_sum == 0 or th == 0 or tw == 0: # If we don't have a template return [] if th > ih or tw > iw: # If the template is bigger than the image return [] # How many cells can we split the image into? max_vert_cells = ih // th max_hor_cells = iw // th # Try to work out how many times we can stack the image usable_factors = {n: factors for n, factors in overlap_table.iteritems() if ((template_sum + 1) ** (n)) < ACCURACY_LIMIT} overlap_options = [(factor, n // factor) for n, factors in usable_factors.iteritems() for factor in factors if (factor <= max_vert_cells and n // factor <= max_hor_cells)] if not overlap_options: # We can't stack the image return convolution(bin_template, bin_image, tollerance=tollerance) best_overlap = min(overlap_options, key=lambda x: ((ih // x[0] + th) * (iw // x[1] + tw))) return overlapped_convolution(bin_template, bin_image, tollerance=tollerance, splits=best_overlap) def convolution(bin_template, bin_image, tollerance=0.5): expected = numpy.count_nonzero(bin_template) ih, iw = bin_image.shape th, tw = bin_template.shape # Padd image to even dimensions if ih % 2 or iw % 2: if ih % 2: ih += 1 if iw % 2: iw += 1 bin_image = pad_bin_image_to_shape(bin_image, (ih, iw)) if expected == 0: return [] # Calculate the convolution of the FFT's of the image & template convolution_freqs = rfft2(bin_image) * rfft2(bin_template[::-1, ::-1], bin_image.shape) # Reverse the FFT to find the result image convolution_image = irfft2(convolution_freqs) # At this point, the maximum point in convolution_image should be the # bottom right (why?) of the area of greatest match # The areas in the result image within expected +- tollerance are where we # saw matches found_bitmap = ((convolution_image > (expected - tollerance)) & (convolution_image < (expected + tollerance))) match_points = numpy.transpose(numpy.nonzero(found_bitmap)) # bottom right # Find the top left point from the template (remember match_point is # inside the template (hence -1) return [((fx - (tw - 1)), (fy - (th - 1))) for (fy, fx) in match_points] def overlapped_convolution(bin_template, bin_image, tollerance=0.5, splits=(4, 2)): """ As each of these images are hold only binary values, and RFFT2 works on float64 greyscale values, we can make the convolution more efficient by breaking the image up into :splits: sectons. Each one of these sections then has its greyscale value adjusted and then stacked. We then apply the convolution to this 'stack' of images, and adjust the resultant position matches. """ th, tw = bin_template.shape ih, iw = bin_image.shape hs, ws = splits h = ih // hs w = iw // ws count = numpy.count_nonzero(bin_template) assert count > 0 assert h >= th assert w >= tw yoffset = [(i * h, ((i + 1) * h) + (th - 1)) for i in range(hs)] xoffset = [(i * w, ((i + 1) * w) + (tw - 1)) for i in range(ws)] # image_stacks is Origin (x,y), array, z (height in stack) image_stacks = [((x1, y1), bin_image[y1:y2, x1:x2], float((count + 1) ** (num))) for num, (x1, x2, y1, y2) in enumerate((x1, x2, y1, y2) for (x1, x2) in xoffset for (y1, y2) in yoffset)] pad_h = max(i.shape[0] for _, i, _ in image_stacks) pad_w = max(i.shape[1] for _, i, _ in image_stacks) # rfft metrics must be an even size - why ... maths? pad_w += pad_w % 2 pad_h += pad_h % 2 overlapped_image = sum_2d_images(pad_bin_image_to_shape(i, (pad_h, pad_w)) * num for _, i, num in image_stacks) #print "Overlap splits %r, Image Size (%d,%d), #Overlapped Size (%d,%d)" % (splits,iw,ih,pad_w,pad_h) # Calculate the convolution of the FFT's of the overlapped image & template convolution_freqs = (rfft2(overlapped_image) * rfft2(bin_template[::-1, ::-1], overlapped_image.shape)) # Reverse the FFT to find the result overlapped image convolution_image = irfft2(convolution_freqs) # At this point, the maximum point in convolution_image should be the # bottom right (why?) of the area of greatest match results = set() for (x, y), _, num in image_stacks[::-1]: test = convolution_image / num filtered = ((test >= (count - tollerance)) & (test <= (count + tollerance))) match_points = numpy.transpose(numpy.nonzero(filtered)) # bottom right for (fy, fx) in match_points: if fx < (tw - 1) or fy < (th - 1): continue results.add((x + fx - (tw - 1), y + fy - (th - 1))) convolution_image %= num return list(results) def get_possible_convolution_regions(bin_template, bin_image, tollerance=0.5, rescale=10): result = [] h, w = bin_template.shape[:2] rects = set((x, y, w, h) for x, y in convolution_r3m_targets( bin_template, bin_image, tollerance)) ih, iw = bin_image.shape[:2] # areas of interest aoi = numpy.zeros((ih // rescale + 1, iw // rescale + 1), bool) for x, y, w, h in rects: aoi[y // rescale:(y + h) // rescale + 1, x // rescale:(x + w) // rescale + 1] = True bp = binary_partition_image(aoi, min_w=w // rescale, min_h=h // rescale) if bp: bp = prune_unbeneficial_partitions(aoi, bp) return result def binary_partition_to_rects(bp, image, template_w, template_h, xoffset=0, yoffset=0): h, w = image.shape[2:] if bp is None: return [(xoffset, yoffset, w, h)] pos, axis, (p1, p2) = bp if axis == 0: new_xoffset, new_yoffset = xoffset, yoffset + pos i1, i2 = image[pos:], image[:pos] else: new_xoffset, new_yoffset = xoffset + pos, yoffset i1, i2 = image[:, pos:], image[:, :pos] def prune_unbeneficial_partitions(image, bp): pos, axis, (p1, p2) = bp if axis == 0: i1, i2 = image[pos:], image[:pos] else: i1, i2 = image[:, pos:], image[:, :pos] if p1 is None: p1_result = numpy.count_nonzero(i1) == 0 else: p1_result = prune_unbeneficial_partitions(i1, p1) if p2 is None: p2_result = numpy.count_nonzero(i2) == 0 else: p2_result = prune_unbeneficial_partitions(i2, p2) if p1_result or p2_result: return [pos, axis, [None if p1_result in [True, False] else p1_result, None if p2_result in [True, False] else p2_result]] else: return None def get_partition_scores(image, min_w=1, min_h=1): """Return list of best to worst binary splits along the x and y axis. """ h, w = image.shape[:2] if w == 0 or h == 0: return [] area = h * w cnz = numpy.count_nonzero total = cnz(image) if total == 0 or area == total: return [] if h < min_h * 2: y_c = [] else: y_c = [(-abs((count / ((h - y) * w)) - ((total - count) / (y * w))), y, 0) for count, y in ((cnz(image[y:]), y) for y in range(min_h, image.shape[0] - min_h))] if w < min_w * 2: x_c = [] else: x_c = [(-abs((count / (h * (w - x))) - ((total - count) / (h * x))), x, 1) for count, x in ((cnz(image[:, x:]), x) for x in range(min_w, image.shape[1] - min_w))] return sorted(x_c + y_c) def get_best_partition(image, min_w=1, min_h=1): partitions = get_partition_scores(image, min_w=min_w, min_h=min_h) if partitions: return partitions[0][-2:] else: return None def binary_partition_image(image, min_w=1, min_h=1, depth=0, max_depth=-1): """Return a bsp of [pos, axis, [before_node, after_node]] nodes where leaf nodes == None. If max_depth < 0 this function will continue until all leaf nodes have been found, if it is >= 0 leaf nodes will be created at that depth. min_w and min_h are the minimum width or height of a partition. """ if max_depth >= 0 and depth >= max_depth: return None partition = get_best_partition(image, min_w=min_w, min_h=min_h) if partition is None: return None pos, axis = partition if axis == 0: p1 = binary_partition_image( image[pos:], min_w, min_h, depth + 1, max_depth) p2 = binary_partition_image( image[:pos], min_w, min_h, depth + 1, max_depth) elif axis == 1: p1 = binary_partition_image( image[:, pos:], min_w, min_h, depth + 1, max_depth) p2 = binary_partition_image( image[:, :pos], min_w, min_h, depth + 1, max_depth) return [pos, axis, [p1, p2]] def draw_binary_partition(image, subdiv, res_image=None, counter=None): if res_image is None: res_image = numpy.zeros(image.shape) if counter is None: counter = itertools.count(15) pos, axis, (p1, p2) = subdiv if axis == 0: s1, s2 = res_image[pos:], res_image[:pos] i1, i2 = image[pos:], image[:pos] else: s1, s2 = res_image[:, pos:], res_image[:, :pos] i1, i2 = image[:, pos:], image[:, :pos] if p1 is None: if numpy.count_nonzero(i1): s1[:] = next(counter) else: draw_binary_partition(i1, p1, s1, counter) if p2 is None: if numpy.count_nonzero(i2): s2[:] = next(counter) else: draw_binary_partition(i2, p2, s2, counter) return res_image def rescale2avg(image): sub1 = image[:-1:2, :-1:2] sub2 = image[1::2, 1::2] res = numpy.zeros(sub1.shape, numpy.uint32) res += sub1 res += sub2 res /= 2 return res.astype(numpy.uint8) def rescale2max(image): sub1 = image[:-1:2, :-1:2] sub2 = image[1::2, 1::2] if len(image.shape) == 3: max_map = grey_scale(sub1) > grey_scale(sub2) else: max_map = sub1 > sub2 inv_max_map = max_map == False res = numpy.zeros(sub1.shape, numpy.uint8) res[max_map] = sub1[max_map] res[inv_max_map] = sub2[inv_max_map] return res def rescale3avg(image): sub1 = image[:-2:3, :-2:3] sub2 = image[1:-1:3, 1:-1:3] sub3 = image[2::3, 2::3] res = numpy.zeros(sub1.shape, numpy.uint32) res += sub1 res += sub2 res += sub3 res /= 3 return res.astype(numpy.uint8) def rescale3max(image): sub1 = image[:-2:3, :-2:3] sub2 = image[1:-1:3, 1:-1:3] sub3 = image[2::3, 2::3] if len(image.shape) == 3: grey1 = grey_scale(sub1) grey2 = grey_scale(sub2) grey3 = grey_scale(sub3) else: grey1, grey2, grey3 = sub1, sub2, sub3 max_map_1 = (grey1 > grey2) & (grey1 > grey3) max_map_2 = (grey2 > grey1) & (grey2 > grey3) max_map_3 = (max_map_1 | max_map_2) == False res = numpy.zeros(sub1.shape, numpy.uint8) res[max_map_1] = sub1[max_map_1] res[max_map_2] = sub2[max_map_2] res[max_map_3] = sub3[max_map_3] return res def or_reduce_rescale3max_offset(image): return reduce(operator.or_, (rescale3max(image[y1:y2, x1:x2]) for (y1, y2), (x1, x2) in itertools.product( *[[(i, -(3 - i)) for i in range(3)]] * 2) ) ) def numpy_or_all(images): it = iter(images) result = numpy.copy(next(it)) for image in it: result |= image return result def grey_scale(image): """Converts RGB image to Greyscale :param image: input image :type image: 3 channel 3d :class:`numpy.ndarray` :rtype: :class:`numpy.ndarray` """ return image.astype(numpy.int32).sum(axis=2) // 3 def find_edges(image): base = image[:, :, numpy.newaxis].astype(numpy.int16) c = base[1:-1, 1:-1] e = base[1:-1, :-2] w = base[1:-1, 2:] s = base[:-2, 1:-1] n = base[2:, 1:-1] diffs = numpy.concatenate([c-e, c-w, c-s, c-n], axis=2).max(axis=2) diffs[diffs < 0] = 0 dh, dw = diffs.shape col = numpy.zeros((dh, 1), numpy.uint8) row = numpy.zeros((1, dw+2), numpy.uint8) return numpy.vstack( (row, numpy.hstack((col, diffs.astype(numpy.uint8), col)), row) ) def find_threshold_near_density(img, density, low=0, high=255): """Find a threshold where the fraction of pixels above the threshold is closest to density where density is (count of pixels above threshold / count of pixels). The highest threshold closest to the desired density will be returned. Use low and high to exclude undesirable thresholds. :param img: target image :type img: 2d :class:`numpy.ndarray` :param density: target density :type density: float between 0.0 and 1.0 :param low: min threshold to test :type low: ubyte :param migh: max threshold to test :type low: ubyte :rtype: ubyte """ size = numpy.size(img) densities = [] last_t = None while True: t = ((high - low) // 2) + low if t == last_t: densities.sort(key=lambda x: (abs(x[0] - density), 256 - x[1])) return densities[0][1] else: last_t = t d = numpy.count_nonzero(img > t) / size densities.append((d, t)) if d < density: high = t elif d >= density: # search away from low low = t def filter_greys_using_image(image, target): """Filter out any values in target not in image :param image: image containing values to appear in filtered image :param target: the image to filter :rtype: 2d :class:`numpy.ndarray` containing only value in image and with the same dimensions as target """ maskbase = numpy.array(range(256), dtype=numpy.uint8) mask = numpy.where(numpy.in1d(maskbase, numpy.unique(image)), maskbase, 0) return mask[target] def correlation_coefficient_normed(template, image): h, w = template.shape[:2] H, W = image.shape[:2] template_size = template.size template_distance = template - (template.sum() / template_size) corr_num = numpy.zeros((H, W), numpy.float64) corr_denum = numpy.zeros((H, W), numpy.float64) for y in xrange(H): for x in xrange(W): image_in_template_area = image[y:y + h, x:x + w] image_distance_of_template_area = ( image_in_template_area - ( image_in_template_area.sum() / template_size ) ) I_H, I_W = image_distance_of_template_area.shape[:2] sum_of_template_by_image_distance_at_xy = ( image_distance_of_template_area * template_distance[:I_H, :I_W] ).sum() corr_num[y, x] = sum_of_template_by_image_distance_at_xy corr_denum[y, x] = numpy.sqrt( (template_distance ** 2).sum() * (image_distance_of_template_area ** 2).sum() ) print y return corr_num / corr_denum
ten10solutions/Geist
geist/match_position_finder_helpers.py
find_potential_match_regions
python
def find_potential_match_regions(template, transformed_array, method='correlation', raw_tolerance=0.666): if method == 'correlation': match_value = np.sum(template**2) # this will be the value of the match in the elif method == 'squared difference': match_value = 0 elif method == 'correlation coefficient': temp_minus_mean = template - np.mean(template) match_value = np.sum(temp_minus_mean**2) else: raise ValueError('Matching method not implemented') condition = ((np.round(transformed_array, decimals=3)>=match_value*raw_tolerance) & (np.round(transformed_array, decimals=3)<=match_value*(1./raw_tolerance))) return np.transpose(condition.nonzero())
To prevent prohibitively slow calculation of normalisation coefficient at each point in image find potential match points, and normalise these only these. This function uses the definitions of the matching functions to calculate the expected match value and finds positions in the transformed array matching these- normalisation will then eliminate false positives
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/match_position_finder_helpers.py#L4-L21
null
import numpy as np # trsnposition and omparison above take most time # correlation coefficient matches at top left- perfect for tiling # correlation matches to bottom right- so requires transformation for tiling def get_tiles_at_potential_match_regions(image, template, transformed_array, method='correlation', raw_tolerance=0.001): if method not in ['correlation', 'correlation coefficient', 'squared difference']: raise ValueError('Matching method not implemented') h, w = template.shape match_points = find_potential_match_regions(template, transformed_array, method=method, raw_tolerance=raw_tolerance) match_points = [(match[0], match[1]) for match in match_points] # create tile for each match point- use dict so we know which match point it applies to # match point here is position of top left pixel of tile image_tiles_dict = {match_points[i]:image[match_points[i][0]:match_points[i][0]+h,match_points[i][1]:match_points[i][1]+w] for i in range(len(match_points))} return image_tiles_dict ############################################### # image tiles dict is of form match_point coord:tile at that point def normalise_correlation(image_tile_dict, transformed_array, template, normed_tolerance=1): """Calculates the normalisation coefficients of potential match positions Then normalises the correlation at these positions, and returns them if they do indeed constitute a match """ template_norm = np.linalg.norm(template) image_norms = {(x,y):np.linalg.norm(image_tile_dict[(x,y)])*template_norm for (x,y) in image_tile_dict.keys()} match_points = image_tile_dict.keys() # for correlation, then need to transofrm back to get correct value for division h, w = template.shape #points_from_transformed_array = [(match[0] + h - 1, match[1] + w - 1) for match in match_points] image_matches_normalised = {match_points[i]:transformed_array[match_points[i][0], match_points[i][1]]/image_norms[match_points[i]] for i in range(len(match_points))} result = {key:value for key, value in image_matches_normalised.items() if np.round(value, decimals=3) >= normed_tolerance} return result.keys() # image tiles dict is of form match_point coord:tile at that point def normalise_correlation_coefficient(image_tile_dict, transformed_array, template, normed_tolerance=1): """As above, but for when the correlation coefficient matching method is used """ template_mean = np.mean(template) template_minus_mean = template - template_mean template_norm = np.linalg.norm(template_minus_mean) image_norms = {(x,y):np.linalg.norm(image_tile_dict[(x,y)]- np.mean(image_tile_dict[(x,y)]))*template_norm for (x,y) in image_tile_dict.keys()} match_points = image_tile_dict.keys() # for correlation, then need to transofrm back to get correct value for division h, w = template.shape image_matches_normalised = {match_points[i]:transformed_array[match_points[i][0], match_points[i][1]]/image_norms[match_points[i]] for i in range(len(match_points))} normalised_matches = {key:value for key, value in image_matches_normalised.items() if np.round(value, decimals=3) >= normed_tolerance} return normalised_matches.keys() def calculate_squared_differences(image_tile_dict, transformed_array, template, sq_diff_tolerance=0.1): """As above, but for when the squared differences matching method is used """ template_norm_squared = np.sum(template**2) image_norms_squared = {(x,y):np.sum(image_tile_dict[(x,y)]**2) for (x,y) in image_tile_dict.keys()} match_points = image_tile_dict.keys() # for correlation, then need to transofrm back to get correct value for division h, w = template.shape image_matches_normalised = {match_points[i]:-2*transformed_array[match_points[i][0], match_points[i][1]] + image_norms_squared[match_points[i]] + template_norm_squared for i in range(len(match_points))} #print image_matches_normalised cutoff = h*w*255**2*sq_diff_tolerance normalised_matches = {key:value for key, value in image_matches_normalised.items() if np.round(value, decimals=3) <= cutoff} return normalised_matches.keys()
ten10solutions/Geist
geist/match_position_finder_helpers.py
normalise_correlation
python
def normalise_correlation(image_tile_dict, transformed_array, template, normed_tolerance=1): template_norm = np.linalg.norm(template) image_norms = {(x,y):np.linalg.norm(image_tile_dict[(x,y)])*template_norm for (x,y) in image_tile_dict.keys()} match_points = image_tile_dict.keys() # for correlation, then need to transofrm back to get correct value for division h, w = template.shape #points_from_transformed_array = [(match[0] + h - 1, match[1] + w - 1) for match in match_points] image_matches_normalised = {match_points[i]:transformed_array[match_points[i][0], match_points[i][1]]/image_norms[match_points[i]] for i in range(len(match_points))} result = {key:value for key, value in image_matches_normalised.items() if np.round(value, decimals=3) >= normed_tolerance} return result.keys()
Calculates the normalisation coefficients of potential match positions Then normalises the correlation at these positions, and returns them if they do indeed constitute a match
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/match_position_finder_helpers.py#L44-L57
null
import numpy as np def find_potential_match_regions(template, transformed_array, method='correlation', raw_tolerance=0.666): """To prevent prohibitively slow calculation of normalisation coefficient at each point in image find potential match points, and normalise these only these. This function uses the definitions of the matching functions to calculate the expected match value and finds positions in the transformed array matching these- normalisation will then eliminate false positives """ if method == 'correlation': match_value = np.sum(template**2) # this will be the value of the match in the elif method == 'squared difference': match_value = 0 elif method == 'correlation coefficient': temp_minus_mean = template - np.mean(template) match_value = np.sum(temp_minus_mean**2) else: raise ValueError('Matching method not implemented') condition = ((np.round(transformed_array, decimals=3)>=match_value*raw_tolerance) & (np.round(transformed_array, decimals=3)<=match_value*(1./raw_tolerance))) return np.transpose(condition.nonzero())# trsnposition and omparison above take most time # correlation coefficient matches at top left- perfect for tiling # correlation matches to bottom right- so requires transformation for tiling def get_tiles_at_potential_match_regions(image, template, transformed_array, method='correlation', raw_tolerance=0.001): if method not in ['correlation', 'correlation coefficient', 'squared difference']: raise ValueError('Matching method not implemented') h, w = template.shape match_points = find_potential_match_regions(template, transformed_array, method=method, raw_tolerance=raw_tolerance) match_points = [(match[0], match[1]) for match in match_points] # create tile for each match point- use dict so we know which match point it applies to # match point here is position of top left pixel of tile image_tiles_dict = {match_points[i]:image[match_points[i][0]:match_points[i][0]+h,match_points[i][1]:match_points[i][1]+w] for i in range(len(match_points))} return image_tiles_dict ############################################### # image tiles dict is of form match_point coord:tile at that point # image tiles dict is of form match_point coord:tile at that point def normalise_correlation_coefficient(image_tile_dict, transformed_array, template, normed_tolerance=1): """As above, but for when the correlation coefficient matching method is used """ template_mean = np.mean(template) template_minus_mean = template - template_mean template_norm = np.linalg.norm(template_minus_mean) image_norms = {(x,y):np.linalg.norm(image_tile_dict[(x,y)]- np.mean(image_tile_dict[(x,y)]))*template_norm for (x,y) in image_tile_dict.keys()} match_points = image_tile_dict.keys() # for correlation, then need to transofrm back to get correct value for division h, w = template.shape image_matches_normalised = {match_points[i]:transformed_array[match_points[i][0], match_points[i][1]]/image_norms[match_points[i]] for i in range(len(match_points))} normalised_matches = {key:value for key, value in image_matches_normalised.items() if np.round(value, decimals=3) >= normed_tolerance} return normalised_matches.keys() def calculate_squared_differences(image_tile_dict, transformed_array, template, sq_diff_tolerance=0.1): """As above, but for when the squared differences matching method is used """ template_norm_squared = np.sum(template**2) image_norms_squared = {(x,y):np.sum(image_tile_dict[(x,y)]**2) for (x,y) in image_tile_dict.keys()} match_points = image_tile_dict.keys() # for correlation, then need to transofrm back to get correct value for division h, w = template.shape image_matches_normalised = {match_points[i]:-2*transformed_array[match_points[i][0], match_points[i][1]] + image_norms_squared[match_points[i]] + template_norm_squared for i in range(len(match_points))} #print image_matches_normalised cutoff = h*w*255**2*sq_diff_tolerance normalised_matches = {key:value for key, value in image_matches_normalised.items() if np.round(value, decimals=3) <= cutoff} return normalised_matches.keys()
ten10solutions/Geist
geist/match_position_finder_helpers.py
normalise_correlation_coefficient
python
def normalise_correlation_coefficient(image_tile_dict, transformed_array, template, normed_tolerance=1): template_mean = np.mean(template) template_minus_mean = template - template_mean template_norm = np.linalg.norm(template_minus_mean) image_norms = {(x,y):np.linalg.norm(image_tile_dict[(x,y)]- np.mean(image_tile_dict[(x,y)]))*template_norm for (x,y) in image_tile_dict.keys()} match_points = image_tile_dict.keys() # for correlation, then need to transofrm back to get correct value for division h, w = template.shape image_matches_normalised = {match_points[i]:transformed_array[match_points[i][0], match_points[i][1]]/image_norms[match_points[i]] for i in range(len(match_points))} normalised_matches = {key:value for key, value in image_matches_normalised.items() if np.round(value, decimals=3) >= normed_tolerance} return normalised_matches.keys()
As above, but for when the correlation coefficient matching method is used
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/match_position_finder_helpers.py#L61-L73
null
import numpy as np def find_potential_match_regions(template, transformed_array, method='correlation', raw_tolerance=0.666): """To prevent prohibitively slow calculation of normalisation coefficient at each point in image find potential match points, and normalise these only these. This function uses the definitions of the matching functions to calculate the expected match value and finds positions in the transformed array matching these- normalisation will then eliminate false positives """ if method == 'correlation': match_value = np.sum(template**2) # this will be the value of the match in the elif method == 'squared difference': match_value = 0 elif method == 'correlation coefficient': temp_minus_mean = template - np.mean(template) match_value = np.sum(temp_minus_mean**2) else: raise ValueError('Matching method not implemented') condition = ((np.round(transformed_array, decimals=3)>=match_value*raw_tolerance) & (np.round(transformed_array, decimals=3)<=match_value*(1./raw_tolerance))) return np.transpose(condition.nonzero())# trsnposition and omparison above take most time # correlation coefficient matches at top left- perfect for tiling # correlation matches to bottom right- so requires transformation for tiling def get_tiles_at_potential_match_regions(image, template, transformed_array, method='correlation', raw_tolerance=0.001): if method not in ['correlation', 'correlation coefficient', 'squared difference']: raise ValueError('Matching method not implemented') h, w = template.shape match_points = find_potential_match_regions(template, transformed_array, method=method, raw_tolerance=raw_tolerance) match_points = [(match[0], match[1]) for match in match_points] # create tile for each match point- use dict so we know which match point it applies to # match point here is position of top left pixel of tile image_tiles_dict = {match_points[i]:image[match_points[i][0]:match_points[i][0]+h,match_points[i][1]:match_points[i][1]+w] for i in range(len(match_points))} return image_tiles_dict ############################################### # image tiles dict is of form match_point coord:tile at that point def normalise_correlation(image_tile_dict, transformed_array, template, normed_tolerance=1): """Calculates the normalisation coefficients of potential match positions Then normalises the correlation at these positions, and returns them if they do indeed constitute a match """ template_norm = np.linalg.norm(template) image_norms = {(x,y):np.linalg.norm(image_tile_dict[(x,y)])*template_norm for (x,y) in image_tile_dict.keys()} match_points = image_tile_dict.keys() # for correlation, then need to transofrm back to get correct value for division h, w = template.shape #points_from_transformed_array = [(match[0] + h - 1, match[1] + w - 1) for match in match_points] image_matches_normalised = {match_points[i]:transformed_array[match_points[i][0], match_points[i][1]]/image_norms[match_points[i]] for i in range(len(match_points))} result = {key:value for key, value in image_matches_normalised.items() if np.round(value, decimals=3) >= normed_tolerance} return result.keys() # image tiles dict is of form match_point coord:tile at that point def calculate_squared_differences(image_tile_dict, transformed_array, template, sq_diff_tolerance=0.1): """As above, but for when the squared differences matching method is used """ template_norm_squared = np.sum(template**2) image_norms_squared = {(x,y):np.sum(image_tile_dict[(x,y)]**2) for (x,y) in image_tile_dict.keys()} match_points = image_tile_dict.keys() # for correlation, then need to transofrm back to get correct value for division h, w = template.shape image_matches_normalised = {match_points[i]:-2*transformed_array[match_points[i][0], match_points[i][1]] + image_norms_squared[match_points[i]] + template_norm_squared for i in range(len(match_points))} #print image_matches_normalised cutoff = h*w*255**2*sq_diff_tolerance normalised_matches = {key:value for key, value in image_matches_normalised.items() if np.round(value, decimals=3) <= cutoff} return normalised_matches.keys()
ten10solutions/Geist
geist/match_position_finder_helpers.py
calculate_squared_differences
python
def calculate_squared_differences(image_tile_dict, transformed_array, template, sq_diff_tolerance=0.1): template_norm_squared = np.sum(template**2) image_norms_squared = {(x,y):np.sum(image_tile_dict[(x,y)]**2) for (x,y) in image_tile_dict.keys()} match_points = image_tile_dict.keys() # for correlation, then need to transofrm back to get correct value for division h, w = template.shape image_matches_normalised = {match_points[i]:-2*transformed_array[match_points[i][0], match_points[i][1]] + image_norms_squared[match_points[i]] + template_norm_squared for i in range(len(match_points))} #print image_matches_normalised cutoff = h*w*255**2*sq_diff_tolerance normalised_matches = {key:value for key, value in image_matches_normalised.items() if np.round(value, decimals=3) <= cutoff} return normalised_matches.keys()
As above, but for when the squared differences matching method is used
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/match_position_finder_helpers.py#L77-L89
null
import numpy as np def find_potential_match_regions(template, transformed_array, method='correlation', raw_tolerance=0.666): """To prevent prohibitively slow calculation of normalisation coefficient at each point in image find potential match points, and normalise these only these. This function uses the definitions of the matching functions to calculate the expected match value and finds positions in the transformed array matching these- normalisation will then eliminate false positives """ if method == 'correlation': match_value = np.sum(template**2) # this will be the value of the match in the elif method == 'squared difference': match_value = 0 elif method == 'correlation coefficient': temp_minus_mean = template - np.mean(template) match_value = np.sum(temp_minus_mean**2) else: raise ValueError('Matching method not implemented') condition = ((np.round(transformed_array, decimals=3)>=match_value*raw_tolerance) & (np.round(transformed_array, decimals=3)<=match_value*(1./raw_tolerance))) return np.transpose(condition.nonzero())# trsnposition and omparison above take most time # correlation coefficient matches at top left- perfect for tiling # correlation matches to bottom right- so requires transformation for tiling def get_tiles_at_potential_match_regions(image, template, transformed_array, method='correlation', raw_tolerance=0.001): if method not in ['correlation', 'correlation coefficient', 'squared difference']: raise ValueError('Matching method not implemented') h, w = template.shape match_points = find_potential_match_regions(template, transformed_array, method=method, raw_tolerance=raw_tolerance) match_points = [(match[0], match[1]) for match in match_points] # create tile for each match point- use dict so we know which match point it applies to # match point here is position of top left pixel of tile image_tiles_dict = {match_points[i]:image[match_points[i][0]:match_points[i][0]+h,match_points[i][1]:match_points[i][1]+w] for i in range(len(match_points))} return image_tiles_dict ############################################### # image tiles dict is of form match_point coord:tile at that point def normalise_correlation(image_tile_dict, transformed_array, template, normed_tolerance=1): """Calculates the normalisation coefficients of potential match positions Then normalises the correlation at these positions, and returns them if they do indeed constitute a match """ template_norm = np.linalg.norm(template) image_norms = {(x,y):np.linalg.norm(image_tile_dict[(x,y)])*template_norm for (x,y) in image_tile_dict.keys()} match_points = image_tile_dict.keys() # for correlation, then need to transofrm back to get correct value for division h, w = template.shape #points_from_transformed_array = [(match[0] + h - 1, match[1] + w - 1) for match in match_points] image_matches_normalised = {match_points[i]:transformed_array[match_points[i][0], match_points[i][1]]/image_norms[match_points[i]] for i in range(len(match_points))} result = {key:value for key, value in image_matches_normalised.items() if np.round(value, decimals=3) >= normed_tolerance} return result.keys() # image tiles dict is of form match_point coord:tile at that point def normalise_correlation_coefficient(image_tile_dict, transformed_array, template, normed_tolerance=1): """As above, but for when the correlation coefficient matching method is used """ template_mean = np.mean(template) template_minus_mean = template - template_mean template_norm = np.linalg.norm(template_minus_mean) image_norms = {(x,y):np.linalg.norm(image_tile_dict[(x,y)]- np.mean(image_tile_dict[(x,y)]))*template_norm for (x,y) in image_tile_dict.keys()} match_points = image_tile_dict.keys() # for correlation, then need to transofrm back to get correct value for division h, w = template.shape image_matches_normalised = {match_points[i]:transformed_array[match_points[i][0], match_points[i][1]]/image_norms[match_points[i]] for i in range(len(match_points))} normalised_matches = {key:value for key, value in image_matches_normalised.items() if np.round(value, decimals=3) >= normed_tolerance} return normalised_matches.keys()
ten10solutions/Geist
geist/backends/_x11_common.py
GeistXBase.create_process
python
def create_process(self, command, shell=True, stdout=None, stderr=None, env=None): env = env if env is not None else dict(os.environ) env['DISPLAY'] = self.display return subprocess.Popen(command, shell=shell, stdout=stdout, stderr=stderr, env=env)
Execute a process using subprocess.Popen, setting the backend's DISPLAY
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/backends/_x11_common.py#L54-L63
null
class GeistXBase(object): KEY_NAME_TO_CODE = keysyms KEY_NAME_TO_CODE_IGNORE_CASE = {name.lower(): value for name, value in keysyms.iteritems()} def __init__(self, **kwargs): display = kwargs.get('display', ':0') self._display = display self._conn = ooxcb.connect(display) self._root = self._conn.setup.roots[self._conn.pref_screen].root @property def display(self): return self._display def actions_transaction(self): return _ActionsTransaction(self) def _get_key_code_from_name(self, name): if name == 'shift': symb = GeistXBase.KEY_NAME_TO_CODE['Shift_L'] elif name in GeistXBase.KEY_NAME_TO_CODE: symb = GeistXBase.KEY_NAME_TO_CODE[name] elif name.lower() in GeistXBase.KEY_NAME_TO_CODE_IGNORE_CASE: symb = GeistXBase.KEY_NAME_TO_CODE_IGNORE_CASE[name] else: raise ValueError('unhandled key %r' % (name,)) return self._conn.keysyms.get_keycode(symb) def key_down(self, name): key_code = self._get_key_code_from_name(name) with self._conn.bunch(): self._conn.xtest.fake_input_checked( KeyPress, detail=key_code ) def key_up(self, name): key_code = self._get_key_code_from_name(name) with self._conn.bunch(): self._conn.xtest.fake_input_checked( KeyRelease, detail=key_code ) def button_down(self, button_num): with self._conn.bunch(): self._conn.xtest.fake_input_checked( ButtonPress, detail=button_num ) def button_up(self, button_num): with self._conn.bunch(): self._conn.xtest.fake_input_checked( ButtonRelease, detail=button_num ) def move(self, point): x, y = point with self._conn.bunch(): self._conn.xtest.fake_input_checked( MotionNotify, rootX=x, rootY=y, ) def cursor_position(self): reply = self._root.query_pointer().reply() return reply.root_x, reply.root_y def close(self): if hasattr(self, '_conn'): self._conn.disconnect() del self._conn def __del__(self): self.close()
ten10solutions/Geist
geist/matchers.py
match_via_correlation
python
def match_via_correlation(image, template, raw_tolerance=1, normed_tolerance=0.9): h, w = image.shape th, tw = template.shape # fft based convolution enables fast matching of large images correlation = fftconvolve(image, template[::-1,::-1]) # trim the returned image, fftconvolve returns an image of width: (Temp_w-1) + Im_w + (Temp_w -1), likewise height correlation = correlation[th-1:h, tw-1:w] # find images regions which are potentially matches match_position_dict = get_tiles_at_potential_match_regions(image, template, correlation, raw_tolerance=raw_tolerance) # bright spots in images can lead to false positivies- the normalisation carried out here eliminates those results = normalise_correlation(match_position_dict, correlation, template, normed_tolerance=normed_tolerance) return results
Matchihng algorithm based on normalised cross correlation. Using this matching prevents false positives occuring for bright patches in the image
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/matchers.py#L8-L22
[ "def get_tiles_at_potential_match_regions(image, template, transformed_array, method='correlation', raw_tolerance=0.001):\n if method not in ['correlation', 'correlation coefficient', 'squared difference']:\n raise ValueError('Matching method not implemented')\n h, w = template.shape\n match_points = find_potential_match_regions(template, transformed_array, method=method, raw_tolerance=raw_tolerance)\n match_points = [(match[0], match[1]) for match in match_points]\n # create tile for each match point- use dict so we know which match point it applies to\n # match point here is position of top left pixel of tile\n image_tiles_dict = {match_points[i]:image[match_points[i][0]:match_points[i][0]+h,match_points[i][1]:match_points[i][1]+w] for i in range(len(match_points))}\n return image_tiles_dict\n", "def normalise_correlation(image_tile_dict, transformed_array, template, normed_tolerance=1):\n \"\"\"Calculates the normalisation coefficients of potential match positions\n Then normalises the correlation at these positions, and returns them\n if they do indeed constitute a match\n \"\"\"\n template_norm = np.linalg.norm(template)\n image_norms = {(x,y):np.linalg.norm(image_tile_dict[(x,y)])*template_norm for (x,y) in image_tile_dict.keys()}\n match_points = image_tile_dict.keys()\n # for correlation, then need to transofrm back to get correct value for division\n h, w = template.shape\n #points_from_transformed_array = [(match[0] + h - 1, match[1] + w - 1) for match in match_points]\n image_matches_normalised = {match_points[i]:transformed_array[match_points[i][0], match_points[i][1]]/image_norms[match_points[i]] for i in range(len(match_points))}\n result = {key:value for key, value in image_matches_normalised.items() if np.round(value, decimals=3) >= normed_tolerance}\n return result.keys()\n" ]
from .match_position_finder_helpers import get_tiles_at_potential_match_regions, normalise_correlation, normalise_correlation_coefficient, find_potential_match_regions from scipy.signal import fftconvolve from scipy.ndimage.measurements import label, find_objects import numpy as np # both these methods return array of points giving bottom right coordinate of match def match_via_squared_difference(image, template, raw_tolerance=1, sq_diff_tolerance=0.1): """ Matchihng algorithm based on normalised cross correlation. Using this matching prevents false positives occuring for bright patches in the image """ h, w = image.shape th, tw = template.shape # fft based convolution enables fast matching of large images correlation = fftconvolve(image, template[::-1,::-1]) # trim the returned image, fftconvolve returns an image of width: (Temp_w-1) + Im_w + (Temp_w -1), likewise height correlation = correlation[th-1:h, tw-1:w] # find images regions which are potentially matches match_position_dict = get_tiles_at_potential_match_regions(image, template, correlation, raw_tolerance=raw_tolerance) # bright spots in images can lead to false positivies- the normalisation carried out here eliminates those results = calculate_squared_differences(match_position_dict, correlation, template, sq_diff_tolerance=sq_diff_tolerance) return results def match_via_correlation_coefficient(image, template, raw_tolerance=1, normed_tolerance=0.9): """ Matching algorithm based on 2-dimensional version of Pearson product-moment correlation coefficient. This is more robust in the case where the match might be scaled or slightly rotated. From experimentation, this method is less prone to false positives than the correlation method. """ h, w = image.shape th, tw = template.shape temp_mean = np.mean(template) temp_minus_mean = template - temp_mean convolution = fftconvolve(image, temp_minus_mean[::-1,::-1]) convolution = convolution[th-1:h, tw-1:w] match_position_dict = get_tiles_at_potential_match_regions(image, template, convolution, method='correlation coefficient', raw_tolerance=raw_tolerance) # this is empty, so think condition is wrong results = normalise_correlation_coefficient(match_position_dict, convolution, template, normed_tolerance=normed_tolerance) return results def fuzzy_match(image, template, normed_tolerance=None, raw_tolerance=None, method='correlation'): """Determines, using one of two methods, whether a match(es) is present and returns the positions of the bottom right corners of the matches. Fuzzy matches returns regions, so the center of each region is returned as the final match location USE THIS FUNCTION IF you need to match, e.g. the same image but rendered slightly different with respect to anti aliasing; the same image on a number of different backgrounds. The method is the name of the matching method used, the details of this do not matter. Use the default method unless you have too many false positives, in this case, use the method 'correlation coefficient.' The correlation coefficient method can also be more robust at matching when the match might not be exact. The raw_tolerance is the proportion of the value at match positions (i.e. the value returned for an exact match) that we count as a match. For fuzzy matching, this value will not be exactly the value returned for an exact match N. B. Lowering raw_tolerance increases the number of potential match tiles requiring normalisation. This DRAMATICALLY slows down matching as normalisation (a process which eliminates false positives) The normed_tolerance is how far a potential match value can differ from one after normalisation. The tolerance values indicated below are from a short investigation, looking to minimise missing items we wish to match, as all as false positives which inevitably occur when performing fuzzy matching. To generate these values, we tested maching letters with different type of antialiasing on a number of backgrounds. """ if method == 'correlation': if not raw_tolerance: raw_tolerance = 0.95 if not normed_tolerance: normed_tolerance = 0.95 results = np.array(match_via_correlation(image, template, raw_tolerance=raw_tolerance, normed_tolerance=normed_tolerance)) elif method == 'correlation coefficient': if not raw_tolerance: raw_tolerance = 0.95 if not normed_tolerance: normed_tolerance = 0.95 results = np.array(match_via_correlation_coefficient(image, template, raw_tolerance=raw_tolerance, normed_tolerance=normed_tolerance)) elif method == 'squared difference': if not raw_tolerance: raw_tolerance = 0.95 if not normed_tolerance: normed_tolerance = 0.05 results = np.array(match_via_squared_difference(image, template, raw_tolerance=raw_tolerance, sq_diff_tolerance=normed_tolerance)) h, w = image.shape th, tw = template.shape results = np.array([(result[0], result[1]) for result in results]) #match_x, match_y = int(np.mean(results[:,1])), int(np.mean(results[:,0])) results_aggregated_mean_match_position = match_positions((h,w), results) return results_aggregated_mean_match_position def match_positions(shape, list_of_coords): """ In cases where we have multiple matches, each highlighted by a region of coordinates, we need to separate matches, and find mean of each to return as match position """ match_array = np.zeros(shape) try: # excpetion hit on this line if nothing in list_of_coords- i.e. no matches match_array[list_of_coords[:,0],list_of_coords[:,1]] = 1 labelled = label(match_array) objects = find_objects(labelled[0]) coords = [{'x':(slice_x.start, slice_x.stop),'y':(slice_y.start, slice_y.stop)} for (slice_y,slice_x) in objects] final_positions = [(int(np.mean(coords[i]['x'])),int(np.mean(coords[i]['y']))) for i in range(len(coords))] return final_positions except IndexError: print 'no matches found' # this error occurs if no matches are found return [] ## not what we want a all!!! only will take exact matches, defeating entire point def post_process(image, template, list_of_coords): h, w = template.shape for x, y in list_of_coords: print x-h + 1, y-w + 1 sub_image = image[x-h + 1:x + 1, y-w + 1:y + 1] print sub_image.shape, template.shape, x, y if not np.allclose(template, sub_image): list_of_coords.remove((x,y)) return list_of_coords def to_rgb(im): return np.dstack([im.astype(np.uint8)] * 3).copy(order='C') def highlight_matched_region_no_normalisation(image, template, method='correlation', raw_tolerance=0.666): conv = fftconvolve(image, template[::-1,::-1]) th, tw = template.shape r = find_potential_match_regions(template, conv, method=method, raw_tolerance=raw_tolerance) r_in_image = [(r_x, r_y) for (r_x, r_y) in r if (r_x < image.shape[0] and r_y < image.shape[1])] im_rgb = to_rgb(image) for (x,y) in r_in_image: try: im_rgb[x-th:x,y-tw:y] = 0, 100, 100 except IndexError: im_rgb[x,y] = 0, 100, 100 return im_rgb def highlight_matched_region_normalised(image, shape, list_of_coords): th, tw = shape im_rgb = to_rgb(image) for (x,y) in list_of_coords: #print (x,y) try: im_rgb[x-th:x,y-tw:y] = 0, 100, 100 except IndexError: im_rgb[x,y] = 0, 100, 100 return im_rgb
ten10solutions/Geist
geist/matchers.py
match_via_squared_difference
python
def match_via_squared_difference(image, template, raw_tolerance=1, sq_diff_tolerance=0.1): h, w = image.shape th, tw = template.shape # fft based convolution enables fast matching of large images correlation = fftconvolve(image, template[::-1,::-1]) # trim the returned image, fftconvolve returns an image of width: (Temp_w-1) + Im_w + (Temp_w -1), likewise height correlation = correlation[th-1:h, tw-1:w] # find images regions which are potentially matches match_position_dict = get_tiles_at_potential_match_regions(image, template, correlation, raw_tolerance=raw_tolerance) # bright spots in images can lead to false positivies- the normalisation carried out here eliminates those results = calculate_squared_differences(match_position_dict, correlation, template, sq_diff_tolerance=sq_diff_tolerance) return results
Matchihng algorithm based on normalised cross correlation. Using this matching prevents false positives occuring for bright patches in the image
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/matchers.py#L25-L39
[ "def get_tiles_at_potential_match_regions(image, template, transformed_array, method='correlation', raw_tolerance=0.001):\n if method not in ['correlation', 'correlation coefficient', 'squared difference']:\n raise ValueError('Matching method not implemented')\n h, w = template.shape\n match_points = find_potential_match_regions(template, transformed_array, method=method, raw_tolerance=raw_tolerance)\n match_points = [(match[0], match[1]) for match in match_points]\n # create tile for each match point- use dict so we know which match point it applies to\n # match point here is position of top left pixel of tile\n image_tiles_dict = {match_points[i]:image[match_points[i][0]:match_points[i][0]+h,match_points[i][1]:match_points[i][1]+w] for i in range(len(match_points))}\n return image_tiles_dict\n" ]
from .match_position_finder_helpers import get_tiles_at_potential_match_regions, normalise_correlation, normalise_correlation_coefficient, find_potential_match_regions from scipy.signal import fftconvolve from scipy.ndimage.measurements import label, find_objects import numpy as np # both these methods return array of points giving bottom right coordinate of match def match_via_correlation(image, template, raw_tolerance=1, normed_tolerance=0.9): """ Matchihng algorithm based on normalised cross correlation. Using this matching prevents false positives occuring for bright patches in the image """ h, w = image.shape th, tw = template.shape # fft based convolution enables fast matching of large images correlation = fftconvolve(image, template[::-1,::-1]) # trim the returned image, fftconvolve returns an image of width: (Temp_w-1) + Im_w + (Temp_w -1), likewise height correlation = correlation[th-1:h, tw-1:w] # find images regions which are potentially matches match_position_dict = get_tiles_at_potential_match_regions(image, template, correlation, raw_tolerance=raw_tolerance) # bright spots in images can lead to false positivies- the normalisation carried out here eliminates those results = normalise_correlation(match_position_dict, correlation, template, normed_tolerance=normed_tolerance) return results def match_via_correlation_coefficient(image, template, raw_tolerance=1, normed_tolerance=0.9): """ Matching algorithm based on 2-dimensional version of Pearson product-moment correlation coefficient. This is more robust in the case where the match might be scaled or slightly rotated. From experimentation, this method is less prone to false positives than the correlation method. """ h, w = image.shape th, tw = template.shape temp_mean = np.mean(template) temp_minus_mean = template - temp_mean convolution = fftconvolve(image, temp_minus_mean[::-1,::-1]) convolution = convolution[th-1:h, tw-1:w] match_position_dict = get_tiles_at_potential_match_regions(image, template, convolution, method='correlation coefficient', raw_tolerance=raw_tolerance) # this is empty, so think condition is wrong results = normalise_correlation_coefficient(match_position_dict, convolution, template, normed_tolerance=normed_tolerance) return results def fuzzy_match(image, template, normed_tolerance=None, raw_tolerance=None, method='correlation'): """Determines, using one of two methods, whether a match(es) is present and returns the positions of the bottom right corners of the matches. Fuzzy matches returns regions, so the center of each region is returned as the final match location USE THIS FUNCTION IF you need to match, e.g. the same image but rendered slightly different with respect to anti aliasing; the same image on a number of different backgrounds. The method is the name of the matching method used, the details of this do not matter. Use the default method unless you have too many false positives, in this case, use the method 'correlation coefficient.' The correlation coefficient method can also be more robust at matching when the match might not be exact. The raw_tolerance is the proportion of the value at match positions (i.e. the value returned for an exact match) that we count as a match. For fuzzy matching, this value will not be exactly the value returned for an exact match N. B. Lowering raw_tolerance increases the number of potential match tiles requiring normalisation. This DRAMATICALLY slows down matching as normalisation (a process which eliminates false positives) The normed_tolerance is how far a potential match value can differ from one after normalisation. The tolerance values indicated below are from a short investigation, looking to minimise missing items we wish to match, as all as false positives which inevitably occur when performing fuzzy matching. To generate these values, we tested maching letters with different type of antialiasing on a number of backgrounds. """ if method == 'correlation': if not raw_tolerance: raw_tolerance = 0.95 if not normed_tolerance: normed_tolerance = 0.95 results = np.array(match_via_correlation(image, template, raw_tolerance=raw_tolerance, normed_tolerance=normed_tolerance)) elif method == 'correlation coefficient': if not raw_tolerance: raw_tolerance = 0.95 if not normed_tolerance: normed_tolerance = 0.95 results = np.array(match_via_correlation_coefficient(image, template, raw_tolerance=raw_tolerance, normed_tolerance=normed_tolerance)) elif method == 'squared difference': if not raw_tolerance: raw_tolerance = 0.95 if not normed_tolerance: normed_tolerance = 0.05 results = np.array(match_via_squared_difference(image, template, raw_tolerance=raw_tolerance, sq_diff_tolerance=normed_tolerance)) h, w = image.shape th, tw = template.shape results = np.array([(result[0], result[1]) for result in results]) #match_x, match_y = int(np.mean(results[:,1])), int(np.mean(results[:,0])) results_aggregated_mean_match_position = match_positions((h,w), results) return results_aggregated_mean_match_position def match_positions(shape, list_of_coords): """ In cases where we have multiple matches, each highlighted by a region of coordinates, we need to separate matches, and find mean of each to return as match position """ match_array = np.zeros(shape) try: # excpetion hit on this line if nothing in list_of_coords- i.e. no matches match_array[list_of_coords[:,0],list_of_coords[:,1]] = 1 labelled = label(match_array) objects = find_objects(labelled[0]) coords = [{'x':(slice_x.start, slice_x.stop),'y':(slice_y.start, slice_y.stop)} for (slice_y,slice_x) in objects] final_positions = [(int(np.mean(coords[i]['x'])),int(np.mean(coords[i]['y']))) for i in range(len(coords))] return final_positions except IndexError: print 'no matches found' # this error occurs if no matches are found return [] ## not what we want a all!!! only will take exact matches, defeating entire point def post_process(image, template, list_of_coords): h, w = template.shape for x, y in list_of_coords: print x-h + 1, y-w + 1 sub_image = image[x-h + 1:x + 1, y-w + 1:y + 1] print sub_image.shape, template.shape, x, y if not np.allclose(template, sub_image): list_of_coords.remove((x,y)) return list_of_coords def to_rgb(im): return np.dstack([im.astype(np.uint8)] * 3).copy(order='C') def highlight_matched_region_no_normalisation(image, template, method='correlation', raw_tolerance=0.666): conv = fftconvolve(image, template[::-1,::-1]) th, tw = template.shape r = find_potential_match_regions(template, conv, method=method, raw_tolerance=raw_tolerance) r_in_image = [(r_x, r_y) for (r_x, r_y) in r if (r_x < image.shape[0] and r_y < image.shape[1])] im_rgb = to_rgb(image) for (x,y) in r_in_image: try: im_rgb[x-th:x,y-tw:y] = 0, 100, 100 except IndexError: im_rgb[x,y] = 0, 100, 100 return im_rgb def highlight_matched_region_normalised(image, shape, list_of_coords): th, tw = shape im_rgb = to_rgb(image) for (x,y) in list_of_coords: #print (x,y) try: im_rgb[x-th:x,y-tw:y] = 0, 100, 100 except IndexError: im_rgb[x,y] = 0, 100, 100 return im_rgb
ten10solutions/Geist
geist/matchers.py
match_via_correlation_coefficient
python
def match_via_correlation_coefficient(image, template, raw_tolerance=1, normed_tolerance=0.9): h, w = image.shape th, tw = template.shape temp_mean = np.mean(template) temp_minus_mean = template - temp_mean convolution = fftconvolve(image, temp_minus_mean[::-1,::-1]) convolution = convolution[th-1:h, tw-1:w] match_position_dict = get_tiles_at_potential_match_regions(image, template, convolution, method='correlation coefficient', raw_tolerance=raw_tolerance) # this is empty, so think condition is wrong results = normalise_correlation_coefficient(match_position_dict, convolution, template, normed_tolerance=normed_tolerance) return results
Matching algorithm based on 2-dimensional version of Pearson product-moment correlation coefficient. This is more robust in the case where the match might be scaled or slightly rotated. From experimentation, this method is less prone to false positives than the correlation method.
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/matchers.py#L43-L59
[ "def get_tiles_at_potential_match_regions(image, template, transformed_array, method='correlation', raw_tolerance=0.001):\n if method not in ['correlation', 'correlation coefficient', 'squared difference']:\n raise ValueError('Matching method not implemented')\n h, w = template.shape\n match_points = find_potential_match_regions(template, transformed_array, method=method, raw_tolerance=raw_tolerance)\n match_points = [(match[0], match[1]) for match in match_points]\n # create tile for each match point- use dict so we know which match point it applies to\n # match point here is position of top left pixel of tile\n image_tiles_dict = {match_points[i]:image[match_points[i][0]:match_points[i][0]+h,match_points[i][1]:match_points[i][1]+w] for i in range(len(match_points))}\n return image_tiles_dict\n", "def normalise_correlation_coefficient(image_tile_dict, transformed_array, template, normed_tolerance=1):\n \"\"\"As above, but for when the correlation coefficient matching method is used\n \"\"\"\n template_mean = np.mean(template)\n template_minus_mean = template - template_mean\n template_norm = np.linalg.norm(template_minus_mean)\n image_norms = {(x,y):np.linalg.norm(image_tile_dict[(x,y)]- np.mean(image_tile_dict[(x,y)]))*template_norm for (x,y) in image_tile_dict.keys()}\n match_points = image_tile_dict.keys()\n # for correlation, then need to transofrm back to get correct value for division\n h, w = template.shape\n image_matches_normalised = {match_points[i]:transformed_array[match_points[i][0], match_points[i][1]]/image_norms[match_points[i]] for i in range(len(match_points))}\n normalised_matches = {key:value for key, value in image_matches_normalised.items() if np.round(value, decimals=3) >= normed_tolerance}\n return normalised_matches.keys()\n" ]
from .match_position_finder_helpers import get_tiles_at_potential_match_regions, normalise_correlation, normalise_correlation_coefficient, find_potential_match_regions from scipy.signal import fftconvolve from scipy.ndimage.measurements import label, find_objects import numpy as np # both these methods return array of points giving bottom right coordinate of match def match_via_correlation(image, template, raw_tolerance=1, normed_tolerance=0.9): """ Matchihng algorithm based on normalised cross correlation. Using this matching prevents false positives occuring for bright patches in the image """ h, w = image.shape th, tw = template.shape # fft based convolution enables fast matching of large images correlation = fftconvolve(image, template[::-1,::-1]) # trim the returned image, fftconvolve returns an image of width: (Temp_w-1) + Im_w + (Temp_w -1), likewise height correlation = correlation[th-1:h, tw-1:w] # find images regions which are potentially matches match_position_dict = get_tiles_at_potential_match_regions(image, template, correlation, raw_tolerance=raw_tolerance) # bright spots in images can lead to false positivies- the normalisation carried out here eliminates those results = normalise_correlation(match_position_dict, correlation, template, normed_tolerance=normed_tolerance) return results def match_via_squared_difference(image, template, raw_tolerance=1, sq_diff_tolerance=0.1): """ Matchihng algorithm based on normalised cross correlation. Using this matching prevents false positives occuring for bright patches in the image """ h, w = image.shape th, tw = template.shape # fft based convolution enables fast matching of large images correlation = fftconvolve(image, template[::-1,::-1]) # trim the returned image, fftconvolve returns an image of width: (Temp_w-1) + Im_w + (Temp_w -1), likewise height correlation = correlation[th-1:h, tw-1:w] # find images regions which are potentially matches match_position_dict = get_tiles_at_potential_match_regions(image, template, correlation, raw_tolerance=raw_tolerance) # bright spots in images can lead to false positivies- the normalisation carried out here eliminates those results = calculate_squared_differences(match_position_dict, correlation, template, sq_diff_tolerance=sq_diff_tolerance) return results def fuzzy_match(image, template, normed_tolerance=None, raw_tolerance=None, method='correlation'): """Determines, using one of two methods, whether a match(es) is present and returns the positions of the bottom right corners of the matches. Fuzzy matches returns regions, so the center of each region is returned as the final match location USE THIS FUNCTION IF you need to match, e.g. the same image but rendered slightly different with respect to anti aliasing; the same image on a number of different backgrounds. The method is the name of the matching method used, the details of this do not matter. Use the default method unless you have too many false positives, in this case, use the method 'correlation coefficient.' The correlation coefficient method can also be more robust at matching when the match might not be exact. The raw_tolerance is the proportion of the value at match positions (i.e. the value returned for an exact match) that we count as a match. For fuzzy matching, this value will not be exactly the value returned for an exact match N. B. Lowering raw_tolerance increases the number of potential match tiles requiring normalisation. This DRAMATICALLY slows down matching as normalisation (a process which eliminates false positives) The normed_tolerance is how far a potential match value can differ from one after normalisation. The tolerance values indicated below are from a short investigation, looking to minimise missing items we wish to match, as all as false positives which inevitably occur when performing fuzzy matching. To generate these values, we tested maching letters with different type of antialiasing on a number of backgrounds. """ if method == 'correlation': if not raw_tolerance: raw_tolerance = 0.95 if not normed_tolerance: normed_tolerance = 0.95 results = np.array(match_via_correlation(image, template, raw_tolerance=raw_tolerance, normed_tolerance=normed_tolerance)) elif method == 'correlation coefficient': if not raw_tolerance: raw_tolerance = 0.95 if not normed_tolerance: normed_tolerance = 0.95 results = np.array(match_via_correlation_coefficient(image, template, raw_tolerance=raw_tolerance, normed_tolerance=normed_tolerance)) elif method == 'squared difference': if not raw_tolerance: raw_tolerance = 0.95 if not normed_tolerance: normed_tolerance = 0.05 results = np.array(match_via_squared_difference(image, template, raw_tolerance=raw_tolerance, sq_diff_tolerance=normed_tolerance)) h, w = image.shape th, tw = template.shape results = np.array([(result[0], result[1]) for result in results]) #match_x, match_y = int(np.mean(results[:,1])), int(np.mean(results[:,0])) results_aggregated_mean_match_position = match_positions((h,w), results) return results_aggregated_mean_match_position def match_positions(shape, list_of_coords): """ In cases where we have multiple matches, each highlighted by a region of coordinates, we need to separate matches, and find mean of each to return as match position """ match_array = np.zeros(shape) try: # excpetion hit on this line if nothing in list_of_coords- i.e. no matches match_array[list_of_coords[:,0],list_of_coords[:,1]] = 1 labelled = label(match_array) objects = find_objects(labelled[0]) coords = [{'x':(slice_x.start, slice_x.stop),'y':(slice_y.start, slice_y.stop)} for (slice_y,slice_x) in objects] final_positions = [(int(np.mean(coords[i]['x'])),int(np.mean(coords[i]['y']))) for i in range(len(coords))] return final_positions except IndexError: print 'no matches found' # this error occurs if no matches are found return [] ## not what we want a all!!! only will take exact matches, defeating entire point def post_process(image, template, list_of_coords): h, w = template.shape for x, y in list_of_coords: print x-h + 1, y-w + 1 sub_image = image[x-h + 1:x + 1, y-w + 1:y + 1] print sub_image.shape, template.shape, x, y if not np.allclose(template, sub_image): list_of_coords.remove((x,y)) return list_of_coords def to_rgb(im): return np.dstack([im.astype(np.uint8)] * 3).copy(order='C') def highlight_matched_region_no_normalisation(image, template, method='correlation', raw_tolerance=0.666): conv = fftconvolve(image, template[::-1,::-1]) th, tw = template.shape r = find_potential_match_regions(template, conv, method=method, raw_tolerance=raw_tolerance) r_in_image = [(r_x, r_y) for (r_x, r_y) in r if (r_x < image.shape[0] and r_y < image.shape[1])] im_rgb = to_rgb(image) for (x,y) in r_in_image: try: im_rgb[x-th:x,y-tw:y] = 0, 100, 100 except IndexError: im_rgb[x,y] = 0, 100, 100 return im_rgb def highlight_matched_region_normalised(image, shape, list_of_coords): th, tw = shape im_rgb = to_rgb(image) for (x,y) in list_of_coords: #print (x,y) try: im_rgb[x-th:x,y-tw:y] = 0, 100, 100 except IndexError: im_rgb[x,y] = 0, 100, 100 return im_rgb
ten10solutions/Geist
geist/matchers.py
fuzzy_match
python
def fuzzy_match(image, template, normed_tolerance=None, raw_tolerance=None, method='correlation'): if method == 'correlation': if not raw_tolerance: raw_tolerance = 0.95 if not normed_tolerance: normed_tolerance = 0.95 results = np.array(match_via_correlation(image, template, raw_tolerance=raw_tolerance, normed_tolerance=normed_tolerance)) elif method == 'correlation coefficient': if not raw_tolerance: raw_tolerance = 0.95 if not normed_tolerance: normed_tolerance = 0.95 results = np.array(match_via_correlation_coefficient(image, template, raw_tolerance=raw_tolerance, normed_tolerance=normed_tolerance)) elif method == 'squared difference': if not raw_tolerance: raw_tolerance = 0.95 if not normed_tolerance: normed_tolerance = 0.05 results = np.array(match_via_squared_difference(image, template, raw_tolerance=raw_tolerance, sq_diff_tolerance=normed_tolerance)) h, w = image.shape th, tw = template.shape results = np.array([(result[0], result[1]) for result in results]) #match_x, match_y = int(np.mean(results[:,1])), int(np.mean(results[:,0])) results_aggregated_mean_match_position = match_positions((h,w), results) return results_aggregated_mean_match_position
Determines, using one of two methods, whether a match(es) is present and returns the positions of the bottom right corners of the matches. Fuzzy matches returns regions, so the center of each region is returned as the final match location USE THIS FUNCTION IF you need to match, e.g. the same image but rendered slightly different with respect to anti aliasing; the same image on a number of different backgrounds. The method is the name of the matching method used, the details of this do not matter. Use the default method unless you have too many false positives, in this case, use the method 'correlation coefficient.' The correlation coefficient method can also be more robust at matching when the match might not be exact. The raw_tolerance is the proportion of the value at match positions (i.e. the value returned for an exact match) that we count as a match. For fuzzy matching, this value will not be exactly the value returned for an exact match N. B. Lowering raw_tolerance increases the number of potential match tiles requiring normalisation. This DRAMATICALLY slows down matching as normalisation (a process which eliminates false positives) The normed_tolerance is how far a potential match value can differ from one after normalisation. The tolerance values indicated below are from a short investigation, looking to minimise missing items we wish to match, as all as false positives which inevitably occur when performing fuzzy matching. To generate these values, we tested maching letters with different type of antialiasing on a number of backgrounds.
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/matchers.py#L64-L110
[ "def match_positions(shape, list_of_coords):\n \"\"\" In cases where we have multiple matches, each highlighted by a region of coordinates,\n we need to separate matches, and find mean of each to return as match position\n \"\"\"\n match_array = np.zeros(shape)\n try:\n # excpetion hit on this line if nothing in list_of_coords- i.e. no matches\n match_array[list_of_coords[:,0],list_of_coords[:,1]] = 1\n labelled = label(match_array)\n objects = find_objects(labelled[0])\n coords = [{'x':(slice_x.start, slice_x.stop),'y':(slice_y.start, slice_y.stop)} for (slice_y,slice_x) in objects]\n final_positions = [(int(np.mean(coords[i]['x'])),int(np.mean(coords[i]['y']))) for i in range(len(coords))]\n return final_positions\n except IndexError:\n print 'no matches found'\n # this error occurs if no matches are found\n return []\n", "def match_via_correlation(image, template, raw_tolerance=1, normed_tolerance=0.9):\n \"\"\" Matchihng algorithm based on normalised cross correlation.\n Using this matching prevents false positives occuring for bright patches in the image\n \"\"\"\n h, w = image.shape\n th, tw = template.shape\n # fft based convolution enables fast matching of large images\n correlation = fftconvolve(image, template[::-1,::-1])\n # trim the returned image, fftconvolve returns an image of width: (Temp_w-1) + Im_w + (Temp_w -1), likewise height\n correlation = correlation[th-1:h, tw-1:w]\n # find images regions which are potentially matches\n match_position_dict = get_tiles_at_potential_match_regions(image, template, correlation, raw_tolerance=raw_tolerance)\n # bright spots in images can lead to false positivies- the normalisation carried out here eliminates those\n results = normalise_correlation(match_position_dict, correlation, template, normed_tolerance=normed_tolerance)\n return results\n", "def match_via_squared_difference(image, template, raw_tolerance=1, sq_diff_tolerance=0.1):\n \"\"\" Matchihng algorithm based on normalised cross correlation.\n Using this matching prevents false positives occuring for bright patches in the image\n \"\"\"\n h, w = image.shape\n th, tw = template.shape\n # fft based convolution enables fast matching of large images\n correlation = fftconvolve(image, template[::-1,::-1])\n # trim the returned image, fftconvolve returns an image of width: (Temp_w-1) + Im_w + (Temp_w -1), likewise height\n correlation = correlation[th-1:h, tw-1:w]\n # find images regions which are potentially matches\n match_position_dict = get_tiles_at_potential_match_regions(image, template, correlation, raw_tolerance=raw_tolerance)\n # bright spots in images can lead to false positivies- the normalisation carried out here eliminates those\n results = calculate_squared_differences(match_position_dict, correlation, template, sq_diff_tolerance=sq_diff_tolerance)\n return results\n", "def match_via_correlation_coefficient(image, template, raw_tolerance=1, normed_tolerance=0.9):\n \"\"\" Matching algorithm based on 2-dimensional version of Pearson product-moment correlation coefficient.\n\n This is more robust in the case where the match might be scaled or slightly rotated.\n\n From experimentation, this method is less prone to false positives than the correlation method.\n \"\"\"\n h, w = image.shape\n th, tw = template.shape\n temp_mean = np.mean(template)\n temp_minus_mean = template - temp_mean\n convolution = fftconvolve(image, temp_minus_mean[::-1,::-1])\n convolution = convolution[th-1:h, tw-1:w]\n match_position_dict = get_tiles_at_potential_match_regions(image, template, convolution, method='correlation coefficient', raw_tolerance=raw_tolerance)\n # this is empty, so think condition is wrong\n results = normalise_correlation_coefficient(match_position_dict, convolution, template, normed_tolerance=normed_tolerance)\n return results\n" ]
from .match_position_finder_helpers import get_tiles_at_potential_match_regions, normalise_correlation, normalise_correlation_coefficient, find_potential_match_regions from scipy.signal import fftconvolve from scipy.ndimage.measurements import label, find_objects import numpy as np # both these methods return array of points giving bottom right coordinate of match def match_via_correlation(image, template, raw_tolerance=1, normed_tolerance=0.9): """ Matchihng algorithm based on normalised cross correlation. Using this matching prevents false positives occuring for bright patches in the image """ h, w = image.shape th, tw = template.shape # fft based convolution enables fast matching of large images correlation = fftconvolve(image, template[::-1,::-1]) # trim the returned image, fftconvolve returns an image of width: (Temp_w-1) + Im_w + (Temp_w -1), likewise height correlation = correlation[th-1:h, tw-1:w] # find images regions which are potentially matches match_position_dict = get_tiles_at_potential_match_regions(image, template, correlation, raw_tolerance=raw_tolerance) # bright spots in images can lead to false positivies- the normalisation carried out here eliminates those results = normalise_correlation(match_position_dict, correlation, template, normed_tolerance=normed_tolerance) return results def match_via_squared_difference(image, template, raw_tolerance=1, sq_diff_tolerance=0.1): """ Matchihng algorithm based on normalised cross correlation. Using this matching prevents false positives occuring for bright patches in the image """ h, w = image.shape th, tw = template.shape # fft based convolution enables fast matching of large images correlation = fftconvolve(image, template[::-1,::-1]) # trim the returned image, fftconvolve returns an image of width: (Temp_w-1) + Im_w + (Temp_w -1), likewise height correlation = correlation[th-1:h, tw-1:w] # find images regions which are potentially matches match_position_dict = get_tiles_at_potential_match_regions(image, template, correlation, raw_tolerance=raw_tolerance) # bright spots in images can lead to false positivies- the normalisation carried out here eliminates those results = calculate_squared_differences(match_position_dict, correlation, template, sq_diff_tolerance=sq_diff_tolerance) return results def match_via_correlation_coefficient(image, template, raw_tolerance=1, normed_tolerance=0.9): """ Matching algorithm based on 2-dimensional version of Pearson product-moment correlation coefficient. This is more robust in the case where the match might be scaled or slightly rotated. From experimentation, this method is less prone to false positives than the correlation method. """ h, w = image.shape th, tw = template.shape temp_mean = np.mean(template) temp_minus_mean = template - temp_mean convolution = fftconvolve(image, temp_minus_mean[::-1,::-1]) convolution = convolution[th-1:h, tw-1:w] match_position_dict = get_tiles_at_potential_match_regions(image, template, convolution, method='correlation coefficient', raw_tolerance=raw_tolerance) # this is empty, so think condition is wrong results = normalise_correlation_coefficient(match_position_dict, convolution, template, normed_tolerance=normed_tolerance) return results def match_positions(shape, list_of_coords): """ In cases where we have multiple matches, each highlighted by a region of coordinates, we need to separate matches, and find mean of each to return as match position """ match_array = np.zeros(shape) try: # excpetion hit on this line if nothing in list_of_coords- i.e. no matches match_array[list_of_coords[:,0],list_of_coords[:,1]] = 1 labelled = label(match_array) objects = find_objects(labelled[0]) coords = [{'x':(slice_x.start, slice_x.stop),'y':(slice_y.start, slice_y.stop)} for (slice_y,slice_x) in objects] final_positions = [(int(np.mean(coords[i]['x'])),int(np.mean(coords[i]['y']))) for i in range(len(coords))] return final_positions except IndexError: print 'no matches found' # this error occurs if no matches are found return [] ## not what we want a all!!! only will take exact matches, defeating entire point def post_process(image, template, list_of_coords): h, w = template.shape for x, y in list_of_coords: print x-h + 1, y-w + 1 sub_image = image[x-h + 1:x + 1, y-w + 1:y + 1] print sub_image.shape, template.shape, x, y if not np.allclose(template, sub_image): list_of_coords.remove((x,y)) return list_of_coords def to_rgb(im): return np.dstack([im.astype(np.uint8)] * 3).copy(order='C') def highlight_matched_region_no_normalisation(image, template, method='correlation', raw_tolerance=0.666): conv = fftconvolve(image, template[::-1,::-1]) th, tw = template.shape r = find_potential_match_regions(template, conv, method=method, raw_tolerance=raw_tolerance) r_in_image = [(r_x, r_y) for (r_x, r_y) in r if (r_x < image.shape[0] and r_y < image.shape[1])] im_rgb = to_rgb(image) for (x,y) in r_in_image: try: im_rgb[x-th:x,y-tw:y] = 0, 100, 100 except IndexError: im_rgb[x,y] = 0, 100, 100 return im_rgb def highlight_matched_region_normalised(image, shape, list_of_coords): th, tw = shape im_rgb = to_rgb(image) for (x,y) in list_of_coords: #print (x,y) try: im_rgb[x-th:x,y-tw:y] = 0, 100, 100 except IndexError: im_rgb[x,y] = 0, 100, 100 return im_rgb
ten10solutions/Geist
geist/matchers.py
match_positions
python
def match_positions(shape, list_of_coords): match_array = np.zeros(shape) try: # excpetion hit on this line if nothing in list_of_coords- i.e. no matches match_array[list_of_coords[:,0],list_of_coords[:,1]] = 1 labelled = label(match_array) objects = find_objects(labelled[0]) coords = [{'x':(slice_x.start, slice_x.stop),'y':(slice_y.start, slice_y.stop)} for (slice_y,slice_x) in objects] final_positions = [(int(np.mean(coords[i]['x'])),int(np.mean(coords[i]['y']))) for i in range(len(coords))] return final_positions except IndexError: print 'no matches found' # this error occurs if no matches are found return []
In cases where we have multiple matches, each highlighted by a region of coordinates, we need to separate matches, and find mean of each to return as match position
train
https://github.com/ten10solutions/Geist/blob/a1ef16d8b4c3777735008b671a50acfde3ce7bf1/geist/matchers.py#L114-L130
null
from .match_position_finder_helpers import get_tiles_at_potential_match_regions, normalise_correlation, normalise_correlation_coefficient, find_potential_match_regions from scipy.signal import fftconvolve from scipy.ndimage.measurements import label, find_objects import numpy as np # both these methods return array of points giving bottom right coordinate of match def match_via_correlation(image, template, raw_tolerance=1, normed_tolerance=0.9): """ Matchihng algorithm based on normalised cross correlation. Using this matching prevents false positives occuring for bright patches in the image """ h, w = image.shape th, tw = template.shape # fft based convolution enables fast matching of large images correlation = fftconvolve(image, template[::-1,::-1]) # trim the returned image, fftconvolve returns an image of width: (Temp_w-1) + Im_w + (Temp_w -1), likewise height correlation = correlation[th-1:h, tw-1:w] # find images regions which are potentially matches match_position_dict = get_tiles_at_potential_match_regions(image, template, correlation, raw_tolerance=raw_tolerance) # bright spots in images can lead to false positivies- the normalisation carried out here eliminates those results = normalise_correlation(match_position_dict, correlation, template, normed_tolerance=normed_tolerance) return results def match_via_squared_difference(image, template, raw_tolerance=1, sq_diff_tolerance=0.1): """ Matchihng algorithm based on normalised cross correlation. Using this matching prevents false positives occuring for bright patches in the image """ h, w = image.shape th, tw = template.shape # fft based convolution enables fast matching of large images correlation = fftconvolve(image, template[::-1,::-1]) # trim the returned image, fftconvolve returns an image of width: (Temp_w-1) + Im_w + (Temp_w -1), likewise height correlation = correlation[th-1:h, tw-1:w] # find images regions which are potentially matches match_position_dict = get_tiles_at_potential_match_regions(image, template, correlation, raw_tolerance=raw_tolerance) # bright spots in images can lead to false positivies- the normalisation carried out here eliminates those results = calculate_squared_differences(match_position_dict, correlation, template, sq_diff_tolerance=sq_diff_tolerance) return results def match_via_correlation_coefficient(image, template, raw_tolerance=1, normed_tolerance=0.9): """ Matching algorithm based on 2-dimensional version of Pearson product-moment correlation coefficient. This is more robust in the case where the match might be scaled or slightly rotated. From experimentation, this method is less prone to false positives than the correlation method. """ h, w = image.shape th, tw = template.shape temp_mean = np.mean(template) temp_minus_mean = template - temp_mean convolution = fftconvolve(image, temp_minus_mean[::-1,::-1]) convolution = convolution[th-1:h, tw-1:w] match_position_dict = get_tiles_at_potential_match_regions(image, template, convolution, method='correlation coefficient', raw_tolerance=raw_tolerance) # this is empty, so think condition is wrong results = normalise_correlation_coefficient(match_position_dict, convolution, template, normed_tolerance=normed_tolerance) return results def fuzzy_match(image, template, normed_tolerance=None, raw_tolerance=None, method='correlation'): """Determines, using one of two methods, whether a match(es) is present and returns the positions of the bottom right corners of the matches. Fuzzy matches returns regions, so the center of each region is returned as the final match location USE THIS FUNCTION IF you need to match, e.g. the same image but rendered slightly different with respect to anti aliasing; the same image on a number of different backgrounds. The method is the name of the matching method used, the details of this do not matter. Use the default method unless you have too many false positives, in this case, use the method 'correlation coefficient.' The correlation coefficient method can also be more robust at matching when the match might not be exact. The raw_tolerance is the proportion of the value at match positions (i.e. the value returned for an exact match) that we count as a match. For fuzzy matching, this value will not be exactly the value returned for an exact match N. B. Lowering raw_tolerance increases the number of potential match tiles requiring normalisation. This DRAMATICALLY slows down matching as normalisation (a process which eliminates false positives) The normed_tolerance is how far a potential match value can differ from one after normalisation. The tolerance values indicated below are from a short investigation, looking to minimise missing items we wish to match, as all as false positives which inevitably occur when performing fuzzy matching. To generate these values, we tested maching letters with different type of antialiasing on a number of backgrounds. """ if method == 'correlation': if not raw_tolerance: raw_tolerance = 0.95 if not normed_tolerance: normed_tolerance = 0.95 results = np.array(match_via_correlation(image, template, raw_tolerance=raw_tolerance, normed_tolerance=normed_tolerance)) elif method == 'correlation coefficient': if not raw_tolerance: raw_tolerance = 0.95 if not normed_tolerance: normed_tolerance = 0.95 results = np.array(match_via_correlation_coefficient(image, template, raw_tolerance=raw_tolerance, normed_tolerance=normed_tolerance)) elif method == 'squared difference': if not raw_tolerance: raw_tolerance = 0.95 if not normed_tolerance: normed_tolerance = 0.05 results = np.array(match_via_squared_difference(image, template, raw_tolerance=raw_tolerance, sq_diff_tolerance=normed_tolerance)) h, w = image.shape th, tw = template.shape results = np.array([(result[0], result[1]) for result in results]) #match_x, match_y = int(np.mean(results[:,1])), int(np.mean(results[:,0])) results_aggregated_mean_match_position = match_positions((h,w), results) return results_aggregated_mean_match_position ## not what we want a all!!! only will take exact matches, defeating entire point def post_process(image, template, list_of_coords): h, w = template.shape for x, y in list_of_coords: print x-h + 1, y-w + 1 sub_image = image[x-h + 1:x + 1, y-w + 1:y + 1] print sub_image.shape, template.shape, x, y if not np.allclose(template, sub_image): list_of_coords.remove((x,y)) return list_of_coords def to_rgb(im): return np.dstack([im.astype(np.uint8)] * 3).copy(order='C') def highlight_matched_region_no_normalisation(image, template, method='correlation', raw_tolerance=0.666): conv = fftconvolve(image, template[::-1,::-1]) th, tw = template.shape r = find_potential_match_regions(template, conv, method=method, raw_tolerance=raw_tolerance) r_in_image = [(r_x, r_y) for (r_x, r_y) in r if (r_x < image.shape[0] and r_y < image.shape[1])] im_rgb = to_rgb(image) for (x,y) in r_in_image: try: im_rgb[x-th:x,y-tw:y] = 0, 100, 100 except IndexError: im_rgb[x,y] = 0, 100, 100 return im_rgb def highlight_matched_region_normalised(image, shape, list_of_coords): th, tw = shape im_rgb = to_rgb(image) for (x,y) in list_of_coords: #print (x,y) try: im_rgb[x-th:x,y-tw:y] = 0, 100, 100 except IndexError: im_rgb[x,y] = 0, 100, 100 return im_rgb
stxnext/mappet
mappet/helpers.py
no_empty_value
python
def no_empty_value(func): @wraps(func) def wrapper(value): if not value: raise Exception("Empty value not allowed") return func(value) return wrapper
Raises an exception if function argument is empty.
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/helpers.py#L38-L45
null
# -*- coding: utf-8 -*- u"""Helper functions. .. :module: helpers :synopsis: Helper functions. """ from collections import defaultdict from decimal import Decimal from functools import partial, wraps import datetime from lxml import etree import dateutil.parser __all__ = [ 'to_bool', 'to_date', 'to_datetime', 'to_decimal', 'to_float', 'to_int', 'to_str', 'to_time', 'from_bool', 'from_date', 'from_datetime', 'from_time', 'CAST_DICT', 'normalize_tag', 'etree_to_dict', 'dict_to_etree', ] def to_bool(value): """Converts human boolean-like values to Python boolean. Falls back to :class:`bool` when ``value`` is not recognized. :param value: the value to convert :returns: ``True`` if value is truthy, ``False`` otherwise :rtype: bool """ cases = { '0': False, 'false': False, 'no': False, '1': True, 'true': True, 'yes': True, } value = value.lower() if isinstance(value, basestring) else value return cases.get(value, bool(value)) def to_str(value): u"""Represents values as unicode strings to support diacritics.""" return unicode(value) def to_int(value): return int(value) def to_float(value): return float(value) def to_decimal(value): return Decimal(value) @no_empty_value def to_time(value): value = str(value) # dateutil.parse has problems parsing full hours without minutes sep = value[2:3] if not (sep == ':' or sep.isdigit()): value = value[:2] + ':00' + value[2:] return dateutil.parser.parse(value).time() @no_empty_value def to_datetime(value): return parse_datetime(value) def parse_datetime(value): value = str(value) return dateutil.parser.parse(value) @no_empty_value def to_date(value): return parse_datetime(value) def from_bool(value): cases = { True: 'YES', False: 'NO', } try: return cases.get(value, bool(value)) except Exception: return False def from_time(value): if not isinstance(value, datetime.time): raise Exception('Value {} is not datetime.time object'.format(value)) return value.isoformat() @no_empty_value def from_datetime(value): if not isinstance(value, datetime.datetime): raise Exception('Unexpected type {} of value {} (expected datetime.datetime)'.format(type(value), value)) if value.tzinfo is None: value = value.replace(tzinfo=dateutil.tz.tzlocal()) # pragma: nocover return value.replace(microsecond=0).isoformat() @no_empty_value def from_date(value): if not isinstance(value, datetime.date) and not isinstance(value, datetime.datetime): raise Exception('Not datetime.date object but {}: {}'.format(type(value), value)) return value.isoformat() CAST_DICT = { bool: from_bool, int: str, str: str, unicode: str, float: str, datetime.time: from_time, datetime.datetime: from_datetime, datetime.date: from_date, } def normalize_tag(tag): u"""Normalizes tag name. :param str tag: tag name to normalize :rtype: str :returns: normalized tag name >>> normalize_tag('tag-NaMe') 'tag_name' """ return tag.lower().replace('-', '_') def etree_to_dict(t, trim=True, **kw): u"""Converts an lxml.etree object to Python dict. >>> etree_to_dict(etree.Element('root')) {'root': None} :param etree.Element t: lxml tree to convert :returns d: a dict representing the lxml tree ``t`` :rtype: dict """ d = {t.tag: {} if t.attrib else None} children = list(t) etree_to_dict_w_args = partial(etree_to_dict, trim=trim, **kw) if children: dd = defaultdict(list) d = {t.tag: {}} for dc in map(etree_to_dict_w_args, children): for k, v in dc.iteritems(): # do not add Comment instance to the key if k is not etree.Comment: dd[k].append(v) d[t.tag] = {k: v[0] if len(v) == 1 else v for k, v in dd.iteritems()} if t.attrib: d[t.tag].update(('@' + k, v) for k, v in t.attrib.iteritems()) if trim and t.text: t.text = t.text.strip() if t.text: if t.tag is etree.Comment and not kw.get('without_comments'): # adds a comments node d['#comments'] = t.text elif children or t.attrib: d[t.tag]['#text'] = t.text else: d[t.tag] = t.text return d def dict_to_etree(d, root): u"""Converts a dict to lxml.etree object. >>> dict_to_etree({'root': {'#text': 'node_text', '@attr': 'val'}}, etree.Element('root')) # doctest: +ELLIPSIS <Element root at 0x...> :param dict d: dict representing the XML tree :param etree.Element root: XML node which will be assigned the resulting tree :returns: Textual representation of the XML tree :rtype: str """ def _to_etree(d, node): if d is None or len(d) == 0: return elif isinstance(d, basestring): node.text = d elif isinstance(d, dict): for k, v in d.items(): assert isinstance(k, basestring) if k.startswith('#'): assert k == '#text' assert isinstance(v, basestring) node.text = v elif k.startswith('@'): assert isinstance(v, basestring) node.set(k[1:], v) elif isinstance(v, list): # No matter the child count, their parent will be the same. sub_elem = etree.SubElement(node, k) for child_num, e in enumerate(v): if e is None: if child_num == 0: # Found the first occurrence of an empty child, # skip creating of its XML repr, since it would be # the same as ``sub_element`` higher up. continue # A list with None element means an empty child node # in its parent, thus, recreating tags we have to go # up one level. # <node><child/></child></node> <=> {'node': 'child': [None, None]} _to_etree(node, k) else: # If this isn't first child and it's a complex # value (dict), we need to check if it's value # is equivalent to None. if child_num != 0 and not (isinstance(e, dict) and not all(e.values())): # At least one child was None, we have to create # a new parent-node, which will not be empty. sub_elem = etree.SubElement(node, k) _to_etree(e, sub_elem) else: _to_etree(v, etree.SubElement(node, k)) elif etree.iselement(d): # Supports the case, when we got an empty child and want to recreate it. etree.SubElement(d, node) else: raise AttributeError('Argument is neither dict nor basestring.') _to_etree(d, root) return root
stxnext/mappet
mappet/helpers.py
to_bool
python
def to_bool(value): cases = { '0': False, 'false': False, 'no': False, '1': True, 'true': True, 'yes': True, } value = value.lower() if isinstance(value, basestring) else value return cases.get(value, bool(value))
Converts human boolean-like values to Python boolean. Falls back to :class:`bool` when ``value`` is not recognized. :param value: the value to convert :returns: ``True`` if value is truthy, ``False`` otherwise :rtype: bool
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/helpers.py#L48-L67
null
# -*- coding: utf-8 -*- u"""Helper functions. .. :module: helpers :synopsis: Helper functions. """ from collections import defaultdict from decimal import Decimal from functools import partial, wraps import datetime from lxml import etree import dateutil.parser __all__ = [ 'to_bool', 'to_date', 'to_datetime', 'to_decimal', 'to_float', 'to_int', 'to_str', 'to_time', 'from_bool', 'from_date', 'from_datetime', 'from_time', 'CAST_DICT', 'normalize_tag', 'etree_to_dict', 'dict_to_etree', ] def no_empty_value(func): """Raises an exception if function argument is empty.""" @wraps(func) def wrapper(value): if not value: raise Exception("Empty value not allowed") return func(value) return wrapper def to_str(value): u"""Represents values as unicode strings to support diacritics.""" return unicode(value) def to_int(value): return int(value) def to_float(value): return float(value) def to_decimal(value): return Decimal(value) @no_empty_value def to_time(value): value = str(value) # dateutil.parse has problems parsing full hours without minutes sep = value[2:3] if not (sep == ':' or sep.isdigit()): value = value[:2] + ':00' + value[2:] return dateutil.parser.parse(value).time() @no_empty_value def to_datetime(value): return parse_datetime(value) def parse_datetime(value): value = str(value) return dateutil.parser.parse(value) @no_empty_value def to_date(value): return parse_datetime(value) def from_bool(value): cases = { True: 'YES', False: 'NO', } try: return cases.get(value, bool(value)) except Exception: return False def from_time(value): if not isinstance(value, datetime.time): raise Exception('Value {} is not datetime.time object'.format(value)) return value.isoformat() @no_empty_value def from_datetime(value): if not isinstance(value, datetime.datetime): raise Exception('Unexpected type {} of value {} (expected datetime.datetime)'.format(type(value), value)) if value.tzinfo is None: value = value.replace(tzinfo=dateutil.tz.tzlocal()) # pragma: nocover return value.replace(microsecond=0).isoformat() @no_empty_value def from_date(value): if not isinstance(value, datetime.date) and not isinstance(value, datetime.datetime): raise Exception('Not datetime.date object but {}: {}'.format(type(value), value)) return value.isoformat() CAST_DICT = { bool: from_bool, int: str, str: str, unicode: str, float: str, datetime.time: from_time, datetime.datetime: from_datetime, datetime.date: from_date, } def normalize_tag(tag): u"""Normalizes tag name. :param str tag: tag name to normalize :rtype: str :returns: normalized tag name >>> normalize_tag('tag-NaMe') 'tag_name' """ return tag.lower().replace('-', '_') def etree_to_dict(t, trim=True, **kw): u"""Converts an lxml.etree object to Python dict. >>> etree_to_dict(etree.Element('root')) {'root': None} :param etree.Element t: lxml tree to convert :returns d: a dict representing the lxml tree ``t`` :rtype: dict """ d = {t.tag: {} if t.attrib else None} children = list(t) etree_to_dict_w_args = partial(etree_to_dict, trim=trim, **kw) if children: dd = defaultdict(list) d = {t.tag: {}} for dc in map(etree_to_dict_w_args, children): for k, v in dc.iteritems(): # do not add Comment instance to the key if k is not etree.Comment: dd[k].append(v) d[t.tag] = {k: v[0] if len(v) == 1 else v for k, v in dd.iteritems()} if t.attrib: d[t.tag].update(('@' + k, v) for k, v in t.attrib.iteritems()) if trim and t.text: t.text = t.text.strip() if t.text: if t.tag is etree.Comment and not kw.get('without_comments'): # adds a comments node d['#comments'] = t.text elif children or t.attrib: d[t.tag]['#text'] = t.text else: d[t.tag] = t.text return d def dict_to_etree(d, root): u"""Converts a dict to lxml.etree object. >>> dict_to_etree({'root': {'#text': 'node_text', '@attr': 'val'}}, etree.Element('root')) # doctest: +ELLIPSIS <Element root at 0x...> :param dict d: dict representing the XML tree :param etree.Element root: XML node which will be assigned the resulting tree :returns: Textual representation of the XML tree :rtype: str """ def _to_etree(d, node): if d is None or len(d) == 0: return elif isinstance(d, basestring): node.text = d elif isinstance(d, dict): for k, v in d.items(): assert isinstance(k, basestring) if k.startswith('#'): assert k == '#text' assert isinstance(v, basestring) node.text = v elif k.startswith('@'): assert isinstance(v, basestring) node.set(k[1:], v) elif isinstance(v, list): # No matter the child count, their parent will be the same. sub_elem = etree.SubElement(node, k) for child_num, e in enumerate(v): if e is None: if child_num == 0: # Found the first occurrence of an empty child, # skip creating of its XML repr, since it would be # the same as ``sub_element`` higher up. continue # A list with None element means an empty child node # in its parent, thus, recreating tags we have to go # up one level. # <node><child/></child></node> <=> {'node': 'child': [None, None]} _to_etree(node, k) else: # If this isn't first child and it's a complex # value (dict), we need to check if it's value # is equivalent to None. if child_num != 0 and not (isinstance(e, dict) and not all(e.values())): # At least one child was None, we have to create # a new parent-node, which will not be empty. sub_elem = etree.SubElement(node, k) _to_etree(e, sub_elem) else: _to_etree(v, etree.SubElement(node, k)) elif etree.iselement(d): # Supports the case, when we got an empty child and want to recreate it. etree.SubElement(d, node) else: raise AttributeError('Argument is neither dict nor basestring.') _to_etree(d, root) return root
stxnext/mappet
mappet/helpers.py
etree_to_dict
python
def etree_to_dict(t, trim=True, **kw): u"""Converts an lxml.etree object to Python dict. >>> etree_to_dict(etree.Element('root')) {'root': None} :param etree.Element t: lxml tree to convert :returns d: a dict representing the lxml tree ``t`` :rtype: dict """ d = {t.tag: {} if t.attrib else None} children = list(t) etree_to_dict_w_args = partial(etree_to_dict, trim=trim, **kw) if children: dd = defaultdict(list) d = {t.tag: {}} for dc in map(etree_to_dict_w_args, children): for k, v in dc.iteritems(): # do not add Comment instance to the key if k is not etree.Comment: dd[k].append(v) d[t.tag] = {k: v[0] if len(v) == 1 else v for k, v in dd.iteritems()} if t.attrib: d[t.tag].update(('@' + k, v) for k, v in t.attrib.iteritems()) if trim and t.text: t.text = t.text.strip() if t.text: if t.tag is etree.Comment and not kw.get('without_comments'): # adds a comments node d['#comments'] = t.text elif children or t.attrib: d[t.tag]['#text'] = t.text else: d[t.tag] = t.text return d
u"""Converts an lxml.etree object to Python dict. >>> etree_to_dict(etree.Element('root')) {'root': None} :param etree.Element t: lxml tree to convert :returns d: a dict representing the lxml tree ``t`` :rtype: dict
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/helpers.py#L174-L212
null
# -*- coding: utf-8 -*- u"""Helper functions. .. :module: helpers :synopsis: Helper functions. """ from collections import defaultdict from decimal import Decimal from functools import partial, wraps import datetime from lxml import etree import dateutil.parser __all__ = [ 'to_bool', 'to_date', 'to_datetime', 'to_decimal', 'to_float', 'to_int', 'to_str', 'to_time', 'from_bool', 'from_date', 'from_datetime', 'from_time', 'CAST_DICT', 'normalize_tag', 'etree_to_dict', 'dict_to_etree', ] def no_empty_value(func): """Raises an exception if function argument is empty.""" @wraps(func) def wrapper(value): if not value: raise Exception("Empty value not allowed") return func(value) return wrapper def to_bool(value): """Converts human boolean-like values to Python boolean. Falls back to :class:`bool` when ``value`` is not recognized. :param value: the value to convert :returns: ``True`` if value is truthy, ``False`` otherwise :rtype: bool """ cases = { '0': False, 'false': False, 'no': False, '1': True, 'true': True, 'yes': True, } value = value.lower() if isinstance(value, basestring) else value return cases.get(value, bool(value)) def to_str(value): u"""Represents values as unicode strings to support diacritics.""" return unicode(value) def to_int(value): return int(value) def to_float(value): return float(value) def to_decimal(value): return Decimal(value) @no_empty_value def to_time(value): value = str(value) # dateutil.parse has problems parsing full hours without minutes sep = value[2:3] if not (sep == ':' or sep.isdigit()): value = value[:2] + ':00' + value[2:] return dateutil.parser.parse(value).time() @no_empty_value def to_datetime(value): return parse_datetime(value) def parse_datetime(value): value = str(value) return dateutil.parser.parse(value) @no_empty_value def to_date(value): return parse_datetime(value) def from_bool(value): cases = { True: 'YES', False: 'NO', } try: return cases.get(value, bool(value)) except Exception: return False def from_time(value): if not isinstance(value, datetime.time): raise Exception('Value {} is not datetime.time object'.format(value)) return value.isoformat() @no_empty_value def from_datetime(value): if not isinstance(value, datetime.datetime): raise Exception('Unexpected type {} of value {} (expected datetime.datetime)'.format(type(value), value)) if value.tzinfo is None: value = value.replace(tzinfo=dateutil.tz.tzlocal()) # pragma: nocover return value.replace(microsecond=0).isoformat() @no_empty_value def from_date(value): if not isinstance(value, datetime.date) and not isinstance(value, datetime.datetime): raise Exception('Not datetime.date object but {}: {}'.format(type(value), value)) return value.isoformat() CAST_DICT = { bool: from_bool, int: str, str: str, unicode: str, float: str, datetime.time: from_time, datetime.datetime: from_datetime, datetime.date: from_date, } def normalize_tag(tag): u"""Normalizes tag name. :param str tag: tag name to normalize :rtype: str :returns: normalized tag name >>> normalize_tag('tag-NaMe') 'tag_name' """ return tag.lower().replace('-', '_') def dict_to_etree(d, root): u"""Converts a dict to lxml.etree object. >>> dict_to_etree({'root': {'#text': 'node_text', '@attr': 'val'}}, etree.Element('root')) # doctest: +ELLIPSIS <Element root at 0x...> :param dict d: dict representing the XML tree :param etree.Element root: XML node which will be assigned the resulting tree :returns: Textual representation of the XML tree :rtype: str """ def _to_etree(d, node): if d is None or len(d) == 0: return elif isinstance(d, basestring): node.text = d elif isinstance(d, dict): for k, v in d.items(): assert isinstance(k, basestring) if k.startswith('#'): assert k == '#text' assert isinstance(v, basestring) node.text = v elif k.startswith('@'): assert isinstance(v, basestring) node.set(k[1:], v) elif isinstance(v, list): # No matter the child count, their parent will be the same. sub_elem = etree.SubElement(node, k) for child_num, e in enumerate(v): if e is None: if child_num == 0: # Found the first occurrence of an empty child, # skip creating of its XML repr, since it would be # the same as ``sub_element`` higher up. continue # A list with None element means an empty child node # in its parent, thus, recreating tags we have to go # up one level. # <node><child/></child></node> <=> {'node': 'child': [None, None]} _to_etree(node, k) else: # If this isn't first child and it's a complex # value (dict), we need to check if it's value # is equivalent to None. if child_num != 0 and not (isinstance(e, dict) and not all(e.values())): # At least one child was None, we have to create # a new parent-node, which will not be empty. sub_elem = etree.SubElement(node, k) _to_etree(e, sub_elem) else: _to_etree(v, etree.SubElement(node, k)) elif etree.iselement(d): # Supports the case, when we got an empty child and want to recreate it. etree.SubElement(d, node) else: raise AttributeError('Argument is neither dict nor basestring.') _to_etree(d, root) return root
stxnext/mappet
mappet/helpers.py
dict_to_etree
python
def dict_to_etree(d, root): u"""Converts a dict to lxml.etree object. >>> dict_to_etree({'root': {'#text': 'node_text', '@attr': 'val'}}, etree.Element('root')) # doctest: +ELLIPSIS <Element root at 0x...> :param dict d: dict representing the XML tree :param etree.Element root: XML node which will be assigned the resulting tree :returns: Textual representation of the XML tree :rtype: str """ def _to_etree(d, node): if d is None or len(d) == 0: return elif isinstance(d, basestring): node.text = d elif isinstance(d, dict): for k, v in d.items(): assert isinstance(k, basestring) if k.startswith('#'): assert k == '#text' assert isinstance(v, basestring) node.text = v elif k.startswith('@'): assert isinstance(v, basestring) node.set(k[1:], v) elif isinstance(v, list): # No matter the child count, their parent will be the same. sub_elem = etree.SubElement(node, k) for child_num, e in enumerate(v): if e is None: if child_num == 0: # Found the first occurrence of an empty child, # skip creating of its XML repr, since it would be # the same as ``sub_element`` higher up. continue # A list with None element means an empty child node # in its parent, thus, recreating tags we have to go # up one level. # <node><child/></child></node> <=> {'node': 'child': [None, None]} _to_etree(node, k) else: # If this isn't first child and it's a complex # value (dict), we need to check if it's value # is equivalent to None. if child_num != 0 and not (isinstance(e, dict) and not all(e.values())): # At least one child was None, we have to create # a new parent-node, which will not be empty. sub_elem = etree.SubElement(node, k) _to_etree(e, sub_elem) else: _to_etree(v, etree.SubElement(node, k)) elif etree.iselement(d): # Supports the case, when we got an empty child and want to recreate it. etree.SubElement(d, node) else: raise AttributeError('Argument is neither dict nor basestring.') _to_etree(d, root) return root
u"""Converts a dict to lxml.etree object. >>> dict_to_etree({'root': {'#text': 'node_text', '@attr': 'val'}}, etree.Element('root')) # doctest: +ELLIPSIS <Element root at 0x...> :param dict d: dict representing the XML tree :param etree.Element root: XML node which will be assigned the resulting tree :returns: Textual representation of the XML tree :rtype: str
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/helpers.py#L215-L275
[ "def _to_etree(d, node):\n if d is None or len(d) == 0:\n return\n elif isinstance(d, basestring):\n node.text = d\n elif isinstance(d, dict):\n for k, v in d.items():\n assert isinstance(k, basestring)\n if k.startswith('#'):\n assert k == '#text'\n assert isinstance(v, basestring)\n node.text = v\n elif k.startswith('@'):\n assert isinstance(v, basestring)\n node.set(k[1:], v)\n elif isinstance(v, list):\n # No matter the child count, their parent will be the same.\n sub_elem = etree.SubElement(node, k)\n\n for child_num, e in enumerate(v):\n if e is None:\n if child_num == 0:\n # Found the first occurrence of an empty child,\n # skip creating of its XML repr, since it would be\n # the same as ``sub_element`` higher up.\n continue\n # A list with None element means an empty child node\n # in its parent, thus, recreating tags we have to go\n # up one level.\n # <node><child/></child></node> <=> {'node': 'child': [None, None]}\n _to_etree(node, k)\n else:\n # If this isn't first child and it's a complex\n # value (dict), we need to check if it's value\n # is equivalent to None.\n if child_num != 0 and not (isinstance(e, dict) and not all(e.values())):\n # At least one child was None, we have to create\n # a new parent-node, which will not be empty.\n sub_elem = etree.SubElement(node, k)\n _to_etree(e, sub_elem)\n else:\n _to_etree(v, etree.SubElement(node, k))\n elif etree.iselement(d):\n # Supports the case, when we got an empty child and want to recreate it.\n etree.SubElement(d, node)\n else:\n raise AttributeError('Argument is neither dict nor basestring.')\n" ]
# -*- coding: utf-8 -*- u"""Helper functions. .. :module: helpers :synopsis: Helper functions. """ from collections import defaultdict from decimal import Decimal from functools import partial, wraps import datetime from lxml import etree import dateutil.parser __all__ = [ 'to_bool', 'to_date', 'to_datetime', 'to_decimal', 'to_float', 'to_int', 'to_str', 'to_time', 'from_bool', 'from_date', 'from_datetime', 'from_time', 'CAST_DICT', 'normalize_tag', 'etree_to_dict', 'dict_to_etree', ] def no_empty_value(func): """Raises an exception if function argument is empty.""" @wraps(func) def wrapper(value): if not value: raise Exception("Empty value not allowed") return func(value) return wrapper def to_bool(value): """Converts human boolean-like values to Python boolean. Falls back to :class:`bool` when ``value`` is not recognized. :param value: the value to convert :returns: ``True`` if value is truthy, ``False`` otherwise :rtype: bool """ cases = { '0': False, 'false': False, 'no': False, '1': True, 'true': True, 'yes': True, } value = value.lower() if isinstance(value, basestring) else value return cases.get(value, bool(value)) def to_str(value): u"""Represents values as unicode strings to support diacritics.""" return unicode(value) def to_int(value): return int(value) def to_float(value): return float(value) def to_decimal(value): return Decimal(value) @no_empty_value def to_time(value): value = str(value) # dateutil.parse has problems parsing full hours without minutes sep = value[2:3] if not (sep == ':' or sep.isdigit()): value = value[:2] + ':00' + value[2:] return dateutil.parser.parse(value).time() @no_empty_value def to_datetime(value): return parse_datetime(value) def parse_datetime(value): value = str(value) return dateutil.parser.parse(value) @no_empty_value def to_date(value): return parse_datetime(value) def from_bool(value): cases = { True: 'YES', False: 'NO', } try: return cases.get(value, bool(value)) except Exception: return False def from_time(value): if not isinstance(value, datetime.time): raise Exception('Value {} is not datetime.time object'.format(value)) return value.isoformat() @no_empty_value def from_datetime(value): if not isinstance(value, datetime.datetime): raise Exception('Unexpected type {} of value {} (expected datetime.datetime)'.format(type(value), value)) if value.tzinfo is None: value = value.replace(tzinfo=dateutil.tz.tzlocal()) # pragma: nocover return value.replace(microsecond=0).isoformat() @no_empty_value def from_date(value): if not isinstance(value, datetime.date) and not isinstance(value, datetime.datetime): raise Exception('Not datetime.date object but {}: {}'.format(type(value), value)) return value.isoformat() CAST_DICT = { bool: from_bool, int: str, str: str, unicode: str, float: str, datetime.time: from_time, datetime.datetime: from_datetime, datetime.date: from_date, } def normalize_tag(tag): u"""Normalizes tag name. :param str tag: tag name to normalize :rtype: str :returns: normalized tag name >>> normalize_tag('tag-NaMe') 'tag_name' """ return tag.lower().replace('-', '_') def etree_to_dict(t, trim=True, **kw): u"""Converts an lxml.etree object to Python dict. >>> etree_to_dict(etree.Element('root')) {'root': None} :param etree.Element t: lxml tree to convert :returns d: a dict representing the lxml tree ``t`` :rtype: dict """ d = {t.tag: {} if t.attrib else None} children = list(t) etree_to_dict_w_args = partial(etree_to_dict, trim=trim, **kw) if children: dd = defaultdict(list) d = {t.tag: {}} for dc in map(etree_to_dict_w_args, children): for k, v in dc.iteritems(): # do not add Comment instance to the key if k is not etree.Comment: dd[k].append(v) d[t.tag] = {k: v[0] if len(v) == 1 else v for k, v in dd.iteritems()} if t.attrib: d[t.tag].update(('@' + k, v) for k, v in t.attrib.iteritems()) if trim and t.text: t.text = t.text.strip() if t.text: if t.tag is etree.Comment and not kw.get('without_comments'): # adds a comments node d['#comments'] = t.text elif children or t.attrib: d[t.tag]['#text'] = t.text else: d[t.tag] = t.text return d
stxnext/mappet
mappet/mappet.py
Node.getattr
python
def getattr(self, key, default=None, callback=None): u"""Getting the attribute of an element. >>> xml = etree.Element('root') >>> xml.text = 'text' >>> Node(xml).getattr('text') 'text' >>> Node(xml).getattr('text', callback=str.upper) 'TEXT' >>> Node(xml).getattr('wrong_attr', default='default') 'default' """ value = self._xml.text if key == 'text' else self._xml.get(key, default) return callback(value) if callback else value
u"""Getting the attribute of an element. >>> xml = etree.Element('root') >>> xml.text = 'text' >>> Node(xml).getattr('text') 'text' >>> Node(xml).getattr('text', callback=str.upper) 'TEXT' >>> Node(xml).getattr('wrong_attr', default='default') 'default'
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/mappet.py#L75-L88
null
class Node(object): u"""Base class representing an XML node.""" #: The lxml object representing parsed XML. _xml = None def __init__(self, xml): self._xml = xml def __repr__(self): u"""Represent an XML node as a string with child count. >>> xml = etree.Element('root') >>> xml.set('attr1', 'val1') >>> _ = etree.SubElement(xml, 'child') >>> repr(Node(xml)) '<root attr1="val1"> (1)' """ return '<{tagname}{attributes}{closing_paren}> ({children})'.format( tagname=self._xml.tag, attributes=''.join( [' {}="{}"'.format( attr, self._xml.attrib[attr] ) for attr in self._xml.attrib] ), closing_paren='' if len(self._xml) else '/', children=len(self._xml) ) def __getitem__(self, key): u"""Call to a list element. Only calls to node attributes (i.e. starting with `@`) or text nodes (starting with `#`) are allowed. >>> xml = etree.Element('root') >>> xml.set('attr1', 'val1') >>> _ = etree.SubElement(xml, 'child') >>> Node(xml)[0] Traceback (most recent call last): ... KeyError: 0 >>> Node(xml)['@attr1'] 'val1' """ if self.is_key_attr_or_text(key): return self.getattr(key[1:]) raise KeyError(key) def setattr(self, key, value): u"""Sets an attribute on a node. >>> xml = etree.Element('root') >>> Node(xml).setattr('text', 'text2') >>> Node(xml).getattr('text') 'text2' >>> Node(xml).setattr('attr', 'val') >>> Node(xml).getattr('attr') 'val' """ if key == 'text': self._xml.text = str(value) else: self._xml.set(key, str(value)) @property def tag(self): u"""Returns node's tag name.""" return self._xml.tag @staticmethod def is_key_attr_or_text(key): return isinstance(key, basestring) and key.startswith(('@', '#'))
stxnext/mappet
mappet/mappet.py
Node.setattr
python
def setattr(self, key, value): u"""Sets an attribute on a node. >>> xml = etree.Element('root') >>> Node(xml).setattr('text', 'text2') >>> Node(xml).getattr('text') 'text2' >>> Node(xml).setattr('attr', 'val') >>> Node(xml).getattr('attr') 'val' """ if key == 'text': self._xml.text = str(value) else: self._xml.set(key, str(value))
u"""Sets an attribute on a node. >>> xml = etree.Element('root') >>> Node(xml).setattr('text', 'text2') >>> Node(xml).getattr('text') 'text2' >>> Node(xml).setattr('attr', 'val') >>> Node(xml).getattr('attr') 'val'
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/mappet.py#L90-L104
null
class Node(object): u"""Base class representing an XML node.""" #: The lxml object representing parsed XML. _xml = None def __init__(self, xml): self._xml = xml def __repr__(self): u"""Represent an XML node as a string with child count. >>> xml = etree.Element('root') >>> xml.set('attr1', 'val1') >>> _ = etree.SubElement(xml, 'child') >>> repr(Node(xml)) '<root attr1="val1"> (1)' """ return '<{tagname}{attributes}{closing_paren}> ({children})'.format( tagname=self._xml.tag, attributes=''.join( [' {}="{}"'.format( attr, self._xml.attrib[attr] ) for attr in self._xml.attrib] ), closing_paren='' if len(self._xml) else '/', children=len(self._xml) ) def __getitem__(self, key): u"""Call to a list element. Only calls to node attributes (i.e. starting with `@`) or text nodes (starting with `#`) are allowed. >>> xml = etree.Element('root') >>> xml.set('attr1', 'val1') >>> _ = etree.SubElement(xml, 'child') >>> Node(xml)[0] Traceback (most recent call last): ... KeyError: 0 >>> Node(xml)['@attr1'] 'val1' """ if self.is_key_attr_or_text(key): return self.getattr(key[1:]) raise KeyError(key) def getattr(self, key, default=None, callback=None): u"""Getting the attribute of an element. >>> xml = etree.Element('root') >>> xml.text = 'text' >>> Node(xml).getattr('text') 'text' >>> Node(xml).getattr('text', callback=str.upper) 'TEXT' >>> Node(xml).getattr('wrong_attr', default='default') 'default' """ value = self._xml.text if key == 'text' else self._xml.get(key, default) return callback(value) if callback else value @property def tag(self): u"""Returns node's tag name.""" return self._xml.tag @staticmethod def is_key_attr_or_text(key): return isinstance(key, basestring) and key.startswith(('@', '#'))
stxnext/mappet
mappet/mappet.py
Literal.get
python
def get(self, default=None, callback=None): u"""Returns leaf's value.""" value = self._xml.text if self._xml.text else default return callback(value) if callback else value
u"""Returns leaf's value.
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/mappet.py#L209-L212
null
class Literal(Node): u"""Represents a leaf in an XML tree.""" def __str__(self): u"""Represents a leaf as a str. Returns node's text as a string, if it's not None.""" return str(self.get()) #: Represents the leaf as unicode. __unicode__ = __str__ def __repr__(self): u"""Represents the leaf as its textual value.""" return str(self) def __int__(self): u"""Represents the literal as an int.""" return self.to_int() def __float__(self): u"""Represents the literal as an float.""" return self.to_float() def __nonzero__(self): u"""Represents the literal as an bool.""" return True if self._xml.text else False def __eq__(self, other): u"""Compares two leafs. Assumes they are equal if the same are their: * tagname, * parent, * text, * attributes, * position among parent's children. """ self_parent = self._xml.getparent() other_parent = other._xml.getparent() is_same_tag = self._xml.tag == other._xml.tag is_same_parent = self_parent == other_parent is_same_text = str(self) == str(other) are_attrs_equal = (self._xml.attrib == other._xml.attrib) is_same_position = self_parent.index(self._xml) == other_parent.index(other._xml) return all(( is_same_tag, is_same_parent, is_same_text, are_attrs_equal, is_same_position, )) def __hash__(self): return hash(self._xml) def __len__(self): u"""Returns the length of node's text.""" return len(self._xml.text) @staticmethod def __dir__(): u"""Lists available casting methods.""" return sorted(set([fnc for fnc in helpers.__all__ if fnc.startswith('to_')])) def __getattr__(self, name): u"""Returns a function for converting node's value. A leaf has no children, thus accessing its attributes returns a function. """ if name.startswith('to_') and name in dir(helpers): fn = getattr(helpers, name) return lambda: fn(self._xml.text) raise AttributeError(name) def __setitem__(self, key, value): u"""Attribute assignment by dict access. Extending the leaf in this case is not possible, since a string is returned. """ if self.is_key_attr_or_text(key): self.setattr(key[1:], value) def __add__(self, other): u"""String concatenation.""" return self.to_str() + str(other) def __radd__(self, other): u"""Reverse string concatenation.""" return str(other) + self.to_str()
stxnext/mappet
mappet/mappet.py
Mappet.to_str
python
def to_str(self, pretty_print=False, encoding=None, **kw): u"""Converts a node with all of it's children to a string. Remaining arguments are passed to etree.tostring as is. kwarg without_comments: bool because it works only in C14N flags: 'pretty print' and 'encoding' are ignored. :param bool pretty_print: whether to format the output :param str encoding: which encoding to use (ASCII by default) :rtype: str :returns: node's representation as a string """ if kw.get('without_comments') and not kw.get('method'): kw.pop('without_comments') kw['method'] = 'c14n' kw['with_comments'] = False return etree.tostring( self._xml, pretty_print=pretty_print, encoding=encoding, **kw )
u"""Converts a node with all of it's children to a string. Remaining arguments are passed to etree.tostring as is. kwarg without_comments: bool because it works only in C14N flags: 'pretty print' and 'encoding' are ignored. :param bool pretty_print: whether to format the output :param str encoding: which encoding to use (ASCII by default) :rtype: str :returns: node's representation as a string
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/mappet.py#L384-L406
null
class Mappet(Node): u"""A node that may have children.""" _aliases = None u"""Dictionary with node aliases. The keys are normalized tagnames, values are the original tagnames. _aliases = { 'car_model_desc': 'car-model-desc', 'car': 'Car', } """ def __init__(self, xml): u"""Creates the mappet object from either lxml object, a string or a dict. If you pass a dict without root element, one will be created for you with 'root' as tag name. >>> Mappet({'a': {'#text': 'list_elem_1', '@attr1': 'val1'}}).to_str() '<a attr1="val1">list_elem_1</a>' >>> Mappet({'#text': 'list_elem_1', '@attr1': 'val1'}).to_str() '<root attr1="val1">list_elem_1</root>' """ if etree.iselement(xml): self._xml = xml elif isinstance(xml, basestring): self._xml = etree.fromstring(xml) elif isinstance(xml, dict): if len(xml) == 1: root_name = xml.keys()[0] body = xml[root_name] else: root_name = 'root' body = xml self._xml = helpers.dict_to_etree(body, etree.Element(root_name)) else: raise AttributeError('Specified data cannot be used to construct a Mappet object.') def __nonzero__(self): u"""Checks if this node has children, otherwise returns False.""" return self.has_children() def __len__(self): u"""Returns the children count.""" return len(self._xml) def __dir__(self): u"""Returns a list of children and available helper methods.""" return sorted(self.keys() | {m for m in dir(self.__class__) if m.startswith('to_')}) def __deepcopy__(self, memodict): u"""Performs a deepcopy on the underlying XML tree.""" return self.__class__(deepcopy(self._xml)) def __getattr__(self, name): u"""Attribute access. Returns a list o children, if there is more than 1. Returns a child, if there is exactly 1. """ children = self.children(name) if len(children) > 1: return children elif len(children) == 1: return children[0] def __setattr__(self, name, value): u"""Node attribute assignment. Calls ``set`` in the end. """ # Only elements that aren't a part of class definition are overwritten. if name not in dir(self.__class__): return self.set(name, value) return super(Mappet, self).__setattr__(name, value) def __delattr__(self, key): u"""Node removal.""" # Searches among aliases, if none is found returns the original key. tag = self._get_aliases().get(key, key) self.__delitem__(tag) def __getitem__(self, key): u"""Dictionary access.""" # Checks if the call isn't to an attribute. if isinstance(key, basestring) and not key.startswith('@'): children = self.children(key) if len(children) == 1: children = children[0] # Return the value if it's a leaf. if isinstance(children, Literal): return children.get() return children return super(Mappet, self).__getitem__(key) def __delitem__(self, key): u"""Removes all children with a given key.""" # Checks if name is not a part of class definition. if key not in dir(self.__class__): for child in self._xml.iterchildren(tag=key): self._xml.remove(child) def __eq__(self, other): u"""Compares mappet objects. Two mappet objects are deemed equal if the lxmls object they represent are equal. """ return etree.tostring(self._xml) == etree.tostring(other._xml) def __contains__(self, path): u"""Check if object contains given path.""" elem = self.sget(path) return not (elem is None or elem is NONE_NODE) def __getstate__(self): u"""Converts the lxml to string for Pickling.""" return { '_xml': etree.tostring(self._xml, pretty_print=False) } def __setstate__(self, dict_): u"""Restores a Pickled mappet object.""" self._xml = etree.fromstring(dict_['_xml']) def __iter__(self): u"""Returns children as an iterator.""" return self.iter_children() def has_children(self): u"""Returns true if a node has children.""" return bool(len(self)) def iter_children(self, key=None): u"""Iterates over children. :param key: A key for filtering children by tagname. """ tag = None if key: tag = self._get_aliases().get(key) if not tag: raise KeyError(key) for child in self._xml.iterchildren(tag=tag): if len(child): yield self.__class__(child) else: yield Literal(child) def children(self, key=None): u"""Returns node's children. :param key: A key for filtering children by tagname. """ return list(self.iter_children(key)) def update(self, **kwargs): u"""Updating or creation of new simple nodes. Each dict key is used as a tagname and value as text. """ for key, value in kwargs.items(): helper = helpers.CAST_DICT.get(type(value), str) tag = self._get_aliases().get(key, key) elements = list(self._xml.iterchildren(tag=tag)) if elements: for element in elements: element.text = helper(value) else: element = etree.Element(key) element.text = helper(value) self._xml.append(element) self._aliases = None def sget(self, path, default=NONE_NODE): u"""Enables access to nodes if one or more of them don't exist. Example: >>> m = Mappet('<root><tag attr1="attr text">text value</tag></root>') >>> m.sget('tag') text value >>> m.sget('tag.@attr1') 'attr text' >>> m.sget('tag.#text') 'text value' >>> m.sget('reply.vms_model_cars.car.0.params.doors') NONE_NODE Accessing nonexistent path returns None-like object with mocked converting functions which returns None: >>> m.sget('reply.fake_node').to_dict() is None True """ attrs = str(path).split(".") text_or_attr = None last_attr = attrs[-1] # Case of getting text or attribute if last_attr == '#text' or last_attr.startswith('@'): # #text => text, @attr => attr text_or_attr = last_attr[1:] attrs = attrs[:-1] # When getting #text and @attr we want default value to be None. if default is NONE_NODE: default = None my_object = self for attr in attrs: try: if isinstance(my_object, (list, tuple)) and re.match('^\-?\d+$', attr): my_object_next = my_object[int(attr)] else: my_object_next = getattr(my_object, attr) my_object = my_object_next except (AttributeError, KeyError, IndexError): return default # Return #text or @attr if text_or_attr: try: return my_object.getattr(text_or_attr) except AttributeError: # myObject can be a list. return None else: return my_object def create(self, tag, value): u"""Creates a node, if it doesn't exist yet. Unlike attribute access, this allows to pass a node's name with hyphens. Those hyphens will be normalized automatically. In case the required element already exists, raises an exception. Updating/overwriting should be done using `update``. """ child_tags = {child.tag for child in self._xml} if tag in child_tags: raise KeyError('Node {} already exists in XML tree.'.format(tag)) self.set(tag, value) def set(self, name, value): u"""Assigns a new XML structure to the node. A literal value, dict or list can be passed in. Works for all nested levels. Dictionary: >>> m = Mappet('<root/>') >>> m.head = {'a': 'A', 'b': {'#text': 'B', '@attr': 'val'}} >>> m.head.to_str() '<head><a>A</a><b attr="val">B</b></head>' List: >>> m.head = [{'a': i} for i in 'ABC'] >>> m.head.to_str() '<head><a>A</a><a>B</a><a>C</a></head>' Literals: >>> m.head.leaf = 'A' >>> m.head.leaf.get() 'A' """ try: # Searches for a node to assign to. element = next(self._xml.iterchildren(tag=name)) except StopIteration: # There is no such node in the XML tree. We create a new one # with current root as parent (self._xml). element = etree.SubElement(self._xml, name) if isinstance(value, dict): self.assign_dict(element, value) elif isinstance(value, (list, tuple, set)): self.assign_sequence_or_set(element, value) else: # Literal value. self.assign_literal(element, value) # Clear the aliases. self._aliases = None def assign_dict(self, node, xml_dict): """Assigns a Python dict to a ``lxml`` node. :param node: A node to assign the dict to. :param xml_dict: The dict with attributes/children to use. """ new_node = etree.Element(node.tag) # Replaces the previous node with the new one self._xml.replace(node, new_node) # Copies #text and @attrs from the xml_dict helpers.dict_to_etree(xml_dict, new_node) @staticmethod def assign_sequence_or_set(element, value): element.clear() for item in value: temp_element = etree.Element('temp') helpers.dict_to_etree(item, temp_element) for child in temp_element.iterchildren(): element.append(child) del temp_element @staticmethod def assign_literal(element, value): u"""Assigns a literal. If a given node doesn't exist, it will be created. :param etree.Element element: element to which we assign. :param value: the value to assign """ # Searches for a conversion method specific to the type of value. helper = helpers.CAST_DICT.get(type(value), str) # Removes all children and attributes. element.clear() element.text = helper(value) def to_dict(self, **kw): u"""Converts the lxml object to a dict. possible kwargs: without_comments: bool """ _, value = helpers.etree_to_dict(self._xml, **kw).popitem() return value def _get_aliases(self): u"""Creates a dict with aliases. The key is a normalized tagname, value the original tagname. """ if self._aliases is None: self._aliases = {} if self._xml is not None: for child in self._xml.iterchildren(): self._aliases[helpers.normalize_tag(child.tag)] = child.tag return self._aliases def xpath( self, path, namespaces=None, regexp=False, smart_strings=True, single_use=False, ): u"""Executes XPath query on the ``lxml`` object and returns a correct object. :param str path: XPath string e.g., 'cars'/'car' :param str/dict namespaces: e.g., 'exslt', 're' or ``{'re': "http://exslt.org/regular-expressions"}`` :param bool regexp: if ``True`` and no namespaces is provided, it will use ``exslt`` namespace :param bool smart_strings: :param bool single_use: faster method for using only once. Does not create ``XPathEvaluator`` instance. >>> root = mappet.Mappet("<root><a>aB</a><b>aBc</b></root>") >>> root.XPath( "//*[re:test(., '^abc$', 'i')]", namespaces='exslt', regexp=True, ) """ if ( namespaces in ['exslt', 're'] or (regexp and not namespaces) ): namespaces = {'re': "http://exslt.org/regular-expressions"} if single_use: node = self._xml.xpath(path) else: xpe = self.xpath_evaluator( namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) node = xpe(path) if len(node) == 1: node = node[0] if len(node): return self.__class__(node) else: return Literal(node) return node def xpath_evaluator(self, namespaces=None, regexp=False, smart_strings=True): u"""Creates an XPathEvaluator instance for an ElementTree or an Element. :returns: ``XPathEvaluator`` instance """ return etree.XPathEvaluator( self._xml, namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) def keys(self): """Returns a set of node's keys.""" return set(self._get_aliases().keys())
stxnext/mappet
mappet/mappet.py
Mappet.iter_children
python
def iter_children(self, key=None): u"""Iterates over children. :param key: A key for filtering children by tagname. """ tag = None if key: tag = self._get_aliases().get(key) if not tag: raise KeyError(key) for child in self._xml.iterchildren(tag=tag): if len(child): yield self.__class__(child) else: yield Literal(child)
u"""Iterates over children. :param key: A key for filtering children by tagname.
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/mappet.py#L412-L429
[ "def _get_aliases(self):\n u\"\"\"Creates a dict with aliases.\n\n The key is a normalized tagname, value the original tagname.\n \"\"\"\n if self._aliases is None:\n self._aliases = {}\n\n if self._xml is not None:\n for child in self._xml.iterchildren():\n self._aliases[helpers.normalize_tag(child.tag)] = child.tag\n\n return self._aliases\n" ]
class Mappet(Node): u"""A node that may have children.""" _aliases = None u"""Dictionary with node aliases. The keys are normalized tagnames, values are the original tagnames. _aliases = { 'car_model_desc': 'car-model-desc', 'car': 'Car', } """ def __init__(self, xml): u"""Creates the mappet object from either lxml object, a string or a dict. If you pass a dict without root element, one will be created for you with 'root' as tag name. >>> Mappet({'a': {'#text': 'list_elem_1', '@attr1': 'val1'}}).to_str() '<a attr1="val1">list_elem_1</a>' >>> Mappet({'#text': 'list_elem_1', '@attr1': 'val1'}).to_str() '<root attr1="val1">list_elem_1</root>' """ if etree.iselement(xml): self._xml = xml elif isinstance(xml, basestring): self._xml = etree.fromstring(xml) elif isinstance(xml, dict): if len(xml) == 1: root_name = xml.keys()[0] body = xml[root_name] else: root_name = 'root' body = xml self._xml = helpers.dict_to_etree(body, etree.Element(root_name)) else: raise AttributeError('Specified data cannot be used to construct a Mappet object.') def __nonzero__(self): u"""Checks if this node has children, otherwise returns False.""" return self.has_children() def __len__(self): u"""Returns the children count.""" return len(self._xml) def __dir__(self): u"""Returns a list of children and available helper methods.""" return sorted(self.keys() | {m for m in dir(self.__class__) if m.startswith('to_')}) def __deepcopy__(self, memodict): u"""Performs a deepcopy on the underlying XML tree.""" return self.__class__(deepcopy(self._xml)) def __getattr__(self, name): u"""Attribute access. Returns a list o children, if there is more than 1. Returns a child, if there is exactly 1. """ children = self.children(name) if len(children) > 1: return children elif len(children) == 1: return children[0] def __setattr__(self, name, value): u"""Node attribute assignment. Calls ``set`` in the end. """ # Only elements that aren't a part of class definition are overwritten. if name not in dir(self.__class__): return self.set(name, value) return super(Mappet, self).__setattr__(name, value) def __delattr__(self, key): u"""Node removal.""" # Searches among aliases, if none is found returns the original key. tag = self._get_aliases().get(key, key) self.__delitem__(tag) def __getitem__(self, key): u"""Dictionary access.""" # Checks if the call isn't to an attribute. if isinstance(key, basestring) and not key.startswith('@'): children = self.children(key) if len(children) == 1: children = children[0] # Return the value if it's a leaf. if isinstance(children, Literal): return children.get() return children return super(Mappet, self).__getitem__(key) def __delitem__(self, key): u"""Removes all children with a given key.""" # Checks if name is not a part of class definition. if key not in dir(self.__class__): for child in self._xml.iterchildren(tag=key): self._xml.remove(child) def __eq__(self, other): u"""Compares mappet objects. Two mappet objects are deemed equal if the lxmls object they represent are equal. """ return etree.tostring(self._xml) == etree.tostring(other._xml) def __contains__(self, path): u"""Check if object contains given path.""" elem = self.sget(path) return not (elem is None or elem is NONE_NODE) def __getstate__(self): u"""Converts the lxml to string for Pickling.""" return { '_xml': etree.tostring(self._xml, pretty_print=False) } def __setstate__(self, dict_): u"""Restores a Pickled mappet object.""" self._xml = etree.fromstring(dict_['_xml']) def __iter__(self): u"""Returns children as an iterator.""" return self.iter_children() def to_str(self, pretty_print=False, encoding=None, **kw): u"""Converts a node with all of it's children to a string. Remaining arguments are passed to etree.tostring as is. kwarg without_comments: bool because it works only in C14N flags: 'pretty print' and 'encoding' are ignored. :param bool pretty_print: whether to format the output :param str encoding: which encoding to use (ASCII by default) :rtype: str :returns: node's representation as a string """ if kw.get('without_comments') and not kw.get('method'): kw.pop('without_comments') kw['method'] = 'c14n' kw['with_comments'] = False return etree.tostring( self._xml, pretty_print=pretty_print, encoding=encoding, **kw ) def has_children(self): u"""Returns true if a node has children.""" return bool(len(self)) def children(self, key=None): u"""Returns node's children. :param key: A key for filtering children by tagname. """ return list(self.iter_children(key)) def update(self, **kwargs): u"""Updating or creation of new simple nodes. Each dict key is used as a tagname and value as text. """ for key, value in kwargs.items(): helper = helpers.CAST_DICT.get(type(value), str) tag = self._get_aliases().get(key, key) elements = list(self._xml.iterchildren(tag=tag)) if elements: for element in elements: element.text = helper(value) else: element = etree.Element(key) element.text = helper(value) self._xml.append(element) self._aliases = None def sget(self, path, default=NONE_NODE): u"""Enables access to nodes if one or more of them don't exist. Example: >>> m = Mappet('<root><tag attr1="attr text">text value</tag></root>') >>> m.sget('tag') text value >>> m.sget('tag.@attr1') 'attr text' >>> m.sget('tag.#text') 'text value' >>> m.sget('reply.vms_model_cars.car.0.params.doors') NONE_NODE Accessing nonexistent path returns None-like object with mocked converting functions which returns None: >>> m.sget('reply.fake_node').to_dict() is None True """ attrs = str(path).split(".") text_or_attr = None last_attr = attrs[-1] # Case of getting text or attribute if last_attr == '#text' or last_attr.startswith('@'): # #text => text, @attr => attr text_or_attr = last_attr[1:] attrs = attrs[:-1] # When getting #text and @attr we want default value to be None. if default is NONE_NODE: default = None my_object = self for attr in attrs: try: if isinstance(my_object, (list, tuple)) and re.match('^\-?\d+$', attr): my_object_next = my_object[int(attr)] else: my_object_next = getattr(my_object, attr) my_object = my_object_next except (AttributeError, KeyError, IndexError): return default # Return #text or @attr if text_or_attr: try: return my_object.getattr(text_or_attr) except AttributeError: # myObject can be a list. return None else: return my_object def create(self, tag, value): u"""Creates a node, if it doesn't exist yet. Unlike attribute access, this allows to pass a node's name with hyphens. Those hyphens will be normalized automatically. In case the required element already exists, raises an exception. Updating/overwriting should be done using `update``. """ child_tags = {child.tag for child in self._xml} if tag in child_tags: raise KeyError('Node {} already exists in XML tree.'.format(tag)) self.set(tag, value) def set(self, name, value): u"""Assigns a new XML structure to the node. A literal value, dict or list can be passed in. Works for all nested levels. Dictionary: >>> m = Mappet('<root/>') >>> m.head = {'a': 'A', 'b': {'#text': 'B', '@attr': 'val'}} >>> m.head.to_str() '<head><a>A</a><b attr="val">B</b></head>' List: >>> m.head = [{'a': i} for i in 'ABC'] >>> m.head.to_str() '<head><a>A</a><a>B</a><a>C</a></head>' Literals: >>> m.head.leaf = 'A' >>> m.head.leaf.get() 'A' """ try: # Searches for a node to assign to. element = next(self._xml.iterchildren(tag=name)) except StopIteration: # There is no such node in the XML tree. We create a new one # with current root as parent (self._xml). element = etree.SubElement(self._xml, name) if isinstance(value, dict): self.assign_dict(element, value) elif isinstance(value, (list, tuple, set)): self.assign_sequence_or_set(element, value) else: # Literal value. self.assign_literal(element, value) # Clear the aliases. self._aliases = None def assign_dict(self, node, xml_dict): """Assigns a Python dict to a ``lxml`` node. :param node: A node to assign the dict to. :param xml_dict: The dict with attributes/children to use. """ new_node = etree.Element(node.tag) # Replaces the previous node with the new one self._xml.replace(node, new_node) # Copies #text and @attrs from the xml_dict helpers.dict_to_etree(xml_dict, new_node) @staticmethod def assign_sequence_or_set(element, value): element.clear() for item in value: temp_element = etree.Element('temp') helpers.dict_to_etree(item, temp_element) for child in temp_element.iterchildren(): element.append(child) del temp_element @staticmethod def assign_literal(element, value): u"""Assigns a literal. If a given node doesn't exist, it will be created. :param etree.Element element: element to which we assign. :param value: the value to assign """ # Searches for a conversion method specific to the type of value. helper = helpers.CAST_DICT.get(type(value), str) # Removes all children and attributes. element.clear() element.text = helper(value) def to_dict(self, **kw): u"""Converts the lxml object to a dict. possible kwargs: without_comments: bool """ _, value = helpers.etree_to_dict(self._xml, **kw).popitem() return value def _get_aliases(self): u"""Creates a dict with aliases. The key is a normalized tagname, value the original tagname. """ if self._aliases is None: self._aliases = {} if self._xml is not None: for child in self._xml.iterchildren(): self._aliases[helpers.normalize_tag(child.tag)] = child.tag return self._aliases def xpath( self, path, namespaces=None, regexp=False, smart_strings=True, single_use=False, ): u"""Executes XPath query on the ``lxml`` object and returns a correct object. :param str path: XPath string e.g., 'cars'/'car' :param str/dict namespaces: e.g., 'exslt', 're' or ``{'re': "http://exslt.org/regular-expressions"}`` :param bool regexp: if ``True`` and no namespaces is provided, it will use ``exslt`` namespace :param bool smart_strings: :param bool single_use: faster method for using only once. Does not create ``XPathEvaluator`` instance. >>> root = mappet.Mappet("<root><a>aB</a><b>aBc</b></root>") >>> root.XPath( "//*[re:test(., '^abc$', 'i')]", namespaces='exslt', regexp=True, ) """ if ( namespaces in ['exslt', 're'] or (regexp and not namespaces) ): namespaces = {'re': "http://exslt.org/regular-expressions"} if single_use: node = self._xml.xpath(path) else: xpe = self.xpath_evaluator( namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) node = xpe(path) if len(node) == 1: node = node[0] if len(node): return self.__class__(node) else: return Literal(node) return node def xpath_evaluator(self, namespaces=None, regexp=False, smart_strings=True): u"""Creates an XPathEvaluator instance for an ElementTree or an Element. :returns: ``XPathEvaluator`` instance """ return etree.XPathEvaluator( self._xml, namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) def keys(self): """Returns a set of node's keys.""" return set(self._get_aliases().keys())
stxnext/mappet
mappet/mappet.py
Mappet.update
python
def update(self, **kwargs): u"""Updating or creation of new simple nodes. Each dict key is used as a tagname and value as text. """ for key, value in kwargs.items(): helper = helpers.CAST_DICT.get(type(value), str) tag = self._get_aliases().get(key, key) elements = list(self._xml.iterchildren(tag=tag)) if elements: for element in elements: element.text = helper(value) else: element = etree.Element(key) element.text = helper(value) self._xml.append(element) self._aliases = None
u"""Updating or creation of new simple nodes. Each dict key is used as a tagname and value as text.
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/mappet.py#L438-L455
[ "def _get_aliases(self):\n u\"\"\"Creates a dict with aliases.\n\n The key is a normalized tagname, value the original tagname.\n \"\"\"\n if self._aliases is None:\n self._aliases = {}\n\n if self._xml is not None:\n for child in self._xml.iterchildren():\n self._aliases[helpers.normalize_tag(child.tag)] = child.tag\n\n return self._aliases\n" ]
class Mappet(Node): u"""A node that may have children.""" _aliases = None u"""Dictionary with node aliases. The keys are normalized tagnames, values are the original tagnames. _aliases = { 'car_model_desc': 'car-model-desc', 'car': 'Car', } """ def __init__(self, xml): u"""Creates the mappet object from either lxml object, a string or a dict. If you pass a dict without root element, one will be created for you with 'root' as tag name. >>> Mappet({'a': {'#text': 'list_elem_1', '@attr1': 'val1'}}).to_str() '<a attr1="val1">list_elem_1</a>' >>> Mappet({'#text': 'list_elem_1', '@attr1': 'val1'}).to_str() '<root attr1="val1">list_elem_1</root>' """ if etree.iselement(xml): self._xml = xml elif isinstance(xml, basestring): self._xml = etree.fromstring(xml) elif isinstance(xml, dict): if len(xml) == 1: root_name = xml.keys()[0] body = xml[root_name] else: root_name = 'root' body = xml self._xml = helpers.dict_to_etree(body, etree.Element(root_name)) else: raise AttributeError('Specified data cannot be used to construct a Mappet object.') def __nonzero__(self): u"""Checks if this node has children, otherwise returns False.""" return self.has_children() def __len__(self): u"""Returns the children count.""" return len(self._xml) def __dir__(self): u"""Returns a list of children and available helper methods.""" return sorted(self.keys() | {m for m in dir(self.__class__) if m.startswith('to_')}) def __deepcopy__(self, memodict): u"""Performs a deepcopy on the underlying XML tree.""" return self.__class__(deepcopy(self._xml)) def __getattr__(self, name): u"""Attribute access. Returns a list o children, if there is more than 1. Returns a child, if there is exactly 1. """ children = self.children(name) if len(children) > 1: return children elif len(children) == 1: return children[0] def __setattr__(self, name, value): u"""Node attribute assignment. Calls ``set`` in the end. """ # Only elements that aren't a part of class definition are overwritten. if name not in dir(self.__class__): return self.set(name, value) return super(Mappet, self).__setattr__(name, value) def __delattr__(self, key): u"""Node removal.""" # Searches among aliases, if none is found returns the original key. tag = self._get_aliases().get(key, key) self.__delitem__(tag) def __getitem__(self, key): u"""Dictionary access.""" # Checks if the call isn't to an attribute. if isinstance(key, basestring) and not key.startswith('@'): children = self.children(key) if len(children) == 1: children = children[0] # Return the value if it's a leaf. if isinstance(children, Literal): return children.get() return children return super(Mappet, self).__getitem__(key) def __delitem__(self, key): u"""Removes all children with a given key.""" # Checks if name is not a part of class definition. if key not in dir(self.__class__): for child in self._xml.iterchildren(tag=key): self._xml.remove(child) def __eq__(self, other): u"""Compares mappet objects. Two mappet objects are deemed equal if the lxmls object they represent are equal. """ return etree.tostring(self._xml) == etree.tostring(other._xml) def __contains__(self, path): u"""Check if object contains given path.""" elem = self.sget(path) return not (elem is None or elem is NONE_NODE) def __getstate__(self): u"""Converts the lxml to string for Pickling.""" return { '_xml': etree.tostring(self._xml, pretty_print=False) } def __setstate__(self, dict_): u"""Restores a Pickled mappet object.""" self._xml = etree.fromstring(dict_['_xml']) def __iter__(self): u"""Returns children as an iterator.""" return self.iter_children() def to_str(self, pretty_print=False, encoding=None, **kw): u"""Converts a node with all of it's children to a string. Remaining arguments are passed to etree.tostring as is. kwarg without_comments: bool because it works only in C14N flags: 'pretty print' and 'encoding' are ignored. :param bool pretty_print: whether to format the output :param str encoding: which encoding to use (ASCII by default) :rtype: str :returns: node's representation as a string """ if kw.get('without_comments') and not kw.get('method'): kw.pop('without_comments') kw['method'] = 'c14n' kw['with_comments'] = False return etree.tostring( self._xml, pretty_print=pretty_print, encoding=encoding, **kw ) def has_children(self): u"""Returns true if a node has children.""" return bool(len(self)) def iter_children(self, key=None): u"""Iterates over children. :param key: A key for filtering children by tagname. """ tag = None if key: tag = self._get_aliases().get(key) if not tag: raise KeyError(key) for child in self._xml.iterchildren(tag=tag): if len(child): yield self.__class__(child) else: yield Literal(child) def children(self, key=None): u"""Returns node's children. :param key: A key for filtering children by tagname. """ return list(self.iter_children(key)) def sget(self, path, default=NONE_NODE): u"""Enables access to nodes if one or more of them don't exist. Example: >>> m = Mappet('<root><tag attr1="attr text">text value</tag></root>') >>> m.sget('tag') text value >>> m.sget('tag.@attr1') 'attr text' >>> m.sget('tag.#text') 'text value' >>> m.sget('reply.vms_model_cars.car.0.params.doors') NONE_NODE Accessing nonexistent path returns None-like object with mocked converting functions which returns None: >>> m.sget('reply.fake_node').to_dict() is None True """ attrs = str(path).split(".") text_or_attr = None last_attr = attrs[-1] # Case of getting text or attribute if last_attr == '#text' or last_attr.startswith('@'): # #text => text, @attr => attr text_or_attr = last_attr[1:] attrs = attrs[:-1] # When getting #text and @attr we want default value to be None. if default is NONE_NODE: default = None my_object = self for attr in attrs: try: if isinstance(my_object, (list, tuple)) and re.match('^\-?\d+$', attr): my_object_next = my_object[int(attr)] else: my_object_next = getattr(my_object, attr) my_object = my_object_next except (AttributeError, KeyError, IndexError): return default # Return #text or @attr if text_or_attr: try: return my_object.getattr(text_or_attr) except AttributeError: # myObject can be a list. return None else: return my_object def create(self, tag, value): u"""Creates a node, if it doesn't exist yet. Unlike attribute access, this allows to pass a node's name with hyphens. Those hyphens will be normalized automatically. In case the required element already exists, raises an exception. Updating/overwriting should be done using `update``. """ child_tags = {child.tag for child in self._xml} if tag in child_tags: raise KeyError('Node {} already exists in XML tree.'.format(tag)) self.set(tag, value) def set(self, name, value): u"""Assigns a new XML structure to the node. A literal value, dict or list can be passed in. Works for all nested levels. Dictionary: >>> m = Mappet('<root/>') >>> m.head = {'a': 'A', 'b': {'#text': 'B', '@attr': 'val'}} >>> m.head.to_str() '<head><a>A</a><b attr="val">B</b></head>' List: >>> m.head = [{'a': i} for i in 'ABC'] >>> m.head.to_str() '<head><a>A</a><a>B</a><a>C</a></head>' Literals: >>> m.head.leaf = 'A' >>> m.head.leaf.get() 'A' """ try: # Searches for a node to assign to. element = next(self._xml.iterchildren(tag=name)) except StopIteration: # There is no such node in the XML tree. We create a new one # with current root as parent (self._xml). element = etree.SubElement(self._xml, name) if isinstance(value, dict): self.assign_dict(element, value) elif isinstance(value, (list, tuple, set)): self.assign_sequence_or_set(element, value) else: # Literal value. self.assign_literal(element, value) # Clear the aliases. self._aliases = None def assign_dict(self, node, xml_dict): """Assigns a Python dict to a ``lxml`` node. :param node: A node to assign the dict to. :param xml_dict: The dict with attributes/children to use. """ new_node = etree.Element(node.tag) # Replaces the previous node with the new one self._xml.replace(node, new_node) # Copies #text and @attrs from the xml_dict helpers.dict_to_etree(xml_dict, new_node) @staticmethod def assign_sequence_or_set(element, value): element.clear() for item in value: temp_element = etree.Element('temp') helpers.dict_to_etree(item, temp_element) for child in temp_element.iterchildren(): element.append(child) del temp_element @staticmethod def assign_literal(element, value): u"""Assigns a literal. If a given node doesn't exist, it will be created. :param etree.Element element: element to which we assign. :param value: the value to assign """ # Searches for a conversion method specific to the type of value. helper = helpers.CAST_DICT.get(type(value), str) # Removes all children and attributes. element.clear() element.text = helper(value) def to_dict(self, **kw): u"""Converts the lxml object to a dict. possible kwargs: without_comments: bool """ _, value = helpers.etree_to_dict(self._xml, **kw).popitem() return value def _get_aliases(self): u"""Creates a dict with aliases. The key is a normalized tagname, value the original tagname. """ if self._aliases is None: self._aliases = {} if self._xml is not None: for child in self._xml.iterchildren(): self._aliases[helpers.normalize_tag(child.tag)] = child.tag return self._aliases def xpath( self, path, namespaces=None, regexp=False, smart_strings=True, single_use=False, ): u"""Executes XPath query on the ``lxml`` object and returns a correct object. :param str path: XPath string e.g., 'cars'/'car' :param str/dict namespaces: e.g., 'exslt', 're' or ``{'re': "http://exslt.org/regular-expressions"}`` :param bool regexp: if ``True`` and no namespaces is provided, it will use ``exslt`` namespace :param bool smart_strings: :param bool single_use: faster method for using only once. Does not create ``XPathEvaluator`` instance. >>> root = mappet.Mappet("<root><a>aB</a><b>aBc</b></root>") >>> root.XPath( "//*[re:test(., '^abc$', 'i')]", namespaces='exslt', regexp=True, ) """ if ( namespaces in ['exslt', 're'] or (regexp and not namespaces) ): namespaces = {'re': "http://exslt.org/regular-expressions"} if single_use: node = self._xml.xpath(path) else: xpe = self.xpath_evaluator( namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) node = xpe(path) if len(node) == 1: node = node[0] if len(node): return self.__class__(node) else: return Literal(node) return node def xpath_evaluator(self, namespaces=None, regexp=False, smart_strings=True): u"""Creates an XPathEvaluator instance for an ElementTree or an Element. :returns: ``XPathEvaluator`` instance """ return etree.XPathEvaluator( self._xml, namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) def keys(self): """Returns a set of node's keys.""" return set(self._get_aliases().keys())
stxnext/mappet
mappet/mappet.py
Mappet.sget
python
def sget(self, path, default=NONE_NODE): u"""Enables access to nodes if one or more of them don't exist. Example: >>> m = Mappet('<root><tag attr1="attr text">text value</tag></root>') >>> m.sget('tag') text value >>> m.sget('tag.@attr1') 'attr text' >>> m.sget('tag.#text') 'text value' >>> m.sget('reply.vms_model_cars.car.0.params.doors') NONE_NODE Accessing nonexistent path returns None-like object with mocked converting functions which returns None: >>> m.sget('reply.fake_node').to_dict() is None True """ attrs = str(path).split(".") text_or_attr = None last_attr = attrs[-1] # Case of getting text or attribute if last_attr == '#text' or last_attr.startswith('@'): # #text => text, @attr => attr text_or_attr = last_attr[1:] attrs = attrs[:-1] # When getting #text and @attr we want default value to be None. if default is NONE_NODE: default = None my_object = self for attr in attrs: try: if isinstance(my_object, (list, tuple)) and re.match('^\-?\d+$', attr): my_object_next = my_object[int(attr)] else: my_object_next = getattr(my_object, attr) my_object = my_object_next except (AttributeError, KeyError, IndexError): return default # Return #text or @attr if text_or_attr: try: return my_object.getattr(text_or_attr) except AttributeError: # myObject can be a list. return None else: return my_object
u"""Enables access to nodes if one or more of them don't exist. Example: >>> m = Mappet('<root><tag attr1="attr text">text value</tag></root>') >>> m.sget('tag') text value >>> m.sget('tag.@attr1') 'attr text' >>> m.sget('tag.#text') 'text value' >>> m.sget('reply.vms_model_cars.car.0.params.doors') NONE_NODE Accessing nonexistent path returns None-like object with mocked converting functions which returns None: >>> m.sget('reply.fake_node').to_dict() is None True
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/mappet.py#L457-L507
[ "def getattr(self, key, default=None, callback=None):\n u\"\"\"Getting the attribute of an element.\n\n >>> xml = etree.Element('root')\n >>> xml.text = 'text'\n >>> Node(xml).getattr('text')\n 'text'\n >>> Node(xml).getattr('text', callback=str.upper)\n 'TEXT'\n >>> Node(xml).getattr('wrong_attr', default='default')\n 'default'\n \"\"\"\n value = self._xml.text if key == 'text' else self._xml.get(key, default)\n return callback(value) if callback else value\n" ]
class Mappet(Node): u"""A node that may have children.""" _aliases = None u"""Dictionary with node aliases. The keys are normalized tagnames, values are the original tagnames. _aliases = { 'car_model_desc': 'car-model-desc', 'car': 'Car', } """ def __init__(self, xml): u"""Creates the mappet object from either lxml object, a string or a dict. If you pass a dict without root element, one will be created for you with 'root' as tag name. >>> Mappet({'a': {'#text': 'list_elem_1', '@attr1': 'val1'}}).to_str() '<a attr1="val1">list_elem_1</a>' >>> Mappet({'#text': 'list_elem_1', '@attr1': 'val1'}).to_str() '<root attr1="val1">list_elem_1</root>' """ if etree.iselement(xml): self._xml = xml elif isinstance(xml, basestring): self._xml = etree.fromstring(xml) elif isinstance(xml, dict): if len(xml) == 1: root_name = xml.keys()[0] body = xml[root_name] else: root_name = 'root' body = xml self._xml = helpers.dict_to_etree(body, etree.Element(root_name)) else: raise AttributeError('Specified data cannot be used to construct a Mappet object.') def __nonzero__(self): u"""Checks if this node has children, otherwise returns False.""" return self.has_children() def __len__(self): u"""Returns the children count.""" return len(self._xml) def __dir__(self): u"""Returns a list of children and available helper methods.""" return sorted(self.keys() | {m for m in dir(self.__class__) if m.startswith('to_')}) def __deepcopy__(self, memodict): u"""Performs a deepcopy on the underlying XML tree.""" return self.__class__(deepcopy(self._xml)) def __getattr__(self, name): u"""Attribute access. Returns a list o children, if there is more than 1. Returns a child, if there is exactly 1. """ children = self.children(name) if len(children) > 1: return children elif len(children) == 1: return children[0] def __setattr__(self, name, value): u"""Node attribute assignment. Calls ``set`` in the end. """ # Only elements that aren't a part of class definition are overwritten. if name not in dir(self.__class__): return self.set(name, value) return super(Mappet, self).__setattr__(name, value) def __delattr__(self, key): u"""Node removal.""" # Searches among aliases, if none is found returns the original key. tag = self._get_aliases().get(key, key) self.__delitem__(tag) def __getitem__(self, key): u"""Dictionary access.""" # Checks if the call isn't to an attribute. if isinstance(key, basestring) and not key.startswith('@'): children = self.children(key) if len(children) == 1: children = children[0] # Return the value if it's a leaf. if isinstance(children, Literal): return children.get() return children return super(Mappet, self).__getitem__(key) def __delitem__(self, key): u"""Removes all children with a given key.""" # Checks if name is not a part of class definition. if key not in dir(self.__class__): for child in self._xml.iterchildren(tag=key): self._xml.remove(child) def __eq__(self, other): u"""Compares mappet objects. Two mappet objects are deemed equal if the lxmls object they represent are equal. """ return etree.tostring(self._xml) == etree.tostring(other._xml) def __contains__(self, path): u"""Check if object contains given path.""" elem = self.sget(path) return not (elem is None or elem is NONE_NODE) def __getstate__(self): u"""Converts the lxml to string for Pickling.""" return { '_xml': etree.tostring(self._xml, pretty_print=False) } def __setstate__(self, dict_): u"""Restores a Pickled mappet object.""" self._xml = etree.fromstring(dict_['_xml']) def __iter__(self): u"""Returns children as an iterator.""" return self.iter_children() def to_str(self, pretty_print=False, encoding=None, **kw): u"""Converts a node with all of it's children to a string. Remaining arguments are passed to etree.tostring as is. kwarg without_comments: bool because it works only in C14N flags: 'pretty print' and 'encoding' are ignored. :param bool pretty_print: whether to format the output :param str encoding: which encoding to use (ASCII by default) :rtype: str :returns: node's representation as a string """ if kw.get('without_comments') and not kw.get('method'): kw.pop('without_comments') kw['method'] = 'c14n' kw['with_comments'] = False return etree.tostring( self._xml, pretty_print=pretty_print, encoding=encoding, **kw ) def has_children(self): u"""Returns true if a node has children.""" return bool(len(self)) def iter_children(self, key=None): u"""Iterates over children. :param key: A key for filtering children by tagname. """ tag = None if key: tag = self._get_aliases().get(key) if not tag: raise KeyError(key) for child in self._xml.iterchildren(tag=tag): if len(child): yield self.__class__(child) else: yield Literal(child) def children(self, key=None): u"""Returns node's children. :param key: A key for filtering children by tagname. """ return list(self.iter_children(key)) def update(self, **kwargs): u"""Updating or creation of new simple nodes. Each dict key is used as a tagname and value as text. """ for key, value in kwargs.items(): helper = helpers.CAST_DICT.get(type(value), str) tag = self._get_aliases().get(key, key) elements = list(self._xml.iterchildren(tag=tag)) if elements: for element in elements: element.text = helper(value) else: element = etree.Element(key) element.text = helper(value) self._xml.append(element) self._aliases = None def create(self, tag, value): u"""Creates a node, if it doesn't exist yet. Unlike attribute access, this allows to pass a node's name with hyphens. Those hyphens will be normalized automatically. In case the required element already exists, raises an exception. Updating/overwriting should be done using `update``. """ child_tags = {child.tag for child in self._xml} if tag in child_tags: raise KeyError('Node {} already exists in XML tree.'.format(tag)) self.set(tag, value) def set(self, name, value): u"""Assigns a new XML structure to the node. A literal value, dict or list can be passed in. Works for all nested levels. Dictionary: >>> m = Mappet('<root/>') >>> m.head = {'a': 'A', 'b': {'#text': 'B', '@attr': 'val'}} >>> m.head.to_str() '<head><a>A</a><b attr="val">B</b></head>' List: >>> m.head = [{'a': i} for i in 'ABC'] >>> m.head.to_str() '<head><a>A</a><a>B</a><a>C</a></head>' Literals: >>> m.head.leaf = 'A' >>> m.head.leaf.get() 'A' """ try: # Searches for a node to assign to. element = next(self._xml.iterchildren(tag=name)) except StopIteration: # There is no such node in the XML tree. We create a new one # with current root as parent (self._xml). element = etree.SubElement(self._xml, name) if isinstance(value, dict): self.assign_dict(element, value) elif isinstance(value, (list, tuple, set)): self.assign_sequence_or_set(element, value) else: # Literal value. self.assign_literal(element, value) # Clear the aliases. self._aliases = None def assign_dict(self, node, xml_dict): """Assigns a Python dict to a ``lxml`` node. :param node: A node to assign the dict to. :param xml_dict: The dict with attributes/children to use. """ new_node = etree.Element(node.tag) # Replaces the previous node with the new one self._xml.replace(node, new_node) # Copies #text and @attrs from the xml_dict helpers.dict_to_etree(xml_dict, new_node) @staticmethod def assign_sequence_or_set(element, value): element.clear() for item in value: temp_element = etree.Element('temp') helpers.dict_to_etree(item, temp_element) for child in temp_element.iterchildren(): element.append(child) del temp_element @staticmethod def assign_literal(element, value): u"""Assigns a literal. If a given node doesn't exist, it will be created. :param etree.Element element: element to which we assign. :param value: the value to assign """ # Searches for a conversion method specific to the type of value. helper = helpers.CAST_DICT.get(type(value), str) # Removes all children and attributes. element.clear() element.text = helper(value) def to_dict(self, **kw): u"""Converts the lxml object to a dict. possible kwargs: without_comments: bool """ _, value = helpers.etree_to_dict(self._xml, **kw).popitem() return value def _get_aliases(self): u"""Creates a dict with aliases. The key is a normalized tagname, value the original tagname. """ if self._aliases is None: self._aliases = {} if self._xml is not None: for child in self._xml.iterchildren(): self._aliases[helpers.normalize_tag(child.tag)] = child.tag return self._aliases def xpath( self, path, namespaces=None, regexp=False, smart_strings=True, single_use=False, ): u"""Executes XPath query on the ``lxml`` object and returns a correct object. :param str path: XPath string e.g., 'cars'/'car' :param str/dict namespaces: e.g., 'exslt', 're' or ``{'re': "http://exslt.org/regular-expressions"}`` :param bool regexp: if ``True`` and no namespaces is provided, it will use ``exslt`` namespace :param bool smart_strings: :param bool single_use: faster method for using only once. Does not create ``XPathEvaluator`` instance. >>> root = mappet.Mappet("<root><a>aB</a><b>aBc</b></root>") >>> root.XPath( "//*[re:test(., '^abc$', 'i')]", namespaces='exslt', regexp=True, ) """ if ( namespaces in ['exslt', 're'] or (regexp and not namespaces) ): namespaces = {'re': "http://exslt.org/regular-expressions"} if single_use: node = self._xml.xpath(path) else: xpe = self.xpath_evaluator( namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) node = xpe(path) if len(node) == 1: node = node[0] if len(node): return self.__class__(node) else: return Literal(node) return node def xpath_evaluator(self, namespaces=None, regexp=False, smart_strings=True): u"""Creates an XPathEvaluator instance for an ElementTree or an Element. :returns: ``XPathEvaluator`` instance """ return etree.XPathEvaluator( self._xml, namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) def keys(self): """Returns a set of node's keys.""" return set(self._get_aliases().keys())
stxnext/mappet
mappet/mappet.py
Mappet.create
python
def create(self, tag, value): u"""Creates a node, if it doesn't exist yet. Unlike attribute access, this allows to pass a node's name with hyphens. Those hyphens will be normalized automatically. In case the required element already exists, raises an exception. Updating/overwriting should be done using `update``. """ child_tags = {child.tag for child in self._xml} if tag in child_tags: raise KeyError('Node {} already exists in XML tree.'.format(tag)) self.set(tag, value)
u"""Creates a node, if it doesn't exist yet. Unlike attribute access, this allows to pass a node's name with hyphens. Those hyphens will be normalized automatically. In case the required element already exists, raises an exception. Updating/overwriting should be done using `update``.
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/mappet.py#L509-L523
[ "def set(self, name, value):\n u\"\"\"Assigns a new XML structure to the node.\n\n A literal value, dict or list can be passed in. Works for all nested levels.\n\n Dictionary:\n >>> m = Mappet('<root/>')\n >>> m.head = {'a': 'A', 'b': {'#text': 'B', '@attr': 'val'}}\n >>> m.head.to_str()\n '<head><a>A</a><b attr=\"val\">B</b></head>'\n\n List:\n >>> m.head = [{'a': i} for i in 'ABC']\n >>> m.head.to_str()\n '<head><a>A</a><a>B</a><a>C</a></head>'\n\n Literals:\n >>> m.head.leaf = 'A'\n >>> m.head.leaf.get()\n 'A'\n \"\"\"\n try:\n # Searches for a node to assign to.\n element = next(self._xml.iterchildren(tag=name))\n except StopIteration:\n # There is no such node in the XML tree. We create a new one\n # with current root as parent (self._xml).\n element = etree.SubElement(self._xml, name)\n\n if isinstance(value, dict):\n self.assign_dict(element, value)\n elif isinstance(value, (list, tuple, set)):\n self.assign_sequence_or_set(element, value)\n else:\n # Literal value.\n self.assign_literal(element, value)\n\n # Clear the aliases.\n self._aliases = None\n" ]
class Mappet(Node): u"""A node that may have children.""" _aliases = None u"""Dictionary with node aliases. The keys are normalized tagnames, values are the original tagnames. _aliases = { 'car_model_desc': 'car-model-desc', 'car': 'Car', } """ def __init__(self, xml): u"""Creates the mappet object from either lxml object, a string or a dict. If you pass a dict without root element, one will be created for you with 'root' as tag name. >>> Mappet({'a': {'#text': 'list_elem_1', '@attr1': 'val1'}}).to_str() '<a attr1="val1">list_elem_1</a>' >>> Mappet({'#text': 'list_elem_1', '@attr1': 'val1'}).to_str() '<root attr1="val1">list_elem_1</root>' """ if etree.iselement(xml): self._xml = xml elif isinstance(xml, basestring): self._xml = etree.fromstring(xml) elif isinstance(xml, dict): if len(xml) == 1: root_name = xml.keys()[0] body = xml[root_name] else: root_name = 'root' body = xml self._xml = helpers.dict_to_etree(body, etree.Element(root_name)) else: raise AttributeError('Specified data cannot be used to construct a Mappet object.') def __nonzero__(self): u"""Checks if this node has children, otherwise returns False.""" return self.has_children() def __len__(self): u"""Returns the children count.""" return len(self._xml) def __dir__(self): u"""Returns a list of children and available helper methods.""" return sorted(self.keys() | {m for m in dir(self.__class__) if m.startswith('to_')}) def __deepcopy__(self, memodict): u"""Performs a deepcopy on the underlying XML tree.""" return self.__class__(deepcopy(self._xml)) def __getattr__(self, name): u"""Attribute access. Returns a list o children, if there is more than 1. Returns a child, if there is exactly 1. """ children = self.children(name) if len(children) > 1: return children elif len(children) == 1: return children[0] def __setattr__(self, name, value): u"""Node attribute assignment. Calls ``set`` in the end. """ # Only elements that aren't a part of class definition are overwritten. if name not in dir(self.__class__): return self.set(name, value) return super(Mappet, self).__setattr__(name, value) def __delattr__(self, key): u"""Node removal.""" # Searches among aliases, if none is found returns the original key. tag = self._get_aliases().get(key, key) self.__delitem__(tag) def __getitem__(self, key): u"""Dictionary access.""" # Checks if the call isn't to an attribute. if isinstance(key, basestring) and not key.startswith('@'): children = self.children(key) if len(children) == 1: children = children[0] # Return the value if it's a leaf. if isinstance(children, Literal): return children.get() return children return super(Mappet, self).__getitem__(key) def __delitem__(self, key): u"""Removes all children with a given key.""" # Checks if name is not a part of class definition. if key not in dir(self.__class__): for child in self._xml.iterchildren(tag=key): self._xml.remove(child) def __eq__(self, other): u"""Compares mappet objects. Two mappet objects are deemed equal if the lxmls object they represent are equal. """ return etree.tostring(self._xml) == etree.tostring(other._xml) def __contains__(self, path): u"""Check if object contains given path.""" elem = self.sget(path) return not (elem is None or elem is NONE_NODE) def __getstate__(self): u"""Converts the lxml to string for Pickling.""" return { '_xml': etree.tostring(self._xml, pretty_print=False) } def __setstate__(self, dict_): u"""Restores a Pickled mappet object.""" self._xml = etree.fromstring(dict_['_xml']) def __iter__(self): u"""Returns children as an iterator.""" return self.iter_children() def to_str(self, pretty_print=False, encoding=None, **kw): u"""Converts a node with all of it's children to a string. Remaining arguments are passed to etree.tostring as is. kwarg without_comments: bool because it works only in C14N flags: 'pretty print' and 'encoding' are ignored. :param bool pretty_print: whether to format the output :param str encoding: which encoding to use (ASCII by default) :rtype: str :returns: node's representation as a string """ if kw.get('without_comments') and not kw.get('method'): kw.pop('without_comments') kw['method'] = 'c14n' kw['with_comments'] = False return etree.tostring( self._xml, pretty_print=pretty_print, encoding=encoding, **kw ) def has_children(self): u"""Returns true if a node has children.""" return bool(len(self)) def iter_children(self, key=None): u"""Iterates over children. :param key: A key for filtering children by tagname. """ tag = None if key: tag = self._get_aliases().get(key) if not tag: raise KeyError(key) for child in self._xml.iterchildren(tag=tag): if len(child): yield self.__class__(child) else: yield Literal(child) def children(self, key=None): u"""Returns node's children. :param key: A key for filtering children by tagname. """ return list(self.iter_children(key)) def update(self, **kwargs): u"""Updating or creation of new simple nodes. Each dict key is used as a tagname and value as text. """ for key, value in kwargs.items(): helper = helpers.CAST_DICT.get(type(value), str) tag = self._get_aliases().get(key, key) elements = list(self._xml.iterchildren(tag=tag)) if elements: for element in elements: element.text = helper(value) else: element = etree.Element(key) element.text = helper(value) self._xml.append(element) self._aliases = None def sget(self, path, default=NONE_NODE): u"""Enables access to nodes if one or more of them don't exist. Example: >>> m = Mappet('<root><tag attr1="attr text">text value</tag></root>') >>> m.sget('tag') text value >>> m.sget('tag.@attr1') 'attr text' >>> m.sget('tag.#text') 'text value' >>> m.sget('reply.vms_model_cars.car.0.params.doors') NONE_NODE Accessing nonexistent path returns None-like object with mocked converting functions which returns None: >>> m.sget('reply.fake_node').to_dict() is None True """ attrs = str(path).split(".") text_or_attr = None last_attr = attrs[-1] # Case of getting text or attribute if last_attr == '#text' or last_attr.startswith('@'): # #text => text, @attr => attr text_or_attr = last_attr[1:] attrs = attrs[:-1] # When getting #text and @attr we want default value to be None. if default is NONE_NODE: default = None my_object = self for attr in attrs: try: if isinstance(my_object, (list, tuple)) and re.match('^\-?\d+$', attr): my_object_next = my_object[int(attr)] else: my_object_next = getattr(my_object, attr) my_object = my_object_next except (AttributeError, KeyError, IndexError): return default # Return #text or @attr if text_or_attr: try: return my_object.getattr(text_or_attr) except AttributeError: # myObject can be a list. return None else: return my_object def set(self, name, value): u"""Assigns a new XML structure to the node. A literal value, dict or list can be passed in. Works for all nested levels. Dictionary: >>> m = Mappet('<root/>') >>> m.head = {'a': 'A', 'b': {'#text': 'B', '@attr': 'val'}} >>> m.head.to_str() '<head><a>A</a><b attr="val">B</b></head>' List: >>> m.head = [{'a': i} for i in 'ABC'] >>> m.head.to_str() '<head><a>A</a><a>B</a><a>C</a></head>' Literals: >>> m.head.leaf = 'A' >>> m.head.leaf.get() 'A' """ try: # Searches for a node to assign to. element = next(self._xml.iterchildren(tag=name)) except StopIteration: # There is no such node in the XML tree. We create a new one # with current root as parent (self._xml). element = etree.SubElement(self._xml, name) if isinstance(value, dict): self.assign_dict(element, value) elif isinstance(value, (list, tuple, set)): self.assign_sequence_or_set(element, value) else: # Literal value. self.assign_literal(element, value) # Clear the aliases. self._aliases = None def assign_dict(self, node, xml_dict): """Assigns a Python dict to a ``lxml`` node. :param node: A node to assign the dict to. :param xml_dict: The dict with attributes/children to use. """ new_node = etree.Element(node.tag) # Replaces the previous node with the new one self._xml.replace(node, new_node) # Copies #text and @attrs from the xml_dict helpers.dict_to_etree(xml_dict, new_node) @staticmethod def assign_sequence_or_set(element, value): element.clear() for item in value: temp_element = etree.Element('temp') helpers.dict_to_etree(item, temp_element) for child in temp_element.iterchildren(): element.append(child) del temp_element @staticmethod def assign_literal(element, value): u"""Assigns a literal. If a given node doesn't exist, it will be created. :param etree.Element element: element to which we assign. :param value: the value to assign """ # Searches for a conversion method specific to the type of value. helper = helpers.CAST_DICT.get(type(value), str) # Removes all children and attributes. element.clear() element.text = helper(value) def to_dict(self, **kw): u"""Converts the lxml object to a dict. possible kwargs: without_comments: bool """ _, value = helpers.etree_to_dict(self._xml, **kw).popitem() return value def _get_aliases(self): u"""Creates a dict with aliases. The key is a normalized tagname, value the original tagname. """ if self._aliases is None: self._aliases = {} if self._xml is not None: for child in self._xml.iterchildren(): self._aliases[helpers.normalize_tag(child.tag)] = child.tag return self._aliases def xpath( self, path, namespaces=None, regexp=False, smart_strings=True, single_use=False, ): u"""Executes XPath query on the ``lxml`` object and returns a correct object. :param str path: XPath string e.g., 'cars'/'car' :param str/dict namespaces: e.g., 'exslt', 're' or ``{'re': "http://exslt.org/regular-expressions"}`` :param bool regexp: if ``True`` and no namespaces is provided, it will use ``exslt`` namespace :param bool smart_strings: :param bool single_use: faster method for using only once. Does not create ``XPathEvaluator`` instance. >>> root = mappet.Mappet("<root><a>aB</a><b>aBc</b></root>") >>> root.XPath( "//*[re:test(., '^abc$', 'i')]", namespaces='exslt', regexp=True, ) """ if ( namespaces in ['exslt', 're'] or (regexp and not namespaces) ): namespaces = {'re': "http://exslt.org/regular-expressions"} if single_use: node = self._xml.xpath(path) else: xpe = self.xpath_evaluator( namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) node = xpe(path) if len(node) == 1: node = node[0] if len(node): return self.__class__(node) else: return Literal(node) return node def xpath_evaluator(self, namespaces=None, regexp=False, smart_strings=True): u"""Creates an XPathEvaluator instance for an ElementTree or an Element. :returns: ``XPathEvaluator`` instance """ return etree.XPathEvaluator( self._xml, namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) def keys(self): """Returns a set of node's keys.""" return set(self._get_aliases().keys())
stxnext/mappet
mappet/mappet.py
Mappet.set
python
def set(self, name, value): u"""Assigns a new XML structure to the node. A literal value, dict or list can be passed in. Works for all nested levels. Dictionary: >>> m = Mappet('<root/>') >>> m.head = {'a': 'A', 'b': {'#text': 'B', '@attr': 'val'}} >>> m.head.to_str() '<head><a>A</a><b attr="val">B</b></head>' List: >>> m.head = [{'a': i} for i in 'ABC'] >>> m.head.to_str() '<head><a>A</a><a>B</a><a>C</a></head>' Literals: >>> m.head.leaf = 'A' >>> m.head.leaf.get() 'A' """ try: # Searches for a node to assign to. element = next(self._xml.iterchildren(tag=name)) except StopIteration: # There is no such node in the XML tree. We create a new one # with current root as parent (self._xml). element = etree.SubElement(self._xml, name) if isinstance(value, dict): self.assign_dict(element, value) elif isinstance(value, (list, tuple, set)): self.assign_sequence_or_set(element, value) else: # Literal value. self.assign_literal(element, value) # Clear the aliases. self._aliases = None
u"""Assigns a new XML structure to the node. A literal value, dict or list can be passed in. Works for all nested levels. Dictionary: >>> m = Mappet('<root/>') >>> m.head = {'a': 'A', 'b': {'#text': 'B', '@attr': 'val'}} >>> m.head.to_str() '<head><a>A</a><b attr="val">B</b></head>' List: >>> m.head = [{'a': i} for i in 'ABC'] >>> m.head.to_str() '<head><a>A</a><a>B</a><a>C</a></head>' Literals: >>> m.head.leaf = 'A' >>> m.head.leaf.get() 'A'
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/mappet.py#L525-L563
[ "def assign_dict(self, node, xml_dict):\n \"\"\"Assigns a Python dict to a ``lxml`` node.\n\n :param node: A node to assign the dict to.\n :param xml_dict: The dict with attributes/children to use.\n \"\"\"\n new_node = etree.Element(node.tag)\n\n # Replaces the previous node with the new one\n self._xml.replace(node, new_node)\n\n # Copies #text and @attrs from the xml_dict\n helpers.dict_to_etree(xml_dict, new_node)\n", "def assign_sequence_or_set(element, value):\n element.clear()\n\n for item in value:\n temp_element = etree.Element('temp')\n helpers.dict_to_etree(item, temp_element)\n for child in temp_element.iterchildren():\n element.append(child)\n del temp_element\n", "def assign_literal(element, value):\n u\"\"\"Assigns a literal.\n\n If a given node doesn't exist, it will be created.\n\n :param etree.Element element: element to which we assign.\n :param value: the value to assign\n \"\"\"\n # Searches for a conversion method specific to the type of value.\n helper = helpers.CAST_DICT.get(type(value), str)\n\n # Removes all children and attributes.\n element.clear()\n element.text = helper(value)\n" ]
class Mappet(Node): u"""A node that may have children.""" _aliases = None u"""Dictionary with node aliases. The keys are normalized tagnames, values are the original tagnames. _aliases = { 'car_model_desc': 'car-model-desc', 'car': 'Car', } """ def __init__(self, xml): u"""Creates the mappet object from either lxml object, a string or a dict. If you pass a dict without root element, one will be created for you with 'root' as tag name. >>> Mappet({'a': {'#text': 'list_elem_1', '@attr1': 'val1'}}).to_str() '<a attr1="val1">list_elem_1</a>' >>> Mappet({'#text': 'list_elem_1', '@attr1': 'val1'}).to_str() '<root attr1="val1">list_elem_1</root>' """ if etree.iselement(xml): self._xml = xml elif isinstance(xml, basestring): self._xml = etree.fromstring(xml) elif isinstance(xml, dict): if len(xml) == 1: root_name = xml.keys()[0] body = xml[root_name] else: root_name = 'root' body = xml self._xml = helpers.dict_to_etree(body, etree.Element(root_name)) else: raise AttributeError('Specified data cannot be used to construct a Mappet object.') def __nonzero__(self): u"""Checks if this node has children, otherwise returns False.""" return self.has_children() def __len__(self): u"""Returns the children count.""" return len(self._xml) def __dir__(self): u"""Returns a list of children and available helper methods.""" return sorted(self.keys() | {m for m in dir(self.__class__) if m.startswith('to_')}) def __deepcopy__(self, memodict): u"""Performs a deepcopy on the underlying XML tree.""" return self.__class__(deepcopy(self._xml)) def __getattr__(self, name): u"""Attribute access. Returns a list o children, if there is more than 1. Returns a child, if there is exactly 1. """ children = self.children(name) if len(children) > 1: return children elif len(children) == 1: return children[0] def __setattr__(self, name, value): u"""Node attribute assignment. Calls ``set`` in the end. """ # Only elements that aren't a part of class definition are overwritten. if name not in dir(self.__class__): return self.set(name, value) return super(Mappet, self).__setattr__(name, value) def __delattr__(self, key): u"""Node removal.""" # Searches among aliases, if none is found returns the original key. tag = self._get_aliases().get(key, key) self.__delitem__(tag) def __getitem__(self, key): u"""Dictionary access.""" # Checks if the call isn't to an attribute. if isinstance(key, basestring) and not key.startswith('@'): children = self.children(key) if len(children) == 1: children = children[0] # Return the value if it's a leaf. if isinstance(children, Literal): return children.get() return children return super(Mappet, self).__getitem__(key) def __delitem__(self, key): u"""Removes all children with a given key.""" # Checks if name is not a part of class definition. if key not in dir(self.__class__): for child in self._xml.iterchildren(tag=key): self._xml.remove(child) def __eq__(self, other): u"""Compares mappet objects. Two mappet objects are deemed equal if the lxmls object they represent are equal. """ return etree.tostring(self._xml) == etree.tostring(other._xml) def __contains__(self, path): u"""Check if object contains given path.""" elem = self.sget(path) return not (elem is None or elem is NONE_NODE) def __getstate__(self): u"""Converts the lxml to string for Pickling.""" return { '_xml': etree.tostring(self._xml, pretty_print=False) } def __setstate__(self, dict_): u"""Restores a Pickled mappet object.""" self._xml = etree.fromstring(dict_['_xml']) def __iter__(self): u"""Returns children as an iterator.""" return self.iter_children() def to_str(self, pretty_print=False, encoding=None, **kw): u"""Converts a node with all of it's children to a string. Remaining arguments are passed to etree.tostring as is. kwarg without_comments: bool because it works only in C14N flags: 'pretty print' and 'encoding' are ignored. :param bool pretty_print: whether to format the output :param str encoding: which encoding to use (ASCII by default) :rtype: str :returns: node's representation as a string """ if kw.get('without_comments') and not kw.get('method'): kw.pop('without_comments') kw['method'] = 'c14n' kw['with_comments'] = False return etree.tostring( self._xml, pretty_print=pretty_print, encoding=encoding, **kw ) def has_children(self): u"""Returns true if a node has children.""" return bool(len(self)) def iter_children(self, key=None): u"""Iterates over children. :param key: A key for filtering children by tagname. """ tag = None if key: tag = self._get_aliases().get(key) if not tag: raise KeyError(key) for child in self._xml.iterchildren(tag=tag): if len(child): yield self.__class__(child) else: yield Literal(child) def children(self, key=None): u"""Returns node's children. :param key: A key for filtering children by tagname. """ return list(self.iter_children(key)) def update(self, **kwargs): u"""Updating or creation of new simple nodes. Each dict key is used as a tagname and value as text. """ for key, value in kwargs.items(): helper = helpers.CAST_DICT.get(type(value), str) tag = self._get_aliases().get(key, key) elements = list(self._xml.iterchildren(tag=tag)) if elements: for element in elements: element.text = helper(value) else: element = etree.Element(key) element.text = helper(value) self._xml.append(element) self._aliases = None def sget(self, path, default=NONE_NODE): u"""Enables access to nodes if one or more of them don't exist. Example: >>> m = Mappet('<root><tag attr1="attr text">text value</tag></root>') >>> m.sget('tag') text value >>> m.sget('tag.@attr1') 'attr text' >>> m.sget('tag.#text') 'text value' >>> m.sget('reply.vms_model_cars.car.0.params.doors') NONE_NODE Accessing nonexistent path returns None-like object with mocked converting functions which returns None: >>> m.sget('reply.fake_node').to_dict() is None True """ attrs = str(path).split(".") text_or_attr = None last_attr = attrs[-1] # Case of getting text or attribute if last_attr == '#text' or last_attr.startswith('@'): # #text => text, @attr => attr text_or_attr = last_attr[1:] attrs = attrs[:-1] # When getting #text and @attr we want default value to be None. if default is NONE_NODE: default = None my_object = self for attr in attrs: try: if isinstance(my_object, (list, tuple)) and re.match('^\-?\d+$', attr): my_object_next = my_object[int(attr)] else: my_object_next = getattr(my_object, attr) my_object = my_object_next except (AttributeError, KeyError, IndexError): return default # Return #text or @attr if text_or_attr: try: return my_object.getattr(text_or_attr) except AttributeError: # myObject can be a list. return None else: return my_object def create(self, tag, value): u"""Creates a node, if it doesn't exist yet. Unlike attribute access, this allows to pass a node's name with hyphens. Those hyphens will be normalized automatically. In case the required element already exists, raises an exception. Updating/overwriting should be done using `update``. """ child_tags = {child.tag for child in self._xml} if tag in child_tags: raise KeyError('Node {} already exists in XML tree.'.format(tag)) self.set(tag, value) def assign_dict(self, node, xml_dict): """Assigns a Python dict to a ``lxml`` node. :param node: A node to assign the dict to. :param xml_dict: The dict with attributes/children to use. """ new_node = etree.Element(node.tag) # Replaces the previous node with the new one self._xml.replace(node, new_node) # Copies #text and @attrs from the xml_dict helpers.dict_to_etree(xml_dict, new_node) @staticmethod def assign_sequence_or_set(element, value): element.clear() for item in value: temp_element = etree.Element('temp') helpers.dict_to_etree(item, temp_element) for child in temp_element.iterchildren(): element.append(child) del temp_element @staticmethod def assign_literal(element, value): u"""Assigns a literal. If a given node doesn't exist, it will be created. :param etree.Element element: element to which we assign. :param value: the value to assign """ # Searches for a conversion method specific to the type of value. helper = helpers.CAST_DICT.get(type(value), str) # Removes all children and attributes. element.clear() element.text = helper(value) def to_dict(self, **kw): u"""Converts the lxml object to a dict. possible kwargs: without_comments: bool """ _, value = helpers.etree_to_dict(self._xml, **kw).popitem() return value def _get_aliases(self): u"""Creates a dict with aliases. The key is a normalized tagname, value the original tagname. """ if self._aliases is None: self._aliases = {} if self._xml is not None: for child in self._xml.iterchildren(): self._aliases[helpers.normalize_tag(child.tag)] = child.tag return self._aliases def xpath( self, path, namespaces=None, regexp=False, smart_strings=True, single_use=False, ): u"""Executes XPath query on the ``lxml`` object and returns a correct object. :param str path: XPath string e.g., 'cars'/'car' :param str/dict namespaces: e.g., 'exslt', 're' or ``{'re': "http://exslt.org/regular-expressions"}`` :param bool regexp: if ``True`` and no namespaces is provided, it will use ``exslt`` namespace :param bool smart_strings: :param bool single_use: faster method for using only once. Does not create ``XPathEvaluator`` instance. >>> root = mappet.Mappet("<root><a>aB</a><b>aBc</b></root>") >>> root.XPath( "//*[re:test(., '^abc$', 'i')]", namespaces='exslt', regexp=True, ) """ if ( namespaces in ['exslt', 're'] or (regexp and not namespaces) ): namespaces = {'re': "http://exslt.org/regular-expressions"} if single_use: node = self._xml.xpath(path) else: xpe = self.xpath_evaluator( namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) node = xpe(path) if len(node) == 1: node = node[0] if len(node): return self.__class__(node) else: return Literal(node) return node def xpath_evaluator(self, namespaces=None, regexp=False, smart_strings=True): u"""Creates an XPathEvaluator instance for an ElementTree or an Element. :returns: ``XPathEvaluator`` instance """ return etree.XPathEvaluator( self._xml, namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) def keys(self): """Returns a set of node's keys.""" return set(self._get_aliases().keys())
stxnext/mappet
mappet/mappet.py
Mappet.assign_dict
python
def assign_dict(self, node, xml_dict): new_node = etree.Element(node.tag) # Replaces the previous node with the new one self._xml.replace(node, new_node) # Copies #text and @attrs from the xml_dict helpers.dict_to_etree(xml_dict, new_node)
Assigns a Python dict to a ``lxml`` node. :param node: A node to assign the dict to. :param xml_dict: The dict with attributes/children to use.
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/mappet.py#L565-L577
[ "def dict_to_etree(d, root):\n u\"\"\"Converts a dict to lxml.etree object.\n\n >>> dict_to_etree({'root': {'#text': 'node_text', '@attr': 'val'}}, etree.Element('root')) # doctest: +ELLIPSIS\n <Element root at 0x...>\n\n :param dict d: dict representing the XML tree\n :param etree.Element root: XML node which will be assigned the resulting tree\n :returns: Textual representation of the XML tree\n :rtype: str\n \"\"\"\n def _to_etree(d, node):\n if d is None or len(d) == 0:\n return\n elif isinstance(d, basestring):\n node.text = d\n elif isinstance(d, dict):\n for k, v in d.items():\n assert isinstance(k, basestring)\n if k.startswith('#'):\n assert k == '#text'\n assert isinstance(v, basestring)\n node.text = v\n elif k.startswith('@'):\n assert isinstance(v, basestring)\n node.set(k[1:], v)\n elif isinstance(v, list):\n # No matter the child count, their parent will be the same.\n sub_elem = etree.SubElement(node, k)\n\n for child_num, e in enumerate(v):\n if e is None:\n if child_num == 0:\n # Found the first occurrence of an empty child,\n # skip creating of its XML repr, since it would be\n # the same as ``sub_element`` higher up.\n continue\n # A list with None element means an empty child node\n # in its parent, thus, recreating tags we have to go\n # up one level.\n # <node><child/></child></node> <=> {'node': 'child': [None, None]}\n _to_etree(node, k)\n else:\n # If this isn't first child and it's a complex\n # value (dict), we need to check if it's value\n # is equivalent to None.\n if child_num != 0 and not (isinstance(e, dict) and not all(e.values())):\n # At least one child was None, we have to create\n # a new parent-node, which will not be empty.\n sub_elem = etree.SubElement(node, k)\n _to_etree(e, sub_elem)\n else:\n _to_etree(v, etree.SubElement(node, k))\n elif etree.iselement(d):\n # Supports the case, when we got an empty child and want to recreate it.\n etree.SubElement(d, node)\n else:\n raise AttributeError('Argument is neither dict nor basestring.')\n\n _to_etree(d, root)\n return root\n" ]
class Mappet(Node): u"""A node that may have children.""" _aliases = None u"""Dictionary with node aliases. The keys are normalized tagnames, values are the original tagnames. _aliases = { 'car_model_desc': 'car-model-desc', 'car': 'Car', } """ def __init__(self, xml): u"""Creates the mappet object from either lxml object, a string or a dict. If you pass a dict without root element, one will be created for you with 'root' as tag name. >>> Mappet({'a': {'#text': 'list_elem_1', '@attr1': 'val1'}}).to_str() '<a attr1="val1">list_elem_1</a>' >>> Mappet({'#text': 'list_elem_1', '@attr1': 'val1'}).to_str() '<root attr1="val1">list_elem_1</root>' """ if etree.iselement(xml): self._xml = xml elif isinstance(xml, basestring): self._xml = etree.fromstring(xml) elif isinstance(xml, dict): if len(xml) == 1: root_name = xml.keys()[0] body = xml[root_name] else: root_name = 'root' body = xml self._xml = helpers.dict_to_etree(body, etree.Element(root_name)) else: raise AttributeError('Specified data cannot be used to construct a Mappet object.') def __nonzero__(self): u"""Checks if this node has children, otherwise returns False.""" return self.has_children() def __len__(self): u"""Returns the children count.""" return len(self._xml) def __dir__(self): u"""Returns a list of children and available helper methods.""" return sorted(self.keys() | {m for m in dir(self.__class__) if m.startswith('to_')}) def __deepcopy__(self, memodict): u"""Performs a deepcopy on the underlying XML tree.""" return self.__class__(deepcopy(self._xml)) def __getattr__(self, name): u"""Attribute access. Returns a list o children, if there is more than 1. Returns a child, if there is exactly 1. """ children = self.children(name) if len(children) > 1: return children elif len(children) == 1: return children[0] def __setattr__(self, name, value): u"""Node attribute assignment. Calls ``set`` in the end. """ # Only elements that aren't a part of class definition are overwritten. if name not in dir(self.__class__): return self.set(name, value) return super(Mappet, self).__setattr__(name, value) def __delattr__(self, key): u"""Node removal.""" # Searches among aliases, if none is found returns the original key. tag = self._get_aliases().get(key, key) self.__delitem__(tag) def __getitem__(self, key): u"""Dictionary access.""" # Checks if the call isn't to an attribute. if isinstance(key, basestring) and not key.startswith('@'): children = self.children(key) if len(children) == 1: children = children[0] # Return the value if it's a leaf. if isinstance(children, Literal): return children.get() return children return super(Mappet, self).__getitem__(key) def __delitem__(self, key): u"""Removes all children with a given key.""" # Checks if name is not a part of class definition. if key not in dir(self.__class__): for child in self._xml.iterchildren(tag=key): self._xml.remove(child) def __eq__(self, other): u"""Compares mappet objects. Two mappet objects are deemed equal if the lxmls object they represent are equal. """ return etree.tostring(self._xml) == etree.tostring(other._xml) def __contains__(self, path): u"""Check if object contains given path.""" elem = self.sget(path) return not (elem is None or elem is NONE_NODE) def __getstate__(self): u"""Converts the lxml to string for Pickling.""" return { '_xml': etree.tostring(self._xml, pretty_print=False) } def __setstate__(self, dict_): u"""Restores a Pickled mappet object.""" self._xml = etree.fromstring(dict_['_xml']) def __iter__(self): u"""Returns children as an iterator.""" return self.iter_children() def to_str(self, pretty_print=False, encoding=None, **kw): u"""Converts a node with all of it's children to a string. Remaining arguments are passed to etree.tostring as is. kwarg without_comments: bool because it works only in C14N flags: 'pretty print' and 'encoding' are ignored. :param bool pretty_print: whether to format the output :param str encoding: which encoding to use (ASCII by default) :rtype: str :returns: node's representation as a string """ if kw.get('without_comments') and not kw.get('method'): kw.pop('without_comments') kw['method'] = 'c14n' kw['with_comments'] = False return etree.tostring( self._xml, pretty_print=pretty_print, encoding=encoding, **kw ) def has_children(self): u"""Returns true if a node has children.""" return bool(len(self)) def iter_children(self, key=None): u"""Iterates over children. :param key: A key for filtering children by tagname. """ tag = None if key: tag = self._get_aliases().get(key) if not tag: raise KeyError(key) for child in self._xml.iterchildren(tag=tag): if len(child): yield self.__class__(child) else: yield Literal(child) def children(self, key=None): u"""Returns node's children. :param key: A key for filtering children by tagname. """ return list(self.iter_children(key)) def update(self, **kwargs): u"""Updating or creation of new simple nodes. Each dict key is used as a tagname and value as text. """ for key, value in kwargs.items(): helper = helpers.CAST_DICT.get(type(value), str) tag = self._get_aliases().get(key, key) elements = list(self._xml.iterchildren(tag=tag)) if elements: for element in elements: element.text = helper(value) else: element = etree.Element(key) element.text = helper(value) self._xml.append(element) self._aliases = None def sget(self, path, default=NONE_NODE): u"""Enables access to nodes if one or more of them don't exist. Example: >>> m = Mappet('<root><tag attr1="attr text">text value</tag></root>') >>> m.sget('tag') text value >>> m.sget('tag.@attr1') 'attr text' >>> m.sget('tag.#text') 'text value' >>> m.sget('reply.vms_model_cars.car.0.params.doors') NONE_NODE Accessing nonexistent path returns None-like object with mocked converting functions which returns None: >>> m.sget('reply.fake_node').to_dict() is None True """ attrs = str(path).split(".") text_or_attr = None last_attr = attrs[-1] # Case of getting text or attribute if last_attr == '#text' or last_attr.startswith('@'): # #text => text, @attr => attr text_or_attr = last_attr[1:] attrs = attrs[:-1] # When getting #text and @attr we want default value to be None. if default is NONE_NODE: default = None my_object = self for attr in attrs: try: if isinstance(my_object, (list, tuple)) and re.match('^\-?\d+$', attr): my_object_next = my_object[int(attr)] else: my_object_next = getattr(my_object, attr) my_object = my_object_next except (AttributeError, KeyError, IndexError): return default # Return #text or @attr if text_or_attr: try: return my_object.getattr(text_or_attr) except AttributeError: # myObject can be a list. return None else: return my_object def create(self, tag, value): u"""Creates a node, if it doesn't exist yet. Unlike attribute access, this allows to pass a node's name with hyphens. Those hyphens will be normalized automatically. In case the required element already exists, raises an exception. Updating/overwriting should be done using `update``. """ child_tags = {child.tag for child in self._xml} if tag in child_tags: raise KeyError('Node {} already exists in XML tree.'.format(tag)) self.set(tag, value) def set(self, name, value): u"""Assigns a new XML structure to the node. A literal value, dict or list can be passed in. Works for all nested levels. Dictionary: >>> m = Mappet('<root/>') >>> m.head = {'a': 'A', 'b': {'#text': 'B', '@attr': 'val'}} >>> m.head.to_str() '<head><a>A</a><b attr="val">B</b></head>' List: >>> m.head = [{'a': i} for i in 'ABC'] >>> m.head.to_str() '<head><a>A</a><a>B</a><a>C</a></head>' Literals: >>> m.head.leaf = 'A' >>> m.head.leaf.get() 'A' """ try: # Searches for a node to assign to. element = next(self._xml.iterchildren(tag=name)) except StopIteration: # There is no such node in the XML tree. We create a new one # with current root as parent (self._xml). element = etree.SubElement(self._xml, name) if isinstance(value, dict): self.assign_dict(element, value) elif isinstance(value, (list, tuple, set)): self.assign_sequence_or_set(element, value) else: # Literal value. self.assign_literal(element, value) # Clear the aliases. self._aliases = None @staticmethod def assign_sequence_or_set(element, value): element.clear() for item in value: temp_element = etree.Element('temp') helpers.dict_to_etree(item, temp_element) for child in temp_element.iterchildren(): element.append(child) del temp_element @staticmethod def assign_literal(element, value): u"""Assigns a literal. If a given node doesn't exist, it will be created. :param etree.Element element: element to which we assign. :param value: the value to assign """ # Searches for a conversion method specific to the type of value. helper = helpers.CAST_DICT.get(type(value), str) # Removes all children and attributes. element.clear() element.text = helper(value) def to_dict(self, **kw): u"""Converts the lxml object to a dict. possible kwargs: without_comments: bool """ _, value = helpers.etree_to_dict(self._xml, **kw).popitem() return value def _get_aliases(self): u"""Creates a dict with aliases. The key is a normalized tagname, value the original tagname. """ if self._aliases is None: self._aliases = {} if self._xml is not None: for child in self._xml.iterchildren(): self._aliases[helpers.normalize_tag(child.tag)] = child.tag return self._aliases def xpath( self, path, namespaces=None, regexp=False, smart_strings=True, single_use=False, ): u"""Executes XPath query on the ``lxml`` object and returns a correct object. :param str path: XPath string e.g., 'cars'/'car' :param str/dict namespaces: e.g., 'exslt', 're' or ``{'re': "http://exslt.org/regular-expressions"}`` :param bool regexp: if ``True`` and no namespaces is provided, it will use ``exslt`` namespace :param bool smart_strings: :param bool single_use: faster method for using only once. Does not create ``XPathEvaluator`` instance. >>> root = mappet.Mappet("<root><a>aB</a><b>aBc</b></root>") >>> root.XPath( "//*[re:test(., '^abc$', 'i')]", namespaces='exslt', regexp=True, ) """ if ( namespaces in ['exslt', 're'] or (regexp and not namespaces) ): namespaces = {'re': "http://exslt.org/regular-expressions"} if single_use: node = self._xml.xpath(path) else: xpe = self.xpath_evaluator( namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) node = xpe(path) if len(node) == 1: node = node[0] if len(node): return self.__class__(node) else: return Literal(node) return node def xpath_evaluator(self, namespaces=None, regexp=False, smart_strings=True): u"""Creates an XPathEvaluator instance for an ElementTree or an Element. :returns: ``XPathEvaluator`` instance """ return etree.XPathEvaluator( self._xml, namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) def keys(self): """Returns a set of node's keys.""" return set(self._get_aliases().keys())
stxnext/mappet
mappet/mappet.py
Mappet.assign_literal
python
def assign_literal(element, value): u"""Assigns a literal. If a given node doesn't exist, it will be created. :param etree.Element element: element to which we assign. :param value: the value to assign """ # Searches for a conversion method specific to the type of value. helper = helpers.CAST_DICT.get(type(value), str) # Removes all children and attributes. element.clear() element.text = helper(value)
u"""Assigns a literal. If a given node doesn't exist, it will be created. :param etree.Element element: element to which we assign. :param value: the value to assign
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/mappet.py#L591-L604
null
class Mappet(Node): u"""A node that may have children.""" _aliases = None u"""Dictionary with node aliases. The keys are normalized tagnames, values are the original tagnames. _aliases = { 'car_model_desc': 'car-model-desc', 'car': 'Car', } """ def __init__(self, xml): u"""Creates the mappet object from either lxml object, a string or a dict. If you pass a dict without root element, one will be created for you with 'root' as tag name. >>> Mappet({'a': {'#text': 'list_elem_1', '@attr1': 'val1'}}).to_str() '<a attr1="val1">list_elem_1</a>' >>> Mappet({'#text': 'list_elem_1', '@attr1': 'val1'}).to_str() '<root attr1="val1">list_elem_1</root>' """ if etree.iselement(xml): self._xml = xml elif isinstance(xml, basestring): self._xml = etree.fromstring(xml) elif isinstance(xml, dict): if len(xml) == 1: root_name = xml.keys()[0] body = xml[root_name] else: root_name = 'root' body = xml self._xml = helpers.dict_to_etree(body, etree.Element(root_name)) else: raise AttributeError('Specified data cannot be used to construct a Mappet object.') def __nonzero__(self): u"""Checks if this node has children, otherwise returns False.""" return self.has_children() def __len__(self): u"""Returns the children count.""" return len(self._xml) def __dir__(self): u"""Returns a list of children and available helper methods.""" return sorted(self.keys() | {m for m in dir(self.__class__) if m.startswith('to_')}) def __deepcopy__(self, memodict): u"""Performs a deepcopy on the underlying XML tree.""" return self.__class__(deepcopy(self._xml)) def __getattr__(self, name): u"""Attribute access. Returns a list o children, if there is more than 1. Returns a child, if there is exactly 1. """ children = self.children(name) if len(children) > 1: return children elif len(children) == 1: return children[0] def __setattr__(self, name, value): u"""Node attribute assignment. Calls ``set`` in the end. """ # Only elements that aren't a part of class definition are overwritten. if name not in dir(self.__class__): return self.set(name, value) return super(Mappet, self).__setattr__(name, value) def __delattr__(self, key): u"""Node removal.""" # Searches among aliases, if none is found returns the original key. tag = self._get_aliases().get(key, key) self.__delitem__(tag) def __getitem__(self, key): u"""Dictionary access.""" # Checks if the call isn't to an attribute. if isinstance(key, basestring) and not key.startswith('@'): children = self.children(key) if len(children) == 1: children = children[0] # Return the value if it's a leaf. if isinstance(children, Literal): return children.get() return children return super(Mappet, self).__getitem__(key) def __delitem__(self, key): u"""Removes all children with a given key.""" # Checks if name is not a part of class definition. if key not in dir(self.__class__): for child in self._xml.iterchildren(tag=key): self._xml.remove(child) def __eq__(self, other): u"""Compares mappet objects. Two mappet objects are deemed equal if the lxmls object they represent are equal. """ return etree.tostring(self._xml) == etree.tostring(other._xml) def __contains__(self, path): u"""Check if object contains given path.""" elem = self.sget(path) return not (elem is None or elem is NONE_NODE) def __getstate__(self): u"""Converts the lxml to string for Pickling.""" return { '_xml': etree.tostring(self._xml, pretty_print=False) } def __setstate__(self, dict_): u"""Restores a Pickled mappet object.""" self._xml = etree.fromstring(dict_['_xml']) def __iter__(self): u"""Returns children as an iterator.""" return self.iter_children() def to_str(self, pretty_print=False, encoding=None, **kw): u"""Converts a node with all of it's children to a string. Remaining arguments are passed to etree.tostring as is. kwarg without_comments: bool because it works only in C14N flags: 'pretty print' and 'encoding' are ignored. :param bool pretty_print: whether to format the output :param str encoding: which encoding to use (ASCII by default) :rtype: str :returns: node's representation as a string """ if kw.get('without_comments') and not kw.get('method'): kw.pop('without_comments') kw['method'] = 'c14n' kw['with_comments'] = False return etree.tostring( self._xml, pretty_print=pretty_print, encoding=encoding, **kw ) def has_children(self): u"""Returns true if a node has children.""" return bool(len(self)) def iter_children(self, key=None): u"""Iterates over children. :param key: A key for filtering children by tagname. """ tag = None if key: tag = self._get_aliases().get(key) if not tag: raise KeyError(key) for child in self._xml.iterchildren(tag=tag): if len(child): yield self.__class__(child) else: yield Literal(child) def children(self, key=None): u"""Returns node's children. :param key: A key for filtering children by tagname. """ return list(self.iter_children(key)) def update(self, **kwargs): u"""Updating or creation of new simple nodes. Each dict key is used as a tagname and value as text. """ for key, value in kwargs.items(): helper = helpers.CAST_DICT.get(type(value), str) tag = self._get_aliases().get(key, key) elements = list(self._xml.iterchildren(tag=tag)) if elements: for element in elements: element.text = helper(value) else: element = etree.Element(key) element.text = helper(value) self._xml.append(element) self._aliases = None def sget(self, path, default=NONE_NODE): u"""Enables access to nodes if one or more of them don't exist. Example: >>> m = Mappet('<root><tag attr1="attr text">text value</tag></root>') >>> m.sget('tag') text value >>> m.sget('tag.@attr1') 'attr text' >>> m.sget('tag.#text') 'text value' >>> m.sget('reply.vms_model_cars.car.0.params.doors') NONE_NODE Accessing nonexistent path returns None-like object with mocked converting functions which returns None: >>> m.sget('reply.fake_node').to_dict() is None True """ attrs = str(path).split(".") text_or_attr = None last_attr = attrs[-1] # Case of getting text or attribute if last_attr == '#text' or last_attr.startswith('@'): # #text => text, @attr => attr text_or_attr = last_attr[1:] attrs = attrs[:-1] # When getting #text and @attr we want default value to be None. if default is NONE_NODE: default = None my_object = self for attr in attrs: try: if isinstance(my_object, (list, tuple)) and re.match('^\-?\d+$', attr): my_object_next = my_object[int(attr)] else: my_object_next = getattr(my_object, attr) my_object = my_object_next except (AttributeError, KeyError, IndexError): return default # Return #text or @attr if text_or_attr: try: return my_object.getattr(text_or_attr) except AttributeError: # myObject can be a list. return None else: return my_object def create(self, tag, value): u"""Creates a node, if it doesn't exist yet. Unlike attribute access, this allows to pass a node's name with hyphens. Those hyphens will be normalized automatically. In case the required element already exists, raises an exception. Updating/overwriting should be done using `update``. """ child_tags = {child.tag for child in self._xml} if tag in child_tags: raise KeyError('Node {} already exists in XML tree.'.format(tag)) self.set(tag, value) def set(self, name, value): u"""Assigns a new XML structure to the node. A literal value, dict or list can be passed in. Works for all nested levels. Dictionary: >>> m = Mappet('<root/>') >>> m.head = {'a': 'A', 'b': {'#text': 'B', '@attr': 'val'}} >>> m.head.to_str() '<head><a>A</a><b attr="val">B</b></head>' List: >>> m.head = [{'a': i} for i in 'ABC'] >>> m.head.to_str() '<head><a>A</a><a>B</a><a>C</a></head>' Literals: >>> m.head.leaf = 'A' >>> m.head.leaf.get() 'A' """ try: # Searches for a node to assign to. element = next(self._xml.iterchildren(tag=name)) except StopIteration: # There is no such node in the XML tree. We create a new one # with current root as parent (self._xml). element = etree.SubElement(self._xml, name) if isinstance(value, dict): self.assign_dict(element, value) elif isinstance(value, (list, tuple, set)): self.assign_sequence_or_set(element, value) else: # Literal value. self.assign_literal(element, value) # Clear the aliases. self._aliases = None def assign_dict(self, node, xml_dict): """Assigns a Python dict to a ``lxml`` node. :param node: A node to assign the dict to. :param xml_dict: The dict with attributes/children to use. """ new_node = etree.Element(node.tag) # Replaces the previous node with the new one self._xml.replace(node, new_node) # Copies #text and @attrs from the xml_dict helpers.dict_to_etree(xml_dict, new_node) @staticmethod def assign_sequence_or_set(element, value): element.clear() for item in value: temp_element = etree.Element('temp') helpers.dict_to_etree(item, temp_element) for child in temp_element.iterchildren(): element.append(child) del temp_element @staticmethod def to_dict(self, **kw): u"""Converts the lxml object to a dict. possible kwargs: without_comments: bool """ _, value = helpers.etree_to_dict(self._xml, **kw).popitem() return value def _get_aliases(self): u"""Creates a dict with aliases. The key is a normalized tagname, value the original tagname. """ if self._aliases is None: self._aliases = {} if self._xml is not None: for child in self._xml.iterchildren(): self._aliases[helpers.normalize_tag(child.tag)] = child.tag return self._aliases def xpath( self, path, namespaces=None, regexp=False, smart_strings=True, single_use=False, ): u"""Executes XPath query on the ``lxml`` object and returns a correct object. :param str path: XPath string e.g., 'cars'/'car' :param str/dict namespaces: e.g., 'exslt', 're' or ``{'re': "http://exslt.org/regular-expressions"}`` :param bool regexp: if ``True`` and no namespaces is provided, it will use ``exslt`` namespace :param bool smart_strings: :param bool single_use: faster method for using only once. Does not create ``XPathEvaluator`` instance. >>> root = mappet.Mappet("<root><a>aB</a><b>aBc</b></root>") >>> root.XPath( "//*[re:test(., '^abc$', 'i')]", namespaces='exslt', regexp=True, ) """ if ( namespaces in ['exslt', 're'] or (regexp and not namespaces) ): namespaces = {'re': "http://exslt.org/regular-expressions"} if single_use: node = self._xml.xpath(path) else: xpe = self.xpath_evaluator( namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) node = xpe(path) if len(node) == 1: node = node[0] if len(node): return self.__class__(node) else: return Literal(node) return node def xpath_evaluator(self, namespaces=None, regexp=False, smart_strings=True): u"""Creates an XPathEvaluator instance for an ElementTree or an Element. :returns: ``XPathEvaluator`` instance """ return etree.XPathEvaluator( self._xml, namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) def keys(self): """Returns a set of node's keys.""" return set(self._get_aliases().keys())
stxnext/mappet
mappet/mappet.py
Mappet.to_dict
python
def to_dict(self, **kw): u"""Converts the lxml object to a dict. possible kwargs: without_comments: bool """ _, value = helpers.etree_to_dict(self._xml, **kw).popitem() return value
u"""Converts the lxml object to a dict. possible kwargs: without_comments: bool
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/mappet.py#L606-L613
[ "def etree_to_dict(t, trim=True, **kw):\n u\"\"\"Converts an lxml.etree object to Python dict.\n\n >>> etree_to_dict(etree.Element('root'))\n {'root': None}\n\n :param etree.Element t: lxml tree to convert\n :returns d: a dict representing the lxml tree ``t``\n :rtype: dict\n \"\"\"\n d = {t.tag: {} if t.attrib else None}\n children = list(t)\n etree_to_dict_w_args = partial(etree_to_dict, trim=trim, **kw)\n\n if children:\n dd = defaultdict(list)\n d = {t.tag: {}}\n\n for dc in map(etree_to_dict_w_args, children):\n for k, v in dc.iteritems():\n # do not add Comment instance to the key\n if k is not etree.Comment:\n dd[k].append(v)\n\n d[t.tag] = {k: v[0] if len(v) == 1 else v for k, v in dd.iteritems()}\n\n if t.attrib:\n d[t.tag].update(('@' + k, v) for k, v in t.attrib.iteritems())\n if trim and t.text:\n t.text = t.text.strip()\n if t.text:\n if t.tag is etree.Comment and not kw.get('without_comments'):\n # adds a comments node\n d['#comments'] = t.text\n elif children or t.attrib:\n d[t.tag]['#text'] = t.text\n else:\n d[t.tag] = t.text\n return d\n" ]
class Mappet(Node): u"""A node that may have children.""" _aliases = None u"""Dictionary with node aliases. The keys are normalized tagnames, values are the original tagnames. _aliases = { 'car_model_desc': 'car-model-desc', 'car': 'Car', } """ def __init__(self, xml): u"""Creates the mappet object from either lxml object, a string or a dict. If you pass a dict without root element, one will be created for you with 'root' as tag name. >>> Mappet({'a': {'#text': 'list_elem_1', '@attr1': 'val1'}}).to_str() '<a attr1="val1">list_elem_1</a>' >>> Mappet({'#text': 'list_elem_1', '@attr1': 'val1'}).to_str() '<root attr1="val1">list_elem_1</root>' """ if etree.iselement(xml): self._xml = xml elif isinstance(xml, basestring): self._xml = etree.fromstring(xml) elif isinstance(xml, dict): if len(xml) == 1: root_name = xml.keys()[0] body = xml[root_name] else: root_name = 'root' body = xml self._xml = helpers.dict_to_etree(body, etree.Element(root_name)) else: raise AttributeError('Specified data cannot be used to construct a Mappet object.') def __nonzero__(self): u"""Checks if this node has children, otherwise returns False.""" return self.has_children() def __len__(self): u"""Returns the children count.""" return len(self._xml) def __dir__(self): u"""Returns a list of children and available helper methods.""" return sorted(self.keys() | {m for m in dir(self.__class__) if m.startswith('to_')}) def __deepcopy__(self, memodict): u"""Performs a deepcopy on the underlying XML tree.""" return self.__class__(deepcopy(self._xml)) def __getattr__(self, name): u"""Attribute access. Returns a list o children, if there is more than 1. Returns a child, if there is exactly 1. """ children = self.children(name) if len(children) > 1: return children elif len(children) == 1: return children[0] def __setattr__(self, name, value): u"""Node attribute assignment. Calls ``set`` in the end. """ # Only elements that aren't a part of class definition are overwritten. if name not in dir(self.__class__): return self.set(name, value) return super(Mappet, self).__setattr__(name, value) def __delattr__(self, key): u"""Node removal.""" # Searches among aliases, if none is found returns the original key. tag = self._get_aliases().get(key, key) self.__delitem__(tag) def __getitem__(self, key): u"""Dictionary access.""" # Checks if the call isn't to an attribute. if isinstance(key, basestring) and not key.startswith('@'): children = self.children(key) if len(children) == 1: children = children[0] # Return the value if it's a leaf. if isinstance(children, Literal): return children.get() return children return super(Mappet, self).__getitem__(key) def __delitem__(self, key): u"""Removes all children with a given key.""" # Checks if name is not a part of class definition. if key not in dir(self.__class__): for child in self._xml.iterchildren(tag=key): self._xml.remove(child) def __eq__(self, other): u"""Compares mappet objects. Two mappet objects are deemed equal if the lxmls object they represent are equal. """ return etree.tostring(self._xml) == etree.tostring(other._xml) def __contains__(self, path): u"""Check if object contains given path.""" elem = self.sget(path) return not (elem is None or elem is NONE_NODE) def __getstate__(self): u"""Converts the lxml to string for Pickling.""" return { '_xml': etree.tostring(self._xml, pretty_print=False) } def __setstate__(self, dict_): u"""Restores a Pickled mappet object.""" self._xml = etree.fromstring(dict_['_xml']) def __iter__(self): u"""Returns children as an iterator.""" return self.iter_children() def to_str(self, pretty_print=False, encoding=None, **kw): u"""Converts a node with all of it's children to a string. Remaining arguments are passed to etree.tostring as is. kwarg without_comments: bool because it works only in C14N flags: 'pretty print' and 'encoding' are ignored. :param bool pretty_print: whether to format the output :param str encoding: which encoding to use (ASCII by default) :rtype: str :returns: node's representation as a string """ if kw.get('without_comments') and not kw.get('method'): kw.pop('without_comments') kw['method'] = 'c14n' kw['with_comments'] = False return etree.tostring( self._xml, pretty_print=pretty_print, encoding=encoding, **kw ) def has_children(self): u"""Returns true if a node has children.""" return bool(len(self)) def iter_children(self, key=None): u"""Iterates over children. :param key: A key for filtering children by tagname. """ tag = None if key: tag = self._get_aliases().get(key) if not tag: raise KeyError(key) for child in self._xml.iterchildren(tag=tag): if len(child): yield self.__class__(child) else: yield Literal(child) def children(self, key=None): u"""Returns node's children. :param key: A key for filtering children by tagname. """ return list(self.iter_children(key)) def update(self, **kwargs): u"""Updating or creation of new simple nodes. Each dict key is used as a tagname and value as text. """ for key, value in kwargs.items(): helper = helpers.CAST_DICT.get(type(value), str) tag = self._get_aliases().get(key, key) elements = list(self._xml.iterchildren(tag=tag)) if elements: for element in elements: element.text = helper(value) else: element = etree.Element(key) element.text = helper(value) self._xml.append(element) self._aliases = None def sget(self, path, default=NONE_NODE): u"""Enables access to nodes if one or more of them don't exist. Example: >>> m = Mappet('<root><tag attr1="attr text">text value</tag></root>') >>> m.sget('tag') text value >>> m.sget('tag.@attr1') 'attr text' >>> m.sget('tag.#text') 'text value' >>> m.sget('reply.vms_model_cars.car.0.params.doors') NONE_NODE Accessing nonexistent path returns None-like object with mocked converting functions which returns None: >>> m.sget('reply.fake_node').to_dict() is None True """ attrs = str(path).split(".") text_or_attr = None last_attr = attrs[-1] # Case of getting text or attribute if last_attr == '#text' or last_attr.startswith('@'): # #text => text, @attr => attr text_or_attr = last_attr[1:] attrs = attrs[:-1] # When getting #text and @attr we want default value to be None. if default is NONE_NODE: default = None my_object = self for attr in attrs: try: if isinstance(my_object, (list, tuple)) and re.match('^\-?\d+$', attr): my_object_next = my_object[int(attr)] else: my_object_next = getattr(my_object, attr) my_object = my_object_next except (AttributeError, KeyError, IndexError): return default # Return #text or @attr if text_or_attr: try: return my_object.getattr(text_or_attr) except AttributeError: # myObject can be a list. return None else: return my_object def create(self, tag, value): u"""Creates a node, if it doesn't exist yet. Unlike attribute access, this allows to pass a node's name with hyphens. Those hyphens will be normalized automatically. In case the required element already exists, raises an exception. Updating/overwriting should be done using `update``. """ child_tags = {child.tag for child in self._xml} if tag in child_tags: raise KeyError('Node {} already exists in XML tree.'.format(tag)) self.set(tag, value) def set(self, name, value): u"""Assigns a new XML structure to the node. A literal value, dict or list can be passed in. Works for all nested levels. Dictionary: >>> m = Mappet('<root/>') >>> m.head = {'a': 'A', 'b': {'#text': 'B', '@attr': 'val'}} >>> m.head.to_str() '<head><a>A</a><b attr="val">B</b></head>' List: >>> m.head = [{'a': i} for i in 'ABC'] >>> m.head.to_str() '<head><a>A</a><a>B</a><a>C</a></head>' Literals: >>> m.head.leaf = 'A' >>> m.head.leaf.get() 'A' """ try: # Searches for a node to assign to. element = next(self._xml.iterchildren(tag=name)) except StopIteration: # There is no such node in the XML tree. We create a new one # with current root as parent (self._xml). element = etree.SubElement(self._xml, name) if isinstance(value, dict): self.assign_dict(element, value) elif isinstance(value, (list, tuple, set)): self.assign_sequence_or_set(element, value) else: # Literal value. self.assign_literal(element, value) # Clear the aliases. self._aliases = None def assign_dict(self, node, xml_dict): """Assigns a Python dict to a ``lxml`` node. :param node: A node to assign the dict to. :param xml_dict: The dict with attributes/children to use. """ new_node = etree.Element(node.tag) # Replaces the previous node with the new one self._xml.replace(node, new_node) # Copies #text and @attrs from the xml_dict helpers.dict_to_etree(xml_dict, new_node) @staticmethod def assign_sequence_or_set(element, value): element.clear() for item in value: temp_element = etree.Element('temp') helpers.dict_to_etree(item, temp_element) for child in temp_element.iterchildren(): element.append(child) del temp_element @staticmethod def assign_literal(element, value): u"""Assigns a literal. If a given node doesn't exist, it will be created. :param etree.Element element: element to which we assign. :param value: the value to assign """ # Searches for a conversion method specific to the type of value. helper = helpers.CAST_DICT.get(type(value), str) # Removes all children and attributes. element.clear() element.text = helper(value) def _get_aliases(self): u"""Creates a dict with aliases. The key is a normalized tagname, value the original tagname. """ if self._aliases is None: self._aliases = {} if self._xml is not None: for child in self._xml.iterchildren(): self._aliases[helpers.normalize_tag(child.tag)] = child.tag return self._aliases def xpath( self, path, namespaces=None, regexp=False, smart_strings=True, single_use=False, ): u"""Executes XPath query on the ``lxml`` object and returns a correct object. :param str path: XPath string e.g., 'cars'/'car' :param str/dict namespaces: e.g., 'exslt', 're' or ``{'re': "http://exslt.org/regular-expressions"}`` :param bool regexp: if ``True`` and no namespaces is provided, it will use ``exslt`` namespace :param bool smart_strings: :param bool single_use: faster method for using only once. Does not create ``XPathEvaluator`` instance. >>> root = mappet.Mappet("<root><a>aB</a><b>aBc</b></root>") >>> root.XPath( "//*[re:test(., '^abc$', 'i')]", namespaces='exslt', regexp=True, ) """ if ( namespaces in ['exslt', 're'] or (regexp and not namespaces) ): namespaces = {'re': "http://exslt.org/regular-expressions"} if single_use: node = self._xml.xpath(path) else: xpe = self.xpath_evaluator( namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) node = xpe(path) if len(node) == 1: node = node[0] if len(node): return self.__class__(node) else: return Literal(node) return node def xpath_evaluator(self, namespaces=None, regexp=False, smart_strings=True): u"""Creates an XPathEvaluator instance for an ElementTree or an Element. :returns: ``XPathEvaluator`` instance """ return etree.XPathEvaluator( self._xml, namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) def keys(self): """Returns a set of node's keys.""" return set(self._get_aliases().keys())
stxnext/mappet
mappet/mappet.py
Mappet._get_aliases
python
def _get_aliases(self): u"""Creates a dict with aliases. The key is a normalized tagname, value the original tagname. """ if self._aliases is None: self._aliases = {} if self._xml is not None: for child in self._xml.iterchildren(): self._aliases[helpers.normalize_tag(child.tag)] = child.tag return self._aliases
u"""Creates a dict with aliases. The key is a normalized tagname, value the original tagname.
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/mappet.py#L615-L627
[ "def normalize_tag(tag):\n u\"\"\"Normalizes tag name.\n\n :param str tag: tag name to normalize\n :rtype: str\n :returns: normalized tag name\n\n >>> normalize_tag('tag-NaMe')\n 'tag_name'\n \"\"\"\n return tag.lower().replace('-', '_')\n" ]
class Mappet(Node): u"""A node that may have children.""" _aliases = None u"""Dictionary with node aliases. The keys are normalized tagnames, values are the original tagnames. _aliases = { 'car_model_desc': 'car-model-desc', 'car': 'Car', } """ def __init__(self, xml): u"""Creates the mappet object from either lxml object, a string or a dict. If you pass a dict without root element, one will be created for you with 'root' as tag name. >>> Mappet({'a': {'#text': 'list_elem_1', '@attr1': 'val1'}}).to_str() '<a attr1="val1">list_elem_1</a>' >>> Mappet({'#text': 'list_elem_1', '@attr1': 'val1'}).to_str() '<root attr1="val1">list_elem_1</root>' """ if etree.iselement(xml): self._xml = xml elif isinstance(xml, basestring): self._xml = etree.fromstring(xml) elif isinstance(xml, dict): if len(xml) == 1: root_name = xml.keys()[0] body = xml[root_name] else: root_name = 'root' body = xml self._xml = helpers.dict_to_etree(body, etree.Element(root_name)) else: raise AttributeError('Specified data cannot be used to construct a Mappet object.') def __nonzero__(self): u"""Checks if this node has children, otherwise returns False.""" return self.has_children() def __len__(self): u"""Returns the children count.""" return len(self._xml) def __dir__(self): u"""Returns a list of children and available helper methods.""" return sorted(self.keys() | {m for m in dir(self.__class__) if m.startswith('to_')}) def __deepcopy__(self, memodict): u"""Performs a deepcopy on the underlying XML tree.""" return self.__class__(deepcopy(self._xml)) def __getattr__(self, name): u"""Attribute access. Returns a list o children, if there is more than 1. Returns a child, if there is exactly 1. """ children = self.children(name) if len(children) > 1: return children elif len(children) == 1: return children[0] def __setattr__(self, name, value): u"""Node attribute assignment. Calls ``set`` in the end. """ # Only elements that aren't a part of class definition are overwritten. if name not in dir(self.__class__): return self.set(name, value) return super(Mappet, self).__setattr__(name, value) def __delattr__(self, key): u"""Node removal.""" # Searches among aliases, if none is found returns the original key. tag = self._get_aliases().get(key, key) self.__delitem__(tag) def __getitem__(self, key): u"""Dictionary access.""" # Checks if the call isn't to an attribute. if isinstance(key, basestring) and not key.startswith('@'): children = self.children(key) if len(children) == 1: children = children[0] # Return the value if it's a leaf. if isinstance(children, Literal): return children.get() return children return super(Mappet, self).__getitem__(key) def __delitem__(self, key): u"""Removes all children with a given key.""" # Checks if name is not a part of class definition. if key not in dir(self.__class__): for child in self._xml.iterchildren(tag=key): self._xml.remove(child) def __eq__(self, other): u"""Compares mappet objects. Two mappet objects are deemed equal if the lxmls object they represent are equal. """ return etree.tostring(self._xml) == etree.tostring(other._xml) def __contains__(self, path): u"""Check if object contains given path.""" elem = self.sget(path) return not (elem is None or elem is NONE_NODE) def __getstate__(self): u"""Converts the lxml to string for Pickling.""" return { '_xml': etree.tostring(self._xml, pretty_print=False) } def __setstate__(self, dict_): u"""Restores a Pickled mappet object.""" self._xml = etree.fromstring(dict_['_xml']) def __iter__(self): u"""Returns children as an iterator.""" return self.iter_children() def to_str(self, pretty_print=False, encoding=None, **kw): u"""Converts a node with all of it's children to a string. Remaining arguments are passed to etree.tostring as is. kwarg without_comments: bool because it works only in C14N flags: 'pretty print' and 'encoding' are ignored. :param bool pretty_print: whether to format the output :param str encoding: which encoding to use (ASCII by default) :rtype: str :returns: node's representation as a string """ if kw.get('without_comments') and not kw.get('method'): kw.pop('without_comments') kw['method'] = 'c14n' kw['with_comments'] = False return etree.tostring( self._xml, pretty_print=pretty_print, encoding=encoding, **kw ) def has_children(self): u"""Returns true if a node has children.""" return bool(len(self)) def iter_children(self, key=None): u"""Iterates over children. :param key: A key for filtering children by tagname. """ tag = None if key: tag = self._get_aliases().get(key) if not tag: raise KeyError(key) for child in self._xml.iterchildren(tag=tag): if len(child): yield self.__class__(child) else: yield Literal(child) def children(self, key=None): u"""Returns node's children. :param key: A key for filtering children by tagname. """ return list(self.iter_children(key)) def update(self, **kwargs): u"""Updating or creation of new simple nodes. Each dict key is used as a tagname and value as text. """ for key, value in kwargs.items(): helper = helpers.CAST_DICT.get(type(value), str) tag = self._get_aliases().get(key, key) elements = list(self._xml.iterchildren(tag=tag)) if elements: for element in elements: element.text = helper(value) else: element = etree.Element(key) element.text = helper(value) self._xml.append(element) self._aliases = None def sget(self, path, default=NONE_NODE): u"""Enables access to nodes if one or more of them don't exist. Example: >>> m = Mappet('<root><tag attr1="attr text">text value</tag></root>') >>> m.sget('tag') text value >>> m.sget('tag.@attr1') 'attr text' >>> m.sget('tag.#text') 'text value' >>> m.sget('reply.vms_model_cars.car.0.params.doors') NONE_NODE Accessing nonexistent path returns None-like object with mocked converting functions which returns None: >>> m.sget('reply.fake_node').to_dict() is None True """ attrs = str(path).split(".") text_or_attr = None last_attr = attrs[-1] # Case of getting text or attribute if last_attr == '#text' or last_attr.startswith('@'): # #text => text, @attr => attr text_or_attr = last_attr[1:] attrs = attrs[:-1] # When getting #text and @attr we want default value to be None. if default is NONE_NODE: default = None my_object = self for attr in attrs: try: if isinstance(my_object, (list, tuple)) and re.match('^\-?\d+$', attr): my_object_next = my_object[int(attr)] else: my_object_next = getattr(my_object, attr) my_object = my_object_next except (AttributeError, KeyError, IndexError): return default # Return #text or @attr if text_or_attr: try: return my_object.getattr(text_or_attr) except AttributeError: # myObject can be a list. return None else: return my_object def create(self, tag, value): u"""Creates a node, if it doesn't exist yet. Unlike attribute access, this allows to pass a node's name with hyphens. Those hyphens will be normalized automatically. In case the required element already exists, raises an exception. Updating/overwriting should be done using `update``. """ child_tags = {child.tag for child in self._xml} if tag in child_tags: raise KeyError('Node {} already exists in XML tree.'.format(tag)) self.set(tag, value) def set(self, name, value): u"""Assigns a new XML structure to the node. A literal value, dict or list can be passed in. Works for all nested levels. Dictionary: >>> m = Mappet('<root/>') >>> m.head = {'a': 'A', 'b': {'#text': 'B', '@attr': 'val'}} >>> m.head.to_str() '<head><a>A</a><b attr="val">B</b></head>' List: >>> m.head = [{'a': i} for i in 'ABC'] >>> m.head.to_str() '<head><a>A</a><a>B</a><a>C</a></head>' Literals: >>> m.head.leaf = 'A' >>> m.head.leaf.get() 'A' """ try: # Searches for a node to assign to. element = next(self._xml.iterchildren(tag=name)) except StopIteration: # There is no such node in the XML tree. We create a new one # with current root as parent (self._xml). element = etree.SubElement(self._xml, name) if isinstance(value, dict): self.assign_dict(element, value) elif isinstance(value, (list, tuple, set)): self.assign_sequence_or_set(element, value) else: # Literal value. self.assign_literal(element, value) # Clear the aliases. self._aliases = None def assign_dict(self, node, xml_dict): """Assigns a Python dict to a ``lxml`` node. :param node: A node to assign the dict to. :param xml_dict: The dict with attributes/children to use. """ new_node = etree.Element(node.tag) # Replaces the previous node with the new one self._xml.replace(node, new_node) # Copies #text and @attrs from the xml_dict helpers.dict_to_etree(xml_dict, new_node) @staticmethod def assign_sequence_or_set(element, value): element.clear() for item in value: temp_element = etree.Element('temp') helpers.dict_to_etree(item, temp_element) for child in temp_element.iterchildren(): element.append(child) del temp_element @staticmethod def assign_literal(element, value): u"""Assigns a literal. If a given node doesn't exist, it will be created. :param etree.Element element: element to which we assign. :param value: the value to assign """ # Searches for a conversion method specific to the type of value. helper = helpers.CAST_DICT.get(type(value), str) # Removes all children and attributes. element.clear() element.text = helper(value) def to_dict(self, **kw): u"""Converts the lxml object to a dict. possible kwargs: without_comments: bool """ _, value = helpers.etree_to_dict(self._xml, **kw).popitem() return value def xpath( self, path, namespaces=None, regexp=False, smart_strings=True, single_use=False, ): u"""Executes XPath query on the ``lxml`` object and returns a correct object. :param str path: XPath string e.g., 'cars'/'car' :param str/dict namespaces: e.g., 'exslt', 're' or ``{'re': "http://exslt.org/regular-expressions"}`` :param bool regexp: if ``True`` and no namespaces is provided, it will use ``exslt`` namespace :param bool smart_strings: :param bool single_use: faster method for using only once. Does not create ``XPathEvaluator`` instance. >>> root = mappet.Mappet("<root><a>aB</a><b>aBc</b></root>") >>> root.XPath( "//*[re:test(., '^abc$', 'i')]", namespaces='exslt', regexp=True, ) """ if ( namespaces in ['exslt', 're'] or (regexp and not namespaces) ): namespaces = {'re': "http://exslt.org/regular-expressions"} if single_use: node = self._xml.xpath(path) else: xpe = self.xpath_evaluator( namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) node = xpe(path) if len(node) == 1: node = node[0] if len(node): return self.__class__(node) else: return Literal(node) return node def xpath_evaluator(self, namespaces=None, regexp=False, smart_strings=True): u"""Creates an XPathEvaluator instance for an ElementTree or an Element. :returns: ``XPathEvaluator`` instance """ return etree.XPathEvaluator( self._xml, namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) def keys(self): """Returns a set of node's keys.""" return set(self._get_aliases().keys())
stxnext/mappet
mappet/mappet.py
Mappet.xpath
python
def xpath( self, path, namespaces=None, regexp=False, smart_strings=True, single_use=False, ): u"""Executes XPath query on the ``lxml`` object and returns a correct object. :param str path: XPath string e.g., 'cars'/'car' :param str/dict namespaces: e.g., 'exslt', 're' or ``{'re': "http://exslt.org/regular-expressions"}`` :param bool regexp: if ``True`` and no namespaces is provided, it will use ``exslt`` namespace :param bool smart_strings: :param bool single_use: faster method for using only once. Does not create ``XPathEvaluator`` instance. >>> root = mappet.Mappet("<root><a>aB</a><b>aBc</b></root>") >>> root.XPath( "//*[re:test(., '^abc$', 'i')]", namespaces='exslt', regexp=True, ) """ if ( namespaces in ['exslt', 're'] or (regexp and not namespaces) ): namespaces = {'re': "http://exslt.org/regular-expressions"} if single_use: node = self._xml.xpath(path) else: xpe = self.xpath_evaluator( namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) node = xpe(path) if len(node) == 1: node = node[0] if len(node): return self.__class__(node) else: return Literal(node) return node
u"""Executes XPath query on the ``lxml`` object and returns a correct object. :param str path: XPath string e.g., 'cars'/'car' :param str/dict namespaces: e.g., 'exslt', 're' or ``{'re': "http://exslt.org/regular-expressions"}`` :param bool regexp: if ``True`` and no namespaces is provided, it will use ``exslt`` namespace :param bool smart_strings: :param bool single_use: faster method for using only once. Does not create ``XPathEvaluator`` instance. >>> root = mappet.Mappet("<root><a>aB</a><b>aBc</b></root>") >>> root.XPath( "//*[re:test(., '^abc$', 'i')]", namespaces='exslt', regexp=True, )
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/mappet.py#L629-L677
[ "def xpath_evaluator(self, namespaces=None, regexp=False, smart_strings=True):\n u\"\"\"Creates an XPathEvaluator instance for an ElementTree or an Element.\n\n :returns: ``XPathEvaluator`` instance\n \"\"\"\n return etree.XPathEvaluator(\n self._xml,\n namespaces=namespaces,\n regexp=regexp,\n smart_strings=smart_strings\n )\n" ]
class Mappet(Node): u"""A node that may have children.""" _aliases = None u"""Dictionary with node aliases. The keys are normalized tagnames, values are the original tagnames. _aliases = { 'car_model_desc': 'car-model-desc', 'car': 'Car', } """ def __init__(self, xml): u"""Creates the mappet object from either lxml object, a string or a dict. If you pass a dict without root element, one will be created for you with 'root' as tag name. >>> Mappet({'a': {'#text': 'list_elem_1', '@attr1': 'val1'}}).to_str() '<a attr1="val1">list_elem_1</a>' >>> Mappet({'#text': 'list_elem_1', '@attr1': 'val1'}).to_str() '<root attr1="val1">list_elem_1</root>' """ if etree.iselement(xml): self._xml = xml elif isinstance(xml, basestring): self._xml = etree.fromstring(xml) elif isinstance(xml, dict): if len(xml) == 1: root_name = xml.keys()[0] body = xml[root_name] else: root_name = 'root' body = xml self._xml = helpers.dict_to_etree(body, etree.Element(root_name)) else: raise AttributeError('Specified data cannot be used to construct a Mappet object.') def __nonzero__(self): u"""Checks if this node has children, otherwise returns False.""" return self.has_children() def __len__(self): u"""Returns the children count.""" return len(self._xml) def __dir__(self): u"""Returns a list of children and available helper methods.""" return sorted(self.keys() | {m for m in dir(self.__class__) if m.startswith('to_')}) def __deepcopy__(self, memodict): u"""Performs a deepcopy on the underlying XML tree.""" return self.__class__(deepcopy(self._xml)) def __getattr__(self, name): u"""Attribute access. Returns a list o children, if there is more than 1. Returns a child, if there is exactly 1. """ children = self.children(name) if len(children) > 1: return children elif len(children) == 1: return children[0] def __setattr__(self, name, value): u"""Node attribute assignment. Calls ``set`` in the end. """ # Only elements that aren't a part of class definition are overwritten. if name not in dir(self.__class__): return self.set(name, value) return super(Mappet, self).__setattr__(name, value) def __delattr__(self, key): u"""Node removal.""" # Searches among aliases, if none is found returns the original key. tag = self._get_aliases().get(key, key) self.__delitem__(tag) def __getitem__(self, key): u"""Dictionary access.""" # Checks if the call isn't to an attribute. if isinstance(key, basestring) and not key.startswith('@'): children = self.children(key) if len(children) == 1: children = children[0] # Return the value if it's a leaf. if isinstance(children, Literal): return children.get() return children return super(Mappet, self).__getitem__(key) def __delitem__(self, key): u"""Removes all children with a given key.""" # Checks if name is not a part of class definition. if key not in dir(self.__class__): for child in self._xml.iterchildren(tag=key): self._xml.remove(child) def __eq__(self, other): u"""Compares mappet objects. Two mappet objects are deemed equal if the lxmls object they represent are equal. """ return etree.tostring(self._xml) == etree.tostring(other._xml) def __contains__(self, path): u"""Check if object contains given path.""" elem = self.sget(path) return not (elem is None or elem is NONE_NODE) def __getstate__(self): u"""Converts the lxml to string for Pickling.""" return { '_xml': etree.tostring(self._xml, pretty_print=False) } def __setstate__(self, dict_): u"""Restores a Pickled mappet object.""" self._xml = etree.fromstring(dict_['_xml']) def __iter__(self): u"""Returns children as an iterator.""" return self.iter_children() def to_str(self, pretty_print=False, encoding=None, **kw): u"""Converts a node with all of it's children to a string. Remaining arguments are passed to etree.tostring as is. kwarg without_comments: bool because it works only in C14N flags: 'pretty print' and 'encoding' are ignored. :param bool pretty_print: whether to format the output :param str encoding: which encoding to use (ASCII by default) :rtype: str :returns: node's representation as a string """ if kw.get('without_comments') and not kw.get('method'): kw.pop('without_comments') kw['method'] = 'c14n' kw['with_comments'] = False return etree.tostring( self._xml, pretty_print=pretty_print, encoding=encoding, **kw ) def has_children(self): u"""Returns true if a node has children.""" return bool(len(self)) def iter_children(self, key=None): u"""Iterates over children. :param key: A key for filtering children by tagname. """ tag = None if key: tag = self._get_aliases().get(key) if not tag: raise KeyError(key) for child in self._xml.iterchildren(tag=tag): if len(child): yield self.__class__(child) else: yield Literal(child) def children(self, key=None): u"""Returns node's children. :param key: A key for filtering children by tagname. """ return list(self.iter_children(key)) def update(self, **kwargs): u"""Updating or creation of new simple nodes. Each dict key is used as a tagname and value as text. """ for key, value in kwargs.items(): helper = helpers.CAST_DICT.get(type(value), str) tag = self._get_aliases().get(key, key) elements = list(self._xml.iterchildren(tag=tag)) if elements: for element in elements: element.text = helper(value) else: element = etree.Element(key) element.text = helper(value) self._xml.append(element) self._aliases = None def sget(self, path, default=NONE_NODE): u"""Enables access to nodes if one or more of them don't exist. Example: >>> m = Mappet('<root><tag attr1="attr text">text value</tag></root>') >>> m.sget('tag') text value >>> m.sget('tag.@attr1') 'attr text' >>> m.sget('tag.#text') 'text value' >>> m.sget('reply.vms_model_cars.car.0.params.doors') NONE_NODE Accessing nonexistent path returns None-like object with mocked converting functions which returns None: >>> m.sget('reply.fake_node').to_dict() is None True """ attrs = str(path).split(".") text_or_attr = None last_attr = attrs[-1] # Case of getting text or attribute if last_attr == '#text' or last_attr.startswith('@'): # #text => text, @attr => attr text_or_attr = last_attr[1:] attrs = attrs[:-1] # When getting #text and @attr we want default value to be None. if default is NONE_NODE: default = None my_object = self for attr in attrs: try: if isinstance(my_object, (list, tuple)) and re.match('^\-?\d+$', attr): my_object_next = my_object[int(attr)] else: my_object_next = getattr(my_object, attr) my_object = my_object_next except (AttributeError, KeyError, IndexError): return default # Return #text or @attr if text_or_attr: try: return my_object.getattr(text_or_attr) except AttributeError: # myObject can be a list. return None else: return my_object def create(self, tag, value): u"""Creates a node, if it doesn't exist yet. Unlike attribute access, this allows to pass a node's name with hyphens. Those hyphens will be normalized automatically. In case the required element already exists, raises an exception. Updating/overwriting should be done using `update``. """ child_tags = {child.tag for child in self._xml} if tag in child_tags: raise KeyError('Node {} already exists in XML tree.'.format(tag)) self.set(tag, value) def set(self, name, value): u"""Assigns a new XML structure to the node. A literal value, dict or list can be passed in. Works for all nested levels. Dictionary: >>> m = Mappet('<root/>') >>> m.head = {'a': 'A', 'b': {'#text': 'B', '@attr': 'val'}} >>> m.head.to_str() '<head><a>A</a><b attr="val">B</b></head>' List: >>> m.head = [{'a': i} for i in 'ABC'] >>> m.head.to_str() '<head><a>A</a><a>B</a><a>C</a></head>' Literals: >>> m.head.leaf = 'A' >>> m.head.leaf.get() 'A' """ try: # Searches for a node to assign to. element = next(self._xml.iterchildren(tag=name)) except StopIteration: # There is no such node in the XML tree. We create a new one # with current root as parent (self._xml). element = etree.SubElement(self._xml, name) if isinstance(value, dict): self.assign_dict(element, value) elif isinstance(value, (list, tuple, set)): self.assign_sequence_or_set(element, value) else: # Literal value. self.assign_literal(element, value) # Clear the aliases. self._aliases = None def assign_dict(self, node, xml_dict): """Assigns a Python dict to a ``lxml`` node. :param node: A node to assign the dict to. :param xml_dict: The dict with attributes/children to use. """ new_node = etree.Element(node.tag) # Replaces the previous node with the new one self._xml.replace(node, new_node) # Copies #text and @attrs from the xml_dict helpers.dict_to_etree(xml_dict, new_node) @staticmethod def assign_sequence_or_set(element, value): element.clear() for item in value: temp_element = etree.Element('temp') helpers.dict_to_etree(item, temp_element) for child in temp_element.iterchildren(): element.append(child) del temp_element @staticmethod def assign_literal(element, value): u"""Assigns a literal. If a given node doesn't exist, it will be created. :param etree.Element element: element to which we assign. :param value: the value to assign """ # Searches for a conversion method specific to the type of value. helper = helpers.CAST_DICT.get(type(value), str) # Removes all children and attributes. element.clear() element.text = helper(value) def to_dict(self, **kw): u"""Converts the lxml object to a dict. possible kwargs: without_comments: bool """ _, value = helpers.etree_to_dict(self._xml, **kw).popitem() return value def _get_aliases(self): u"""Creates a dict with aliases. The key is a normalized tagname, value the original tagname. """ if self._aliases is None: self._aliases = {} if self._xml is not None: for child in self._xml.iterchildren(): self._aliases[helpers.normalize_tag(child.tag)] = child.tag return self._aliases def xpath_evaluator(self, namespaces=None, regexp=False, smart_strings=True): u"""Creates an XPathEvaluator instance for an ElementTree or an Element. :returns: ``XPathEvaluator`` instance """ return etree.XPathEvaluator( self._xml, namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) def keys(self): """Returns a set of node's keys.""" return set(self._get_aliases().keys())
stxnext/mappet
mappet/mappet.py
Mappet.xpath_evaluator
python
def xpath_evaluator(self, namespaces=None, regexp=False, smart_strings=True): u"""Creates an XPathEvaluator instance for an ElementTree or an Element. :returns: ``XPathEvaluator`` instance """ return etree.XPathEvaluator( self._xml, namespaces=namespaces, regexp=regexp, smart_strings=smart_strings )
u"""Creates an XPathEvaluator instance for an ElementTree or an Element. :returns: ``XPathEvaluator`` instance
train
https://github.com/stxnext/mappet/blob/ac7468ac28ed82e45065b1e348cf865c8f73f0db/mappet/mappet.py#L679-L689
null
class Mappet(Node): u"""A node that may have children.""" _aliases = None u"""Dictionary with node aliases. The keys are normalized tagnames, values are the original tagnames. _aliases = { 'car_model_desc': 'car-model-desc', 'car': 'Car', } """ def __init__(self, xml): u"""Creates the mappet object from either lxml object, a string or a dict. If you pass a dict without root element, one will be created for you with 'root' as tag name. >>> Mappet({'a': {'#text': 'list_elem_1', '@attr1': 'val1'}}).to_str() '<a attr1="val1">list_elem_1</a>' >>> Mappet({'#text': 'list_elem_1', '@attr1': 'val1'}).to_str() '<root attr1="val1">list_elem_1</root>' """ if etree.iselement(xml): self._xml = xml elif isinstance(xml, basestring): self._xml = etree.fromstring(xml) elif isinstance(xml, dict): if len(xml) == 1: root_name = xml.keys()[0] body = xml[root_name] else: root_name = 'root' body = xml self._xml = helpers.dict_to_etree(body, etree.Element(root_name)) else: raise AttributeError('Specified data cannot be used to construct a Mappet object.') def __nonzero__(self): u"""Checks if this node has children, otherwise returns False.""" return self.has_children() def __len__(self): u"""Returns the children count.""" return len(self._xml) def __dir__(self): u"""Returns a list of children and available helper methods.""" return sorted(self.keys() | {m for m in dir(self.__class__) if m.startswith('to_')}) def __deepcopy__(self, memodict): u"""Performs a deepcopy on the underlying XML tree.""" return self.__class__(deepcopy(self._xml)) def __getattr__(self, name): u"""Attribute access. Returns a list o children, if there is more than 1. Returns a child, if there is exactly 1. """ children = self.children(name) if len(children) > 1: return children elif len(children) == 1: return children[0] def __setattr__(self, name, value): u"""Node attribute assignment. Calls ``set`` in the end. """ # Only elements that aren't a part of class definition are overwritten. if name not in dir(self.__class__): return self.set(name, value) return super(Mappet, self).__setattr__(name, value) def __delattr__(self, key): u"""Node removal.""" # Searches among aliases, if none is found returns the original key. tag = self._get_aliases().get(key, key) self.__delitem__(tag) def __getitem__(self, key): u"""Dictionary access.""" # Checks if the call isn't to an attribute. if isinstance(key, basestring) and not key.startswith('@'): children = self.children(key) if len(children) == 1: children = children[0] # Return the value if it's a leaf. if isinstance(children, Literal): return children.get() return children return super(Mappet, self).__getitem__(key) def __delitem__(self, key): u"""Removes all children with a given key.""" # Checks if name is not a part of class definition. if key not in dir(self.__class__): for child in self._xml.iterchildren(tag=key): self._xml.remove(child) def __eq__(self, other): u"""Compares mappet objects. Two mappet objects are deemed equal if the lxmls object they represent are equal. """ return etree.tostring(self._xml) == etree.tostring(other._xml) def __contains__(self, path): u"""Check if object contains given path.""" elem = self.sget(path) return not (elem is None or elem is NONE_NODE) def __getstate__(self): u"""Converts the lxml to string for Pickling.""" return { '_xml': etree.tostring(self._xml, pretty_print=False) } def __setstate__(self, dict_): u"""Restores a Pickled mappet object.""" self._xml = etree.fromstring(dict_['_xml']) def __iter__(self): u"""Returns children as an iterator.""" return self.iter_children() def to_str(self, pretty_print=False, encoding=None, **kw): u"""Converts a node with all of it's children to a string. Remaining arguments are passed to etree.tostring as is. kwarg without_comments: bool because it works only in C14N flags: 'pretty print' and 'encoding' are ignored. :param bool pretty_print: whether to format the output :param str encoding: which encoding to use (ASCII by default) :rtype: str :returns: node's representation as a string """ if kw.get('without_comments') and not kw.get('method'): kw.pop('without_comments') kw['method'] = 'c14n' kw['with_comments'] = False return etree.tostring( self._xml, pretty_print=pretty_print, encoding=encoding, **kw ) def has_children(self): u"""Returns true if a node has children.""" return bool(len(self)) def iter_children(self, key=None): u"""Iterates over children. :param key: A key for filtering children by tagname. """ tag = None if key: tag = self._get_aliases().get(key) if not tag: raise KeyError(key) for child in self._xml.iterchildren(tag=tag): if len(child): yield self.__class__(child) else: yield Literal(child) def children(self, key=None): u"""Returns node's children. :param key: A key for filtering children by tagname. """ return list(self.iter_children(key)) def update(self, **kwargs): u"""Updating or creation of new simple nodes. Each dict key is used as a tagname and value as text. """ for key, value in kwargs.items(): helper = helpers.CAST_DICT.get(type(value), str) tag = self._get_aliases().get(key, key) elements = list(self._xml.iterchildren(tag=tag)) if elements: for element in elements: element.text = helper(value) else: element = etree.Element(key) element.text = helper(value) self._xml.append(element) self._aliases = None def sget(self, path, default=NONE_NODE): u"""Enables access to nodes if one or more of them don't exist. Example: >>> m = Mappet('<root><tag attr1="attr text">text value</tag></root>') >>> m.sget('tag') text value >>> m.sget('tag.@attr1') 'attr text' >>> m.sget('tag.#text') 'text value' >>> m.sget('reply.vms_model_cars.car.0.params.doors') NONE_NODE Accessing nonexistent path returns None-like object with mocked converting functions which returns None: >>> m.sget('reply.fake_node').to_dict() is None True """ attrs = str(path).split(".") text_or_attr = None last_attr = attrs[-1] # Case of getting text or attribute if last_attr == '#text' or last_attr.startswith('@'): # #text => text, @attr => attr text_or_attr = last_attr[1:] attrs = attrs[:-1] # When getting #text and @attr we want default value to be None. if default is NONE_NODE: default = None my_object = self for attr in attrs: try: if isinstance(my_object, (list, tuple)) and re.match('^\-?\d+$', attr): my_object_next = my_object[int(attr)] else: my_object_next = getattr(my_object, attr) my_object = my_object_next except (AttributeError, KeyError, IndexError): return default # Return #text or @attr if text_or_attr: try: return my_object.getattr(text_or_attr) except AttributeError: # myObject can be a list. return None else: return my_object def create(self, tag, value): u"""Creates a node, if it doesn't exist yet. Unlike attribute access, this allows to pass a node's name with hyphens. Those hyphens will be normalized automatically. In case the required element already exists, raises an exception. Updating/overwriting should be done using `update``. """ child_tags = {child.tag for child in self._xml} if tag in child_tags: raise KeyError('Node {} already exists in XML tree.'.format(tag)) self.set(tag, value) def set(self, name, value): u"""Assigns a new XML structure to the node. A literal value, dict or list can be passed in. Works for all nested levels. Dictionary: >>> m = Mappet('<root/>') >>> m.head = {'a': 'A', 'b': {'#text': 'B', '@attr': 'val'}} >>> m.head.to_str() '<head><a>A</a><b attr="val">B</b></head>' List: >>> m.head = [{'a': i} for i in 'ABC'] >>> m.head.to_str() '<head><a>A</a><a>B</a><a>C</a></head>' Literals: >>> m.head.leaf = 'A' >>> m.head.leaf.get() 'A' """ try: # Searches for a node to assign to. element = next(self._xml.iterchildren(tag=name)) except StopIteration: # There is no such node in the XML tree. We create a new one # with current root as parent (self._xml). element = etree.SubElement(self._xml, name) if isinstance(value, dict): self.assign_dict(element, value) elif isinstance(value, (list, tuple, set)): self.assign_sequence_or_set(element, value) else: # Literal value. self.assign_literal(element, value) # Clear the aliases. self._aliases = None def assign_dict(self, node, xml_dict): """Assigns a Python dict to a ``lxml`` node. :param node: A node to assign the dict to. :param xml_dict: The dict with attributes/children to use. """ new_node = etree.Element(node.tag) # Replaces the previous node with the new one self._xml.replace(node, new_node) # Copies #text and @attrs from the xml_dict helpers.dict_to_etree(xml_dict, new_node) @staticmethod def assign_sequence_or_set(element, value): element.clear() for item in value: temp_element = etree.Element('temp') helpers.dict_to_etree(item, temp_element) for child in temp_element.iterchildren(): element.append(child) del temp_element @staticmethod def assign_literal(element, value): u"""Assigns a literal. If a given node doesn't exist, it will be created. :param etree.Element element: element to which we assign. :param value: the value to assign """ # Searches for a conversion method specific to the type of value. helper = helpers.CAST_DICT.get(type(value), str) # Removes all children and attributes. element.clear() element.text = helper(value) def to_dict(self, **kw): u"""Converts the lxml object to a dict. possible kwargs: without_comments: bool """ _, value = helpers.etree_to_dict(self._xml, **kw).popitem() return value def _get_aliases(self): u"""Creates a dict with aliases. The key is a normalized tagname, value the original tagname. """ if self._aliases is None: self._aliases = {} if self._xml is not None: for child in self._xml.iterchildren(): self._aliases[helpers.normalize_tag(child.tag)] = child.tag return self._aliases def xpath( self, path, namespaces=None, regexp=False, smart_strings=True, single_use=False, ): u"""Executes XPath query on the ``lxml`` object and returns a correct object. :param str path: XPath string e.g., 'cars'/'car' :param str/dict namespaces: e.g., 'exslt', 're' or ``{'re': "http://exslt.org/regular-expressions"}`` :param bool regexp: if ``True`` and no namespaces is provided, it will use ``exslt`` namespace :param bool smart_strings: :param bool single_use: faster method for using only once. Does not create ``XPathEvaluator`` instance. >>> root = mappet.Mappet("<root><a>aB</a><b>aBc</b></root>") >>> root.XPath( "//*[re:test(., '^abc$', 'i')]", namespaces='exslt', regexp=True, ) """ if ( namespaces in ['exslt', 're'] or (regexp and not namespaces) ): namespaces = {'re': "http://exslt.org/regular-expressions"} if single_use: node = self._xml.xpath(path) else: xpe = self.xpath_evaluator( namespaces=namespaces, regexp=regexp, smart_strings=smart_strings ) node = xpe(path) if len(node) == 1: node = node[0] if len(node): return self.__class__(node) else: return Literal(node) return node def keys(self): """Returns a set of node's keys.""" return set(self._get_aliases().keys())
hackedd/gw2api
gw2api/__init__.py
set_cache_dir
python
def set_cache_dir(directory): global cache_dir if directory is None: cache_dir = None return if not os.path.exists(directory): os.makedirs(directory) if not os.path.isdir(directory): raise ValueError("not a directory") cache_dir = directory
Set the directory to cache JSON responses from most API endpoints.
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/__init__.py#L38-L51
null
import os import requests VERSION = "v1" BASE_URL = "https://api.guildwars2.com/%s/" % VERSION LANGUAGES = {"en": "English", "es": "Spanish", "de": "German", "fr": "French"} TYPE_COIN, TYPE_ITEM, TYPE_TEXT, TYPE_MAP = 1, 2, 3, 4 TYPE_PVP_GAME, TYPE_SKILL, TYPE_TRAIT, TYPE_USER = 5, 6, 7, 8 TYPE_RECIPE, TYPE_SKIN, TYPE_OUTFIT, TYPE_OBJECTIVE = 9, 10, 11, 12 LINK_TYPES = { "coin": TYPE_COIN, "item": TYPE_ITEM, "text": TYPE_TEXT, "map": TYPE_MAP, "pvp_game": TYPE_PVP_GAME, "skill": TYPE_SKILL, "trait": TYPE_TRAIT, "user": TYPE_USER, "recipe": TYPE_RECIPE, "skin": TYPE_SKIN, "outfit": TYPE_OUTFIT, "objective": TYPE_OBJECTIVE, } session = requests.Session() cache_dir = None cache_time = 14 * 24 * 3600 def set_session(sess): """Set the requests.Session to use for all API requests. """ global session session = sess def set_cache_time(time): """Set the maximum lifetime for a cached JSON response. """ global cache_time cache_time = time def get_mumble_link(): from .mumble import gw2link return gw2link from .map import * from .misc import * from .items import * from .skins import * from .events import * from .guild import * from .wvw import * from .util import *
hackedd/gw2api
gw2api/v2/endpoint.py
EndpointBase.get_cached
python
def get_cached(self, path, cache_name, **kwargs): if gw2api.cache_dir and gw2api.cache_time and cache_name: cache_file = os.path.join(gw2api.cache_dir, cache_name) if mtime(cache_file) >= time.time() - gw2api.cache_time: with open(cache_file, "r") as fp: tmp = json.load(fp) return self.make_response(tmp["data"], tmp["meta"]) else: cache_file = None meta, data = self._get(path, **kwargs) if cache_file: with open(cache_file, "w") as fp: json.dump({"meta": meta, "data": data}, fp, indent=2) return self.make_response(data, meta)
Request a resource form the API, first checking if there is a cached response available. Returns the parsed JSON data.
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/v2/endpoint.py#L42-L61
[ "def _get(self, path, **kwargs):\n token = kwargs.pop(\"token\") if \"token\" in kwargs else self.token\n if token:\n headers = kwargs.setdefault(\"headers\", {})\n headers.setdefault(\"Authorization\", \"Bearer \" + token)\n return super(AuthenticatedMixin, self)._get(path, **kwargs)\n", "def _get(self, path, **kwargs):\n r = gw2api.session.get(gw2api.v2.BASE_URL + path, **kwargs)\n\n try:\n response = r.json()\n except ValueError: # pragma: no cover\n response = None\n\n if not r.ok and isinstance(response, dict) and \"text\" in response:\n r.reason = response[\"text\"]\n\n r.raise_for_status()\n\n return self.get_metadata(r), response\n", "def make_response(self, data, meta):\n response_type = self.response_types.get(type(data))\n return response_type(data, meta) if response_type else data\n" ]
class EndpointBase(object): response_types = { list: ListResponse, dict: DictResponse, } def __init__(self, name): super(EndpointBase, self).__init__() self.name = name def has_cached(self, cache_name): if gw2api.cache_dir and gw2api.cache_time and cache_name: cache_file = os.path.join(gw2api.cache_dir, cache_name) return mtime(cache_file) >= time.time() - gw2api.cache_time else: return False def _get(self, path, **kwargs): r = gw2api.session.get(gw2api.v2.BASE_URL + path, **kwargs) try: response = r.json() except ValueError: # pragma: no cover response = None if not r.ok and isinstance(response, dict) and "text" in response: r.reason = response["text"] r.raise_for_status() return self.get_metadata(r), response def get_metadata(self, r): metadata = {} for key, link in r.links.items(): metadata[key] = link["url"] if "x-page-total" in r.headers: metadata["page_total"] = int(r.headers["x-page-total"]) if "x-page-size" in r.headers: metadata["page_size"] = int(r.headers["x-page-size"]) if "x-result-total" in r.headers: metadata["result_total"] = int(r.headers["x-result-total"]) if "x-result-count" in r.headers: metadata["result_count"] = int(r.headers["x-result-count"]) return metadata def make_response(self, data, meta): response_type = self.response_types.get(type(data)) return response_type(data, meta) if response_type else data
hackedd/gw2api
gw2api/skins.py
skin_details
python
def skin_details(skin_id, lang="en"): params = {"skin_id": skin_id, "lang": lang} cache_name = "skin_details.%(skin_id)s.%(lang)s.json" % params return get_cached("skin_details.json", cache_name, params=params)
This resource returns details about a single skin. :param skin_id: The skin to query for. :param lang: The language to display the texts in. The response is an object with at least the following properties. Note that the availability of some properties depends on the type of item the skin applies to. skin_id (number): The skin id. name (string): The name of the skin. type (string): The type of item the skin applies to. One of ``Armor``, ``Back`` or ``Weapon``. flags (list): Skin flags. Currently known skin flags are ``ShowInWardrobe``, ``HideIfLocked`` and ``NoCost``. restrictions (list): Race restrictions: ``Asura``, ``Charr``, ``Human``, ``Norn`` and ``Sylvari``. icon_file_id (string): The icon file id to be used with the render service. icon_file_signature (string): The icon file signature to be used with the render service.
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/skins.py#L16-L53
[ "def get_cached(path, cache_name=None, **kwargs):\n \"\"\"Request a resource form the API, first checking if there is a cached\n response available. Returns the parsed JSON data.\n \"\"\"\n if gw2api.cache_dir and gw2api.cache_time and cache_name is not False:\n if cache_name is None:\n cache_name = path\n cache_file = os.path.join(gw2api.cache_dir, cache_name)\n if mtime(cache_file) >= time.time() - gw2api.cache_time:\n with open(cache_file, \"r\") as fp:\n return json.load(fp)\n else:\n cache_file = None\n\n r = gw2api.session.get(gw2api.BASE_URL + path, **kwargs)\n\n if not r.ok:\n try:\n response = r.json()\n except ValueError: # pragma: no cover\n response = None\n if isinstance(response, dict) and \"text\" in response:\n r.reason = response[\"text\"]\n\n r.raise_for_status()\n data = r.json()\n\n if cache_file:\n with open(cache_file, \"w\") as fp:\n json.dump(data, fp, indent=2)\n\n return data\n" ]
from .util import get_cached __all__ = ("skins", "skin_details") def skins(): """This resource returns a list of skins that were discovered by players in the game. Details about a single skin can be obtained using the :func:`skin_details` resource. """ return get_cached("skins.json").get("skins")
hackedd/gw2api
gw2api/mumble.py
GuildWars2FileMapping.get_map_location
python
def get_map_location(self): map_data = self.get_map() (bounds_e, bounds_n), (bounds_w, bounds_s) = map_data["continent_rect"] (map_e, map_n), (map_w, map_s) = map_data["map_rect"] assert bounds_w < bounds_e assert bounds_n < bounds_s assert map_w < map_e assert map_n < map_s meters_to_inches = 39.3701 x, y, z = self.fAvatarPosition map_x = bounds_w + ((x * meters_to_inches - map_w) / (map_e - map_w) * (bounds_e - bounds_w)) map_y = bounds_n + ((-z * meters_to_inches - map_n) / (map_s - map_n) * (bounds_s - bounds_n)) map_z = y * meters_to_inches return map_x, map_y, map_z
Get the location of the player, converted to world coordinates. :return: a tuple (x, y, z).
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/mumble.py#L93-L117
[ "def get_map(self, lang=\"en\"):\n return maps(self.context.mapId, lang)\n" ]
class GuildWars2FileMapping(FileMapping): def __init__(self, name=u"MumbleLink", value_struct=LinkedMem, create=True): super(GuildWars2FileMapping, self).__init__(name, value_struct, create) def get_map(self, lang="en"): return maps(self.context.mapId, lang)
hackedd/gw2api
gw2api/wvw.py
matches
python
def matches(): wvw_matches = get_cached("wvw/matches.json", False).get("wvw_matches") for match in wvw_matches: match["start_time"] = parse_datetime(match["start_time"]) match["end_time"] = parse_datetime(match["end_time"]) return wvw_matches
This resource returns a list of the currently running WvW matches, with the participating worlds included in the result. Further details about a match can be requested using the ``match_details`` function. The response is a list of match objects, each of which contains the following properties: wvw_match_id (string): The WvW match id. red_world_id (number): The world id of the red world. blue_world_id (number): The world id of the blue world. green_world_id (number): The world id of the green world. start_time (datetime): A timestamp of when the match started. end_time (datetime): A timestamp of when the match ends.
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/wvw.py#L20-L51
[ "def get_cached(path, cache_name=None, **kwargs):\n \"\"\"Request a resource form the API, first checking if there is a cached\n response available. Returns the parsed JSON data.\n \"\"\"\n if gw2api.cache_dir and gw2api.cache_time and cache_name is not False:\n if cache_name is None:\n cache_name = path\n cache_file = os.path.join(gw2api.cache_dir, cache_name)\n if mtime(cache_file) >= time.time() - gw2api.cache_time:\n with open(cache_file, \"r\") as fp:\n return json.load(fp)\n else:\n cache_file = None\n\n r = gw2api.session.get(gw2api.BASE_URL + path, **kwargs)\n\n if not r.ok:\n try:\n response = r.json()\n except ValueError: # pragma: no cover\n response = None\n if isinstance(response, dict) and \"text\" in response:\n r.reason = response[\"text\"]\n\n r.raise_for_status()\n data = r.json()\n\n if cache_file:\n with open(cache_file, \"w\") as fp:\n json.dump(data, fp, indent=2)\n\n return data\n", "def parse_datetime(date_string):\n \"\"\"Parse a datetime string as returned by the ``matches`` endpoint to a\n datetime object.\n\n >>> parse_datetime('2014-07-04T18:00:00Z')\n datetime.datetime(2014, 7, 4, 18, 0)\n\n \"\"\"\n return datetime.strptime(date_string, \"%Y-%m-%dT%H:%M:%SZ\")\n" ]
from datetime import datetime from .util import get_cached __all__ = ("matches", "match_details", "objective_names") def parse_datetime(date_string): """Parse a datetime string as returned by the ``matches`` endpoint to a datetime object. >>> parse_datetime('2014-07-04T18:00:00Z') datetime.datetime(2014, 7, 4, 18, 0) """ return datetime.strptime(date_string, "%Y-%m-%dT%H:%M:%SZ") def match_details(match_id): """This resource returns further details about the specified match, including the total score and further details for each map. :param match_id: The WvW match to query for. The response is an object with the following properties: match_id (string): The WvW match id. scores (list): A list of the three total scores (order: red, blue, green). maps (list): A list of objects containing detailed information about each of the four maps. The map detail objects contain the following properties: type (string): The identifier for the map. Can be either RedHome, GreenHome or BlueHome for the borderlands or Center for Eternal Battlegrounds. scores (list): A list of the three individual scores for this map; in the order red, blue, green. objectives (list): A list of objective objects for this map. Each object contains the following properties: id (number): The objective id. owner (string): The current owner of the objective. Can be any one of Red, Green, Blue or Neutral. owner_guild (string): The guild id of the guild currently claiming the objective. This property is missing if the objective is not claimed. bonuses (list): A list of all bonuses being granted by this map. If no player team owns a bonus from the map, this list is empty. type (string): A shorthand name for the bonus. Currently the only known bonus type is ``bloodlust``: `Borderlands Bloodlust`_ owner (string): The current owner of the bonus. Can be any one of Red, Green, or Blue. Neutral-owned bonuses are not listed. .. _Borderlands Bloodlust: http://wiki.guildwars2.com/wiki/Borderlands_Bloodlust """ return get_cached("wvw/match_details.json", False, params={"match_id": match_id}) def objective_names(lang="en"): """This resource returns a list of the localized WvW objective names for the specified language. :param lang: The language to query the names for. :return: A dictionary mapping the objective Ids to the names. *Note that these are not the names displayed in the game, but rather the abstract type.* """ params = {"lang": lang} cache_name = "objective_names.%(lang)s.json" % params data = get_cached("wvw/objective_names.json", cache_name, params=params) return dict([(objective["id"], objective["name"]) for objective in data])
hackedd/gw2api
gw2api/wvw.py
objective_names
python
def objective_names(lang="en"): params = {"lang": lang} cache_name = "objective_names.%(lang)s.json" % params data = get_cached("wvw/objective_names.json", cache_name, params=params) return dict([(objective["id"], objective["name"]) for objective in data])
This resource returns a list of the localized WvW objective names for the specified language. :param lang: The language to query the names for. :return: A dictionary mapping the objective Ids to the names. *Note that these are not the names displayed in the game, but rather the abstract type.*
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/wvw.py#L117-L131
[ "def get_cached(path, cache_name=None, **kwargs):\n \"\"\"Request a resource form the API, first checking if there is a cached\n response available. Returns the parsed JSON data.\n \"\"\"\n if gw2api.cache_dir and gw2api.cache_time and cache_name is not False:\n if cache_name is None:\n cache_name = path\n cache_file = os.path.join(gw2api.cache_dir, cache_name)\n if mtime(cache_file) >= time.time() - gw2api.cache_time:\n with open(cache_file, \"r\") as fp:\n return json.load(fp)\n else:\n cache_file = None\n\n r = gw2api.session.get(gw2api.BASE_URL + path, **kwargs)\n\n if not r.ok:\n try:\n response = r.json()\n except ValueError: # pragma: no cover\n response = None\n if isinstance(response, dict) and \"text\" in response:\n r.reason = response[\"text\"]\n\n r.raise_for_status()\n data = r.json()\n\n if cache_file:\n with open(cache_file, \"w\") as fp:\n json.dump(data, fp, indent=2)\n\n return data\n" ]
from datetime import datetime from .util import get_cached __all__ = ("matches", "match_details", "objective_names") def parse_datetime(date_string): """Parse a datetime string as returned by the ``matches`` endpoint to a datetime object. >>> parse_datetime('2014-07-04T18:00:00Z') datetime.datetime(2014, 7, 4, 18, 0) """ return datetime.strptime(date_string, "%Y-%m-%dT%H:%M:%SZ") def matches(): """This resource returns a list of the currently running WvW matches, with the participating worlds included in the result. Further details about a match can be requested using the ``match_details`` function. The response is a list of match objects, each of which contains the following properties: wvw_match_id (string): The WvW match id. red_world_id (number): The world id of the red world. blue_world_id (number): The world id of the blue world. green_world_id (number): The world id of the green world. start_time (datetime): A timestamp of when the match started. end_time (datetime): A timestamp of when the match ends. """ wvw_matches = get_cached("wvw/matches.json", False).get("wvw_matches") for match in wvw_matches: match["start_time"] = parse_datetime(match["start_time"]) match["end_time"] = parse_datetime(match["end_time"]) return wvw_matches def match_details(match_id): """This resource returns further details about the specified match, including the total score and further details for each map. :param match_id: The WvW match to query for. The response is an object with the following properties: match_id (string): The WvW match id. scores (list): A list of the three total scores (order: red, blue, green). maps (list): A list of objects containing detailed information about each of the four maps. The map detail objects contain the following properties: type (string): The identifier for the map. Can be either RedHome, GreenHome or BlueHome for the borderlands or Center for Eternal Battlegrounds. scores (list): A list of the three individual scores for this map; in the order red, blue, green. objectives (list): A list of objective objects for this map. Each object contains the following properties: id (number): The objective id. owner (string): The current owner of the objective. Can be any one of Red, Green, Blue or Neutral. owner_guild (string): The guild id of the guild currently claiming the objective. This property is missing if the objective is not claimed. bonuses (list): A list of all bonuses being granted by this map. If no player team owns a bonus from the map, this list is empty. type (string): A shorthand name for the bonus. Currently the only known bonus type is ``bloodlust``: `Borderlands Bloodlust`_ owner (string): The current owner of the bonus. Can be any one of Red, Green, or Blue. Neutral-owned bonuses are not listed. .. _Borderlands Bloodlust: http://wiki.guildwars2.com/wiki/Borderlands_Bloodlust """ return get_cached("wvw/match_details.json", False, params={"match_id": match_id})
hackedd/gw2api
gw2api/util.py
mtime
python
def mtime(path): if not os.path.exists(path): return -1 stat = os.stat(path) return stat.st_mtime
Get the modification time of a file, or -1 if the file does not exist.
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/util.py#L14-L20
null
import os import time import json from struct import pack, unpack from base64 import b64encode, b64decode import gw2api __all__ = ("encode_item_link", "encode_coin_link", "encode_chat_link", "decode_chat_link") def get_cached(path, cache_name=None, **kwargs): """Request a resource form the API, first checking if there is a cached response available. Returns the parsed JSON data. """ if gw2api.cache_dir and gw2api.cache_time and cache_name is not False: if cache_name is None: cache_name = path cache_file = os.path.join(gw2api.cache_dir, cache_name) if mtime(cache_file) >= time.time() - gw2api.cache_time: with open(cache_file, "r") as fp: return json.load(fp) else: cache_file = None r = gw2api.session.get(gw2api.BASE_URL + path, **kwargs) if not r.ok: try: response = r.json() except ValueError: # pragma: no cover response = None if isinstance(response, dict) and "text" in response: r.reason = response["text"] r.raise_for_status() data = r.json() if cache_file: with open(cache_file, "w") as fp: json.dump(data, fp, indent=2) return data def encode_item_link(item_id, number=1, skin_id=None, upgrade1=None, upgrade2=None): """Encode a chat link for an item (or a stack of items). :param item_id: the Id of the item :param number: the number of items in the stack :param skin_id: the id of the skin applied to the item :param upgrade1: the id of the first upgrade component :param upgrade2: the id of the second upgrade component """ return encode_chat_link(gw2api.TYPE_ITEM, id=item_id, number=number, skin_id=skin_id, upgrade1=upgrade1, upgrade2=upgrade2) def encode_coin_link(copper, silver=0, gold=0): """Encode a chat link for an amount of coins. """ return encode_chat_link(gw2api.TYPE_COIN, copper=copper, silver=silver, gold=gold) def encode_chat_link(link_type, **kwargs): if link_type in gw2api.LINK_TYPES: link_type = gw2api.LINK_TYPES[link_type] if link_type == gw2api.TYPE_COIN: if "copper" in kwargs or "silver" in kwargs or "gold" in kwargs: amount = (kwargs.get("gold", 0) * 100 * 100 + kwargs.get("silver", 0) * 100 + kwargs.get("copper", 0)) else: amount = kwargs["amount"] data = pack("<BI", link_type, amount) elif link_type == gw2api.TYPE_ITEM: item_id = kwargs["id"] args = [] for i, key in enumerate(("skin_id", "upgrade1", "upgrade2")): value = kwargs.get(key) if value: item_id |= 2 << (28 + i) args.append(value) format = "<BBI" + "I" * len(args) data = pack(format, link_type, kwargs.get("number", 1), item_id, *args) elif link_type in (gw2api.TYPE_TEXT, gw2api.TYPE_MAP, gw2api.TYPE_SKILL, gw2api.TYPE_TRAIT, gw2api.TYPE_RECIPE, gw2api.TYPE_SKIN, gw2api.TYPE_OUTFIT): data = pack("<BI", link_type, kwargs["id"]) elif link_type == gw2api.TYPE_OBJECTIVE: data = pack("<BII", link_type, kwargs["objective_id"], kwargs["map_id"]) elif isinstance(link_type, int): raise Exception("Unknown link type 0x%02x" % link_type) else: raise Exception("Unknown link type '%s'" % link_type) return "[&%s]" % b64encode(data).decode("ascii") def decode_chat_link(string): if string.startswith("[&") and string.endswith("]"): string = string[2:-1] data = b64decode(string.encode("ascii")) link_type, = unpack("<B", data[:1]) if link_type == gw2api.TYPE_COIN: amount, = unpack("<I", data[1:]) return "coin", {"amount": amount} if link_type == gw2api.TYPE_ITEM: number, item_id = unpack("<BI", data[1:6]) flags = (item_id & 0xFF000000) >> 24 item_id &= 0x00FFFFFF values = {"number": number, "id": item_id} o = 6 if flags & 0x80: values["skin_id"], = unpack("<I", data[o:o+4]) o += 4 if flags & 0x40: values["upgrade1"], = unpack("<I", data[o:o+4]) o += 4 if flags & 0x20: values["upgrade2"], = unpack("<I", data[o:o+4]) o += 4 return "item", values link_type_string = None for key, value in gw2api.LINK_TYPES.items(): if value == link_type: link_type_string = key if link_type in (gw2api.TYPE_TEXT, gw2api.TYPE_MAP, gw2api.TYPE_SKILL, gw2api.TYPE_TRAIT, gw2api.TYPE_RECIPE, gw2api.TYPE_SKIN, gw2api.TYPE_OUTFIT): id, = unpack("<I", data[1:]) return link_type_string, {"id": id} if link_type == gw2api.TYPE_OBJECTIVE: objective_id, map_id = unpack("<II", data[1:]) return link_type_string, {"objective_id": objective_id, "map_id": map_id} raise Exception("Unknown link type 0x%02x" % link_type)
hackedd/gw2api
gw2api/util.py
get_cached
python
def get_cached(path, cache_name=None, **kwargs): if gw2api.cache_dir and gw2api.cache_time and cache_name is not False: if cache_name is None: cache_name = path cache_file = os.path.join(gw2api.cache_dir, cache_name) if mtime(cache_file) >= time.time() - gw2api.cache_time: with open(cache_file, "r") as fp: return json.load(fp) else: cache_file = None r = gw2api.session.get(gw2api.BASE_URL + path, **kwargs) if not r.ok: try: response = r.json() except ValueError: # pragma: no cover response = None if isinstance(response, dict) and "text" in response: r.reason = response["text"] r.raise_for_status() data = r.json() if cache_file: with open(cache_file, "w") as fp: json.dump(data, fp, indent=2) return data
Request a resource form the API, first checking if there is a cached response available. Returns the parsed JSON data.
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/util.py#L23-L54
null
import os import time import json from struct import pack, unpack from base64 import b64encode, b64decode import gw2api __all__ = ("encode_item_link", "encode_coin_link", "encode_chat_link", "decode_chat_link") def mtime(path): """Get the modification time of a file, or -1 if the file does not exist. """ if not os.path.exists(path): return -1 stat = os.stat(path) return stat.st_mtime def encode_item_link(item_id, number=1, skin_id=None, upgrade1=None, upgrade2=None): """Encode a chat link for an item (or a stack of items). :param item_id: the Id of the item :param number: the number of items in the stack :param skin_id: the id of the skin applied to the item :param upgrade1: the id of the first upgrade component :param upgrade2: the id of the second upgrade component """ return encode_chat_link(gw2api.TYPE_ITEM, id=item_id, number=number, skin_id=skin_id, upgrade1=upgrade1, upgrade2=upgrade2) def encode_coin_link(copper, silver=0, gold=0): """Encode a chat link for an amount of coins. """ return encode_chat_link(gw2api.TYPE_COIN, copper=copper, silver=silver, gold=gold) def encode_chat_link(link_type, **kwargs): if link_type in gw2api.LINK_TYPES: link_type = gw2api.LINK_TYPES[link_type] if link_type == gw2api.TYPE_COIN: if "copper" in kwargs or "silver" in kwargs or "gold" in kwargs: amount = (kwargs.get("gold", 0) * 100 * 100 + kwargs.get("silver", 0) * 100 + kwargs.get("copper", 0)) else: amount = kwargs["amount"] data = pack("<BI", link_type, amount) elif link_type == gw2api.TYPE_ITEM: item_id = kwargs["id"] args = [] for i, key in enumerate(("skin_id", "upgrade1", "upgrade2")): value = kwargs.get(key) if value: item_id |= 2 << (28 + i) args.append(value) format = "<BBI" + "I" * len(args) data = pack(format, link_type, kwargs.get("number", 1), item_id, *args) elif link_type in (gw2api.TYPE_TEXT, gw2api.TYPE_MAP, gw2api.TYPE_SKILL, gw2api.TYPE_TRAIT, gw2api.TYPE_RECIPE, gw2api.TYPE_SKIN, gw2api.TYPE_OUTFIT): data = pack("<BI", link_type, kwargs["id"]) elif link_type == gw2api.TYPE_OBJECTIVE: data = pack("<BII", link_type, kwargs["objective_id"], kwargs["map_id"]) elif isinstance(link_type, int): raise Exception("Unknown link type 0x%02x" % link_type) else: raise Exception("Unknown link type '%s'" % link_type) return "[&%s]" % b64encode(data).decode("ascii") def decode_chat_link(string): if string.startswith("[&") and string.endswith("]"): string = string[2:-1] data = b64decode(string.encode("ascii")) link_type, = unpack("<B", data[:1]) if link_type == gw2api.TYPE_COIN: amount, = unpack("<I", data[1:]) return "coin", {"amount": amount} if link_type == gw2api.TYPE_ITEM: number, item_id = unpack("<BI", data[1:6]) flags = (item_id & 0xFF000000) >> 24 item_id &= 0x00FFFFFF values = {"number": number, "id": item_id} o = 6 if flags & 0x80: values["skin_id"], = unpack("<I", data[o:o+4]) o += 4 if flags & 0x40: values["upgrade1"], = unpack("<I", data[o:o+4]) o += 4 if flags & 0x20: values["upgrade2"], = unpack("<I", data[o:o+4]) o += 4 return "item", values link_type_string = None for key, value in gw2api.LINK_TYPES.items(): if value == link_type: link_type_string = key if link_type in (gw2api.TYPE_TEXT, gw2api.TYPE_MAP, gw2api.TYPE_SKILL, gw2api.TYPE_TRAIT, gw2api.TYPE_RECIPE, gw2api.TYPE_SKIN, gw2api.TYPE_OUTFIT): id, = unpack("<I", data[1:]) return link_type_string, {"id": id} if link_type == gw2api.TYPE_OBJECTIVE: objective_id, map_id = unpack("<II", data[1:]) return link_type_string, {"objective_id": objective_id, "map_id": map_id} raise Exception("Unknown link type 0x%02x" % link_type)
hackedd/gw2api
gw2api/util.py
encode_item_link
python
def encode_item_link(item_id, number=1, skin_id=None, upgrade1=None, upgrade2=None): return encode_chat_link(gw2api.TYPE_ITEM, id=item_id, number=number, skin_id=skin_id, upgrade1=upgrade1, upgrade2=upgrade2)
Encode a chat link for an item (or a stack of items). :param item_id: the Id of the item :param number: the number of items in the stack :param skin_id: the id of the skin applied to the item :param upgrade1: the id of the first upgrade component :param upgrade2: the id of the second upgrade component
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/util.py#L57-L69
[ "def encode_chat_link(link_type, **kwargs):\n if link_type in gw2api.LINK_TYPES:\n link_type = gw2api.LINK_TYPES[link_type]\n\n if link_type == gw2api.TYPE_COIN:\n if \"copper\" in kwargs or \"silver\" in kwargs or \"gold\" in kwargs:\n amount = (kwargs.get(\"gold\", 0) * 100 * 100 +\n kwargs.get(\"silver\", 0) * 100 +\n kwargs.get(\"copper\", 0))\n else:\n amount = kwargs[\"amount\"]\n data = pack(\"<BI\", link_type, amount)\n\n elif link_type == gw2api.TYPE_ITEM:\n item_id = kwargs[\"id\"]\n\n args = []\n for i, key in enumerate((\"skin_id\", \"upgrade1\", \"upgrade2\")):\n value = kwargs.get(key)\n if value:\n item_id |= 2 << (28 + i)\n args.append(value)\n\n format = \"<BBI\" + \"I\" * len(args)\n data = pack(format, link_type, kwargs.get(\"number\", 1), item_id, *args)\n\n elif link_type in (gw2api.TYPE_TEXT, gw2api.TYPE_MAP, gw2api.TYPE_SKILL,\n gw2api.TYPE_TRAIT, gw2api.TYPE_RECIPE,\n gw2api.TYPE_SKIN, gw2api.TYPE_OUTFIT):\n data = pack(\"<BI\", link_type, kwargs[\"id\"])\n\n elif link_type == gw2api.TYPE_OBJECTIVE:\n data = pack(\"<BII\", link_type,\n kwargs[\"objective_id\"], kwargs[\"map_id\"])\n\n elif isinstance(link_type, int):\n raise Exception(\"Unknown link type 0x%02x\" % link_type)\n\n else:\n raise Exception(\"Unknown link type '%s'\" % link_type)\n\n return \"[&%s]\" % b64encode(data).decode(\"ascii\")\n" ]
import os import time import json from struct import pack, unpack from base64 import b64encode, b64decode import gw2api __all__ = ("encode_item_link", "encode_coin_link", "encode_chat_link", "decode_chat_link") def mtime(path): """Get the modification time of a file, or -1 if the file does not exist. """ if not os.path.exists(path): return -1 stat = os.stat(path) return stat.st_mtime def get_cached(path, cache_name=None, **kwargs): """Request a resource form the API, first checking if there is a cached response available. Returns the parsed JSON data. """ if gw2api.cache_dir and gw2api.cache_time and cache_name is not False: if cache_name is None: cache_name = path cache_file = os.path.join(gw2api.cache_dir, cache_name) if mtime(cache_file) >= time.time() - gw2api.cache_time: with open(cache_file, "r") as fp: return json.load(fp) else: cache_file = None r = gw2api.session.get(gw2api.BASE_URL + path, **kwargs) if not r.ok: try: response = r.json() except ValueError: # pragma: no cover response = None if isinstance(response, dict) and "text" in response: r.reason = response["text"] r.raise_for_status() data = r.json() if cache_file: with open(cache_file, "w") as fp: json.dump(data, fp, indent=2) return data def encode_coin_link(copper, silver=0, gold=0): """Encode a chat link for an amount of coins. """ return encode_chat_link(gw2api.TYPE_COIN, copper=copper, silver=silver, gold=gold) def encode_chat_link(link_type, **kwargs): if link_type in gw2api.LINK_TYPES: link_type = gw2api.LINK_TYPES[link_type] if link_type == gw2api.TYPE_COIN: if "copper" in kwargs or "silver" in kwargs or "gold" in kwargs: amount = (kwargs.get("gold", 0) * 100 * 100 + kwargs.get("silver", 0) * 100 + kwargs.get("copper", 0)) else: amount = kwargs["amount"] data = pack("<BI", link_type, amount) elif link_type == gw2api.TYPE_ITEM: item_id = kwargs["id"] args = [] for i, key in enumerate(("skin_id", "upgrade1", "upgrade2")): value = kwargs.get(key) if value: item_id |= 2 << (28 + i) args.append(value) format = "<BBI" + "I" * len(args) data = pack(format, link_type, kwargs.get("number", 1), item_id, *args) elif link_type in (gw2api.TYPE_TEXT, gw2api.TYPE_MAP, gw2api.TYPE_SKILL, gw2api.TYPE_TRAIT, gw2api.TYPE_RECIPE, gw2api.TYPE_SKIN, gw2api.TYPE_OUTFIT): data = pack("<BI", link_type, kwargs["id"]) elif link_type == gw2api.TYPE_OBJECTIVE: data = pack("<BII", link_type, kwargs["objective_id"], kwargs["map_id"]) elif isinstance(link_type, int): raise Exception("Unknown link type 0x%02x" % link_type) else: raise Exception("Unknown link type '%s'" % link_type) return "[&%s]" % b64encode(data).decode("ascii") def decode_chat_link(string): if string.startswith("[&") and string.endswith("]"): string = string[2:-1] data = b64decode(string.encode("ascii")) link_type, = unpack("<B", data[:1]) if link_type == gw2api.TYPE_COIN: amount, = unpack("<I", data[1:]) return "coin", {"amount": amount} if link_type == gw2api.TYPE_ITEM: number, item_id = unpack("<BI", data[1:6]) flags = (item_id & 0xFF000000) >> 24 item_id &= 0x00FFFFFF values = {"number": number, "id": item_id} o = 6 if flags & 0x80: values["skin_id"], = unpack("<I", data[o:o+4]) o += 4 if flags & 0x40: values["upgrade1"], = unpack("<I", data[o:o+4]) o += 4 if flags & 0x20: values["upgrade2"], = unpack("<I", data[o:o+4]) o += 4 return "item", values link_type_string = None for key, value in gw2api.LINK_TYPES.items(): if value == link_type: link_type_string = key if link_type in (gw2api.TYPE_TEXT, gw2api.TYPE_MAP, gw2api.TYPE_SKILL, gw2api.TYPE_TRAIT, gw2api.TYPE_RECIPE, gw2api.TYPE_SKIN, gw2api.TYPE_OUTFIT): id, = unpack("<I", data[1:]) return link_type_string, {"id": id} if link_type == gw2api.TYPE_OBJECTIVE: objective_id, map_id = unpack("<II", data[1:]) return link_type_string, {"objective_id": objective_id, "map_id": map_id} raise Exception("Unknown link type 0x%02x" % link_type)
hackedd/gw2api
gw2api/util.py
encode_coin_link
python
def encode_coin_link(copper, silver=0, gold=0): return encode_chat_link(gw2api.TYPE_COIN, copper=copper, silver=silver, gold=gold)
Encode a chat link for an amount of coins.
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/util.py#L72-L76
[ "def encode_chat_link(link_type, **kwargs):\n if link_type in gw2api.LINK_TYPES:\n link_type = gw2api.LINK_TYPES[link_type]\n\n if link_type == gw2api.TYPE_COIN:\n if \"copper\" in kwargs or \"silver\" in kwargs or \"gold\" in kwargs:\n amount = (kwargs.get(\"gold\", 0) * 100 * 100 +\n kwargs.get(\"silver\", 0) * 100 +\n kwargs.get(\"copper\", 0))\n else:\n amount = kwargs[\"amount\"]\n data = pack(\"<BI\", link_type, amount)\n\n elif link_type == gw2api.TYPE_ITEM:\n item_id = kwargs[\"id\"]\n\n args = []\n for i, key in enumerate((\"skin_id\", \"upgrade1\", \"upgrade2\")):\n value = kwargs.get(key)\n if value:\n item_id |= 2 << (28 + i)\n args.append(value)\n\n format = \"<BBI\" + \"I\" * len(args)\n data = pack(format, link_type, kwargs.get(\"number\", 1), item_id, *args)\n\n elif link_type in (gw2api.TYPE_TEXT, gw2api.TYPE_MAP, gw2api.TYPE_SKILL,\n gw2api.TYPE_TRAIT, gw2api.TYPE_RECIPE,\n gw2api.TYPE_SKIN, gw2api.TYPE_OUTFIT):\n data = pack(\"<BI\", link_type, kwargs[\"id\"])\n\n elif link_type == gw2api.TYPE_OBJECTIVE:\n data = pack(\"<BII\", link_type,\n kwargs[\"objective_id\"], kwargs[\"map_id\"])\n\n elif isinstance(link_type, int):\n raise Exception(\"Unknown link type 0x%02x\" % link_type)\n\n else:\n raise Exception(\"Unknown link type '%s'\" % link_type)\n\n return \"[&%s]\" % b64encode(data).decode(\"ascii\")\n" ]
import os import time import json from struct import pack, unpack from base64 import b64encode, b64decode import gw2api __all__ = ("encode_item_link", "encode_coin_link", "encode_chat_link", "decode_chat_link") def mtime(path): """Get the modification time of a file, or -1 if the file does not exist. """ if not os.path.exists(path): return -1 stat = os.stat(path) return stat.st_mtime def get_cached(path, cache_name=None, **kwargs): """Request a resource form the API, first checking if there is a cached response available. Returns the parsed JSON data. """ if gw2api.cache_dir and gw2api.cache_time and cache_name is not False: if cache_name is None: cache_name = path cache_file = os.path.join(gw2api.cache_dir, cache_name) if mtime(cache_file) >= time.time() - gw2api.cache_time: with open(cache_file, "r") as fp: return json.load(fp) else: cache_file = None r = gw2api.session.get(gw2api.BASE_URL + path, **kwargs) if not r.ok: try: response = r.json() except ValueError: # pragma: no cover response = None if isinstance(response, dict) and "text" in response: r.reason = response["text"] r.raise_for_status() data = r.json() if cache_file: with open(cache_file, "w") as fp: json.dump(data, fp, indent=2) return data def encode_item_link(item_id, number=1, skin_id=None, upgrade1=None, upgrade2=None): """Encode a chat link for an item (or a stack of items). :param item_id: the Id of the item :param number: the number of items in the stack :param skin_id: the id of the skin applied to the item :param upgrade1: the id of the first upgrade component :param upgrade2: the id of the second upgrade component """ return encode_chat_link(gw2api.TYPE_ITEM, id=item_id, number=number, skin_id=skin_id, upgrade1=upgrade1, upgrade2=upgrade2) def encode_chat_link(link_type, **kwargs): if link_type in gw2api.LINK_TYPES: link_type = gw2api.LINK_TYPES[link_type] if link_type == gw2api.TYPE_COIN: if "copper" in kwargs or "silver" in kwargs or "gold" in kwargs: amount = (kwargs.get("gold", 0) * 100 * 100 + kwargs.get("silver", 0) * 100 + kwargs.get("copper", 0)) else: amount = kwargs["amount"] data = pack("<BI", link_type, amount) elif link_type == gw2api.TYPE_ITEM: item_id = kwargs["id"] args = [] for i, key in enumerate(("skin_id", "upgrade1", "upgrade2")): value = kwargs.get(key) if value: item_id |= 2 << (28 + i) args.append(value) format = "<BBI" + "I" * len(args) data = pack(format, link_type, kwargs.get("number", 1), item_id, *args) elif link_type in (gw2api.TYPE_TEXT, gw2api.TYPE_MAP, gw2api.TYPE_SKILL, gw2api.TYPE_TRAIT, gw2api.TYPE_RECIPE, gw2api.TYPE_SKIN, gw2api.TYPE_OUTFIT): data = pack("<BI", link_type, kwargs["id"]) elif link_type == gw2api.TYPE_OBJECTIVE: data = pack("<BII", link_type, kwargs["objective_id"], kwargs["map_id"]) elif isinstance(link_type, int): raise Exception("Unknown link type 0x%02x" % link_type) else: raise Exception("Unknown link type '%s'" % link_type) return "[&%s]" % b64encode(data).decode("ascii") def decode_chat_link(string): if string.startswith("[&") and string.endswith("]"): string = string[2:-1] data = b64decode(string.encode("ascii")) link_type, = unpack("<B", data[:1]) if link_type == gw2api.TYPE_COIN: amount, = unpack("<I", data[1:]) return "coin", {"amount": amount} if link_type == gw2api.TYPE_ITEM: number, item_id = unpack("<BI", data[1:6]) flags = (item_id & 0xFF000000) >> 24 item_id &= 0x00FFFFFF values = {"number": number, "id": item_id} o = 6 if flags & 0x80: values["skin_id"], = unpack("<I", data[o:o+4]) o += 4 if flags & 0x40: values["upgrade1"], = unpack("<I", data[o:o+4]) o += 4 if flags & 0x20: values["upgrade2"], = unpack("<I", data[o:o+4]) o += 4 return "item", values link_type_string = None for key, value in gw2api.LINK_TYPES.items(): if value == link_type: link_type_string = key if link_type in (gw2api.TYPE_TEXT, gw2api.TYPE_MAP, gw2api.TYPE_SKILL, gw2api.TYPE_TRAIT, gw2api.TYPE_RECIPE, gw2api.TYPE_SKIN, gw2api.TYPE_OUTFIT): id, = unpack("<I", data[1:]) return link_type_string, {"id": id} if link_type == gw2api.TYPE_OBJECTIVE: objective_id, map_id = unpack("<II", data[1:]) return link_type_string, {"objective_id": objective_id, "map_id": map_id} raise Exception("Unknown link type 0x%02x" % link_type)
hackedd/gw2api
gw2api/items.py
item_details
python
def item_details(item_id, lang="en"): params = {"item_id": item_id, "lang": lang} cache_name = "item_details.%(item_id)s.%(lang)s.json" % params return get_cached("item_details.json", cache_name, params=params)
This resource returns a details about a single item. :param item_id: The item to query for. :param lang: The language to display the texts in. The response is an object with at least the following properties. Note that the availability of some properties depends on the type of the item. item_id (number): The item id. name (string): The name of the item. description (string): The item description. type (string): The item type. level (integer): The required level. rarity (string): The rarity. On of ``Junk``, ``Basic``, ``Fine``, ``Masterwork``, ``Rare``, ``Exotic``, ``Ascended`` or ``Legendary``. vendor_value (integer): The value in coins when selling to a vendor. icon_file_id (string): The icon file id to be used with the render service. icon_file_signature (string): The icon file signature to be used with the render service. game_types (list): The game types where the item is usable. Currently known game types are: ``Activity``, ``Dungeon``, ``Pve``, ``Pvp``, ``PvpLobby`` and ``WvW`` flags (list): Additional item flags. Currently known item flags are: ``AccountBound``, ``HideSuffix``, ``NoMysticForge``, ``NoSalvage``, ``NoSell``, ``NotUpgradeable``, ``NoUnderwater``, ``SoulbindOnAcquire``, ``SoulBindOnUse`` and ``Unique`` restrictions (list): Race restrictions: ``Asura``, ``Charr``, ``Human``, ``Norn`` and ``Sylvari``. Each item type has an `additional key`_ with information specific to that item type. .. _additional key: item-properties.html
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/items.py#L24-L85
[ "def get_cached(path, cache_name=None, **kwargs):\n \"\"\"Request a resource form the API, first checking if there is a cached\n response available. Returns the parsed JSON data.\n \"\"\"\n if gw2api.cache_dir and gw2api.cache_time and cache_name is not False:\n if cache_name is None:\n cache_name = path\n cache_file = os.path.join(gw2api.cache_dir, cache_name)\n if mtime(cache_file) >= time.time() - gw2api.cache_time:\n with open(cache_file, \"r\") as fp:\n return json.load(fp)\n else:\n cache_file = None\n\n r = gw2api.session.get(gw2api.BASE_URL + path, **kwargs)\n\n if not r.ok:\n try:\n response = r.json()\n except ValueError: # pragma: no cover\n response = None\n if isinstance(response, dict) and \"text\" in response:\n r.reason = response[\"text\"]\n\n r.raise_for_status()\n data = r.json()\n\n if cache_file:\n with open(cache_file, \"w\") as fp:\n json.dump(data, fp, indent=2)\n\n return data\n" ]
from .util import get_cached __all__ = ("items", "recipes", "item_details", "recipe_details") def items(): """This resource returns a list of items that were discovered by players in the game. Details about a single item can be obtained using the :func:`item_details` resource. """ return get_cached("items.json").get("items") def recipes(): """This resource returns a list of recipes that were discovered by players in the game. Details about a single recipe can be obtained using the :func:`recipe_details` resource. """ return get_cached("recipes.json").get("recipes") def recipe_details(recipe_id, lang="en"): """This resource returns a details about a single recipe. :param recipe_id: The recipe to query for. :param lang: The language to display the texts in. The response is an object with the following properties: recipe_id (number): The recipe id. type (string): The type of the produced item. output_item_id (string): The item id of the produced item. output_item_count (string): The amount of items produced. min_rating (string): The minimum rating of the recipe. time_to_craft_ms (string): The time it takes to craft the item. disciplines (list): A list of crafting disciplines that can use the recipe. flags (list): Additional recipe flags. Known flags: ``AutoLearned``: Set for recipes that don't have to be discovered. ``LearnedFromItem``: Set for recipes that need a recipe sheet. ingredients (list): A list of objects describing the ingredients for this recipe. Each object contains the following properties: item_id (string): The item id of the ingredient. count (string): The amount of ingredients required. """ params = {"recipe_id": recipe_id, "lang": lang} cache_name = "recipe_details.%(recipe_id)s.%(lang)s.json" % params return get_cached("recipe_details.json", cache_name, params=params)
hackedd/gw2api
gw2api/items.py
recipe_details
python
def recipe_details(recipe_id, lang="en"): params = {"recipe_id": recipe_id, "lang": lang} cache_name = "recipe_details.%(recipe_id)s.%(lang)s.json" % params return get_cached("recipe_details.json", cache_name, params=params)
This resource returns a details about a single recipe. :param recipe_id: The recipe to query for. :param lang: The language to display the texts in. The response is an object with the following properties: recipe_id (number): The recipe id. type (string): The type of the produced item. output_item_id (string): The item id of the produced item. output_item_count (string): The amount of items produced. min_rating (string): The minimum rating of the recipe. time_to_craft_ms (string): The time it takes to craft the item. disciplines (list): A list of crafting disciplines that can use the recipe. flags (list): Additional recipe flags. Known flags: ``AutoLearned``: Set for recipes that don't have to be discovered. ``LearnedFromItem``: Set for recipes that need a recipe sheet. ingredients (list): A list of objects describing the ingredients for this recipe. Each object contains the following properties: item_id (string): The item id of the ingredient. count (string): The amount of ingredients required.
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/items.py#L88-L139
[ "def get_cached(path, cache_name=None, **kwargs):\n \"\"\"Request a resource form the API, first checking if there is a cached\n response available. Returns the parsed JSON data.\n \"\"\"\n if gw2api.cache_dir and gw2api.cache_time and cache_name is not False:\n if cache_name is None:\n cache_name = path\n cache_file = os.path.join(gw2api.cache_dir, cache_name)\n if mtime(cache_file) >= time.time() - gw2api.cache_time:\n with open(cache_file, \"r\") as fp:\n return json.load(fp)\n else:\n cache_file = None\n\n r = gw2api.session.get(gw2api.BASE_URL + path, **kwargs)\n\n if not r.ok:\n try:\n response = r.json()\n except ValueError: # pragma: no cover\n response = None\n if isinstance(response, dict) and \"text\" in response:\n r.reason = response[\"text\"]\n\n r.raise_for_status()\n data = r.json()\n\n if cache_file:\n with open(cache_file, \"w\") as fp:\n json.dump(data, fp, indent=2)\n\n return data\n" ]
from .util import get_cached __all__ = ("items", "recipes", "item_details", "recipe_details") def items(): """This resource returns a list of items that were discovered by players in the game. Details about a single item can be obtained using the :func:`item_details` resource. """ return get_cached("items.json").get("items") def recipes(): """This resource returns a list of recipes that were discovered by players in the game. Details about a single recipe can be obtained using the :func:`recipe_details` resource. """ return get_cached("recipes.json").get("recipes") def item_details(item_id, lang="en"): """This resource returns a details about a single item. :param item_id: The item to query for. :param lang: The language to display the texts in. The response is an object with at least the following properties. Note that the availability of some properties depends on the type of the item. item_id (number): The item id. name (string): The name of the item. description (string): The item description. type (string): The item type. level (integer): The required level. rarity (string): The rarity. On of ``Junk``, ``Basic``, ``Fine``, ``Masterwork``, ``Rare``, ``Exotic``, ``Ascended`` or ``Legendary``. vendor_value (integer): The value in coins when selling to a vendor. icon_file_id (string): The icon file id to be used with the render service. icon_file_signature (string): The icon file signature to be used with the render service. game_types (list): The game types where the item is usable. Currently known game types are: ``Activity``, ``Dungeon``, ``Pve``, ``Pvp``, ``PvpLobby`` and ``WvW`` flags (list): Additional item flags. Currently known item flags are: ``AccountBound``, ``HideSuffix``, ``NoMysticForge``, ``NoSalvage``, ``NoSell``, ``NotUpgradeable``, ``NoUnderwater``, ``SoulbindOnAcquire``, ``SoulBindOnUse`` and ``Unique`` restrictions (list): Race restrictions: ``Asura``, ``Charr``, ``Human``, ``Norn`` and ``Sylvari``. Each item type has an `additional key`_ with information specific to that item type. .. _additional key: item-properties.html """ params = {"item_id": item_id, "lang": lang} cache_name = "item_details.%(item_id)s.%(lang)s.json" % params return get_cached("item_details.json", cache_name, params=params)
hackedd/gw2api
gw2api/misc.py
colors
python
def colors(lang="en"): cache_name = "colors.%s.json" % lang data = get_cached("colors.json", cache_name, params=dict(lang=lang)) return data["colors"]
This resource returns all dyes in the game, including localized names and their color component information. :param lang: The language to query the names for. The response is a dictionary where color ids are mapped to an dictionary containing the following properties: name (string): The name of the dye. base_rgb (list): The base RGB values. cloth (object): Detailed information on its appearance when applied on cloth armor. leather (object): Detailed information on its appearance when applied on leather armor. metal (object): Detailed information on its appearance when applied on metal armor. The detailed information object contains the following properties: brightness (number): The brightness. contrast (number): The contrast. hue (number): The hue in the HSL colorspace. saturation (number): The saturation in the HSL colorspace. lightness (number): The lightness in the HSL colorspace. rgb (list): A list containing precalculated RGB values.
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/misc.py#L15-L62
[ "def get_cached(path, cache_name=None, **kwargs):\n \"\"\"Request a resource form the API, first checking if there is a cached\n response available. Returns the parsed JSON data.\n \"\"\"\n if gw2api.cache_dir and gw2api.cache_time and cache_name is not False:\n if cache_name is None:\n cache_name = path\n cache_file = os.path.join(gw2api.cache_dir, cache_name)\n if mtime(cache_file) >= time.time() - gw2api.cache_time:\n with open(cache_file, \"r\") as fp:\n return json.load(fp)\n else:\n cache_file = None\n\n r = gw2api.session.get(gw2api.BASE_URL + path, **kwargs)\n\n if not r.ok:\n try:\n response = r.json()\n except ValueError: # pragma: no cover\n response = None\n if isinstance(response, dict) and \"text\" in response:\n r.reason = response[\"text\"]\n\n r.raise_for_status()\n data = r.json()\n\n if cache_file:\n with open(cache_file, \"w\") as fp:\n json.dump(data, fp, indent=2)\n\n return data\n" ]
from .util import get_cached __all__ = ("build", "colors", "files") def build(): """This resource returns the current build id of the game. The result of this function is not cached. """ return get_cached("build.json", False).get("build_id") def files(): """This resource returns commonly requested in-game assets that may be used to enhance API-derived applications. The returned information can be used with the render service to retrieve assets. The response is an object where file identifiers are mapped to an object containing the following properties: file_id (string): The file id to be used with the render service. signature (string): The file signature to be used with the render service. """ return get_cached("files.json")
hackedd/gw2api
gw2api/events.py
event_names
python
def event_names(lang="en"): cache_name = "event_names.%s.json" % lang data = get_cached("event_names.json", cache_name, params=dict(lang=lang)) return dict([(event["id"], event["name"]) for event in data])
This resource returns an unordered list of the localized event names for the specified language. :param lang: The language to query the names for. :return: A dictionary where the key is the event id and the value is the name of the event in the specified language.
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/events.py#L7-L18
[ "def get_cached(path, cache_name=None, **kwargs):\n \"\"\"Request a resource form the API, first checking if there is a cached\n response available. Returns the parsed JSON data.\n \"\"\"\n if gw2api.cache_dir and gw2api.cache_time and cache_name is not False:\n if cache_name is None:\n cache_name = path\n cache_file = os.path.join(gw2api.cache_dir, cache_name)\n if mtime(cache_file) >= time.time() - gw2api.cache_time:\n with open(cache_file, \"r\") as fp:\n return json.load(fp)\n else:\n cache_file = None\n\n r = gw2api.session.get(gw2api.BASE_URL + path, **kwargs)\n\n if not r.ok:\n try:\n response = r.json()\n except ValueError: # pragma: no cover\n response = None\n if isinstance(response, dict) and \"text\" in response:\n r.reason = response[\"text\"]\n\n r.raise_for_status()\n data = r.json()\n\n if cache_file:\n with open(cache_file, \"w\") as fp:\n json.dump(data, fp, indent=2)\n\n return data\n" ]
from .util import get_cached __all__ = ("event_names", "event_details") def event_details(event_id=None, lang="en"): """This resource returns static details about available events. :param event_id: Only list this event. :param lang: Show localized texts in the specified language. The response is a dictionary where the key is the event id, and the value is a dictionary containing the following properties: name (string) The name of the event. level (int) The event level. map_id (int) The map where the event takes place. flags (list) A list of additional flags. Possible flags are: ``group_event`` For group events ``map_wide`` For map-wide events. location (object) The location of the event. type (string) The type of the event location, can be ``sphere``, ``cylinder`` or ``poly``. center (list) X, Y, Z coordinates of the event location. radius (number) (type ``sphere`` and ``cylinder``) Radius of the event location. z_range (list) (type ``poly``) List of Minimum and Maximum Z coordinate. points (list) (type ``poly``) List of Points (X, Y) denoting the event location perimeter. If a event_id is given, only the values for that event are returned. """ if event_id: cache_name = "event_details.%s.%s.json" % (event_id, lang) params = {"event_id": event_id, "lang": lang} else: cache_name = "event_details.%s.json" % lang params = {"lang": lang} data = get_cached("event_details.json", cache_name, params=params) events = data["events"] return events.get(event_id) if event_id else events
hackedd/gw2api
gw2api/events.py
event_details
python
def event_details(event_id=None, lang="en"): if event_id: cache_name = "event_details.%s.%s.json" % (event_id, lang) params = {"event_id": event_id, "lang": lang} else: cache_name = "event_details.%s.json" % lang params = {"lang": lang} data = get_cached("event_details.json", cache_name, params=params) events = data["events"] return events.get(event_id) if event_id else events
This resource returns static details about available events. :param event_id: Only list this event. :param lang: Show localized texts in the specified language. The response is a dictionary where the key is the event id, and the value is a dictionary containing the following properties: name (string) The name of the event. level (int) The event level. map_id (int) The map where the event takes place. flags (list) A list of additional flags. Possible flags are: ``group_event`` For group events ``map_wide`` For map-wide events. location (object) The location of the event. type (string) The type of the event location, can be ``sphere``, ``cylinder`` or ``poly``. center (list) X, Y, Z coordinates of the event location. radius (number) (type ``sphere`` and ``cylinder``) Radius of the event location. z_range (list) (type ``poly``) List of Minimum and Maximum Z coordinate. points (list) (type ``poly``) List of Points (X, Y) denoting the event location perimeter. If a event_id is given, only the values for that event are returned.
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/events.py#L21-L79
[ "def get_cached(path, cache_name=None, **kwargs):\n \"\"\"Request a resource form the API, first checking if there is a cached\n response available. Returns the parsed JSON data.\n \"\"\"\n if gw2api.cache_dir and gw2api.cache_time and cache_name is not False:\n if cache_name is None:\n cache_name = path\n cache_file = os.path.join(gw2api.cache_dir, cache_name)\n if mtime(cache_file) >= time.time() - gw2api.cache_time:\n with open(cache_file, \"r\") as fp:\n return json.load(fp)\n else:\n cache_file = None\n\n r = gw2api.session.get(gw2api.BASE_URL + path, **kwargs)\n\n if not r.ok:\n try:\n response = r.json()\n except ValueError: # pragma: no cover\n response = None\n if isinstance(response, dict) and \"text\" in response:\n r.reason = response[\"text\"]\n\n r.raise_for_status()\n data = r.json()\n\n if cache_file:\n with open(cache_file, \"w\") as fp:\n json.dump(data, fp, indent=2)\n\n return data\n" ]
from .util import get_cached __all__ = ("event_names", "event_details") def event_names(lang="en"): """This resource returns an unordered list of the localized event names for the specified language. :param lang: The language to query the names for. :return: A dictionary where the key is the event id and the value is the name of the event in the specified language. """ cache_name = "event_names.%s.json" % lang data = get_cached("event_names.json", cache_name, params=dict(lang=lang)) return dict([(event["id"], event["name"]) for event in data])
hackedd/gw2api
gw2api/guild.py
guild_details
python
def guild_details(guild_id=None, name=None): if guild_id and name: warnings.warn("both guild_id and name are specified, " "name will be ignored") if guild_id: params = {"guild_id": guild_id} cache_name = "guild_details.%s.json" % guild_id elif name: params = {"guild_name": name} cache_name = "guild_details.%s.json" % name else: raise Exception("specify either guild_id or name") return get_cached("guild_details.json", cache_name, params=params)
This resource returns details about a guild. :param guild_id: The guild id to query for. :param name: The guild name to query for. *Note: Only one parameter is required; if both are set, the guild Id takes precedence and a warning will be logged.* The response is a dictionary with the following keys: guild_id (string): The guild id. guild_name (string): The guild name. tag (string): The guild tag. emblem (object): If present, it holds detailed information about the guilds emblem. The emblem dictionary contains the following information: background_id (number): The id of the background image. foreground_id (number): The id of the foreground image. flags (list): A list of additional flags, possible values are: ``FlipBackgroundHorizontal``, ``FlipBackgroundVertical``, ``FlipForegroundHorizontal`` and ``FlipForegroundVertical``. background_color_id (number): The background color id. foreground_primary_color_id (number): The primary foreground color id. foreground_secondary_color_id (number): The secondary foreground color id.
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/guild.py#L9-L68
[ "def get_cached(path, cache_name=None, **kwargs):\n \"\"\"Request a resource form the API, first checking if there is a cached\n response available. Returns the parsed JSON data.\n \"\"\"\n if gw2api.cache_dir and gw2api.cache_time and cache_name is not False:\n if cache_name is None:\n cache_name = path\n cache_file = os.path.join(gw2api.cache_dir, cache_name)\n if mtime(cache_file) >= time.time() - gw2api.cache_time:\n with open(cache_file, \"r\") as fp:\n return json.load(fp)\n else:\n cache_file = None\n\n r = gw2api.session.get(gw2api.BASE_URL + path, **kwargs)\n\n if not r.ok:\n try:\n response = r.json()\n except ValueError: # pragma: no cover\n response = None\n if isinstance(response, dict) and \"text\" in response:\n r.reason = response[\"text\"]\n\n r.raise_for_status()\n data = r.json()\n\n if cache_file:\n with open(cache_file, \"w\") as fp:\n json.dump(data, fp, indent=2)\n\n return data\n" ]
import warnings from .util import get_cached __all__ = ("guild_details", )
hackedd/gw2api
gw2api/map.py
map_names
python
def map_names(lang="en"): cache_name = "map_names.%s.json" % lang data = get_cached("map_names.json", cache_name, params=dict(lang=lang)) return dict([(item["id"], item["name"]) for item in data])
This resource returns an dictionary of the localized map names for the specified language. Only maps with events are listed - if you need a list of all maps, use ``maps.json`` instead. :param lang: The language to query the names for. :return: the response is a dictionary where the key is the map id and the value is the name of the map in the specified language.
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/map.py#L35-L47
[ "def get_cached(path, cache_name=None, **kwargs):\n \"\"\"Request a resource form the API, first checking if there is a cached\n response available. Returns the parsed JSON data.\n \"\"\"\n if gw2api.cache_dir and gw2api.cache_time and cache_name is not False:\n if cache_name is None:\n cache_name = path\n cache_file = os.path.join(gw2api.cache_dir, cache_name)\n if mtime(cache_file) >= time.time() - gw2api.cache_time:\n with open(cache_file, \"r\") as fp:\n return json.load(fp)\n else:\n cache_file = None\n\n r = gw2api.session.get(gw2api.BASE_URL + path, **kwargs)\n\n if not r.ok:\n try:\n response = r.json()\n except ValueError: # pragma: no cover\n response = None\n if isinstance(response, dict) and \"text\" in response:\n r.reason = response[\"text\"]\n\n r.raise_for_status()\n data = r.json()\n\n if cache_file:\n with open(cache_file, \"w\") as fp:\n json.dump(data, fp, indent=2)\n\n return data\n" ]
from .util import get_cached __all__ = ("continents", "map_names", "maps", "map_floor") def continents(): """This resource returns static information about the continents used with the map_floor resource. The response is a dictionary where the key is the continent id, and the value is a dictionary containing the following properties: name (string): The name of the continent. continent_dims (dimension): The width and height dimensions of the continent. min_zoom (number): The minimal zoom level for use with the map tile service. max_zoom (number): The maximum zoom level for use with the map tile service. floors (list): A list of floors available for this continent. *Note: There are only two continents, Tyria and Mists.* """ return get_cached("continents.json").get("continents") def maps(map_id=None, lang="en"): """This resource returns details about maps in the game, including details about floor and translation data on how to translate between world coordinates and map coordinates. :param map_id: Only list this map. :param lang: Show localized texts in the specified language. The response is a dictionary where the key is the map id and the value is a dictionary containing the following properties: map_name (string) The map name. min_level (number) The minimal level of this map. max_level (number) The maximum level of this map. default_floor (number) The default floor of this map. floors (list) A list of available floors for this map. region_id (number) The id of the region this map belongs to. region_name (string) The name of the region this map belongs to. continent_id (number) The id of the continent this map belongs to. continent_name (string) The name of the continent this map belongs to. map_rect (rect) The dimensions of the map. continent_rect (rect) The dimensions of the map within the continent coordinate system. If a map_id is given, only the values for that map are returned. """ if map_id: cache_name = "maps.%s.%s.json" % (map_id, lang) params = {"map_id": map_id, "lang": lang} else: cache_name = "maps.%s.json" % lang params = {"lang": lang} data = get_cached("maps.json", cache_name, params=params).get("maps") return data.get(str(map_id)) if map_id else data def map_floor(continent_id, floor, lang="en"): """This resource returns details about a map floor, used to populate a world map. All coordinates are map coordinates. The returned data only contains static content. Dynamic content, such as vendors, is not currently available. :param continent_id: The continent. :param floor: The map floor. :param lang: Show localized texts in the specified language. The response is an object with the following properties: texture_dims (dimension) The dimensions of the texture. clamped_view (rect) If present, it represents a rectangle of downloadable textures. Every tile coordinate outside this rectangle is not available on the tile server. regions (object) A mapping from region id to an object. Each region object contains the following properties: name (string) The region name. label_coord (coordinate) The coordinates of the region label. maps (object) A mapping from the map id to an object. Each map object contains the following properties: name (string) The map name. min_level (number) The minimum level of the map. max_level (number) The maximum level of the map. default_floor (number) The default floor of the map. map_rect (rect) The dimensions of the map. continent_rect (rect) The dimensions of the map within the continent coordinate system. points_of_interest (list) A list of points of interest (landmarks, waypoints and vistas) Each points of interest object contains the following properties: poi_id (number) The point of interest id. name (string) The name of the point of interest. type (string) The type. This can be either "landmark" for actual points of interest, "waypoint" for waypoints, or "vista" for vistas. floor (number) The floor of this object. coord (coordinate) The coordinates of this object. tasks (list) A list of renown hearts. Each task object contains the following properties: task_id (number) The renown heart id. objective (string) The objective or name of the heart. level (number) The level of the heart. coord (coordinate) The coordinates where it takes place. skill_challenges (list) A list of skill challenges. Each skill challenge object contains the following properties: coord (coordinate) The coordinates of this skill challenge. sectors (list) A list of areas within the map. Each sector object contains the following properties: sector_id (number) The area id. name (string) The name of the area. level (number) The level of the area. coord (coordinate) The coordinates of this area (this is usually the center position). Special types: Dimension properties are two-element lists of width and height. Coordinate properties are two-element lists of the x and y position. Rect properties are two-element lists of coordinates of the upper-left and lower-right coordinates. """ cache_name = "map_floor.%s-%s.%s.json" % (continent_id, floor, lang) params = {"continent_id": continent_id, "floor": floor, "lang": lang} return get_cached("map_floor.json", cache_name, params=params)
hackedd/gw2api
gw2api/map.py
maps
python
def maps(map_id=None, lang="en"): if map_id: cache_name = "maps.%s.%s.json" % (map_id, lang) params = {"map_id": map_id, "lang": lang} else: cache_name = "maps.%s.json" % lang params = {"lang": lang} data = get_cached("maps.json", cache_name, params=params).get("maps") return data.get(str(map_id)) if map_id else data
This resource returns details about maps in the game, including details about floor and translation data on how to translate between world coordinates and map coordinates. :param map_id: Only list this map. :param lang: Show localized texts in the specified language. The response is a dictionary where the key is the map id and the value is a dictionary containing the following properties: map_name (string) The map name. min_level (number) The minimal level of this map. max_level (number) The maximum level of this map. default_floor (number) The default floor of this map. floors (list) A list of available floors for this map. region_id (number) The id of the region this map belongs to. region_name (string) The name of the region this map belongs to. continent_id (number) The id of the continent this map belongs to. continent_name (string) The name of the continent this map belongs to. map_rect (rect) The dimensions of the map. continent_rect (rect) The dimensions of the map within the continent coordinate system. If a map_id is given, only the values for that map are returned.
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/map.py#L50-L105
[ "def get_cached(path, cache_name=None, **kwargs):\n \"\"\"Request a resource form the API, first checking if there is a cached\n response available. Returns the parsed JSON data.\n \"\"\"\n if gw2api.cache_dir and gw2api.cache_time and cache_name is not False:\n if cache_name is None:\n cache_name = path\n cache_file = os.path.join(gw2api.cache_dir, cache_name)\n if mtime(cache_file) >= time.time() - gw2api.cache_time:\n with open(cache_file, \"r\") as fp:\n return json.load(fp)\n else:\n cache_file = None\n\n r = gw2api.session.get(gw2api.BASE_URL + path, **kwargs)\n\n if not r.ok:\n try:\n response = r.json()\n except ValueError: # pragma: no cover\n response = None\n if isinstance(response, dict) and \"text\" in response:\n r.reason = response[\"text\"]\n\n r.raise_for_status()\n data = r.json()\n\n if cache_file:\n with open(cache_file, \"w\") as fp:\n json.dump(data, fp, indent=2)\n\n return data\n" ]
from .util import get_cached __all__ = ("continents", "map_names", "maps", "map_floor") def continents(): """This resource returns static information about the continents used with the map_floor resource. The response is a dictionary where the key is the continent id, and the value is a dictionary containing the following properties: name (string): The name of the continent. continent_dims (dimension): The width and height dimensions of the continent. min_zoom (number): The minimal zoom level for use with the map tile service. max_zoom (number): The maximum zoom level for use with the map tile service. floors (list): A list of floors available for this continent. *Note: There are only two continents, Tyria and Mists.* """ return get_cached("continents.json").get("continents") def map_names(lang="en"): """This resource returns an dictionary of the localized map names for the specified language. Only maps with events are listed - if you need a list of all maps, use ``maps.json`` instead. :param lang: The language to query the names for. :return: the response is a dictionary where the key is the map id and the value is the name of the map in the specified language. """ cache_name = "map_names.%s.json" % lang data = get_cached("map_names.json", cache_name, params=dict(lang=lang)) return dict([(item["id"], item["name"]) for item in data]) def map_floor(continent_id, floor, lang="en"): """This resource returns details about a map floor, used to populate a world map. All coordinates are map coordinates. The returned data only contains static content. Dynamic content, such as vendors, is not currently available. :param continent_id: The continent. :param floor: The map floor. :param lang: Show localized texts in the specified language. The response is an object with the following properties: texture_dims (dimension) The dimensions of the texture. clamped_view (rect) If present, it represents a rectangle of downloadable textures. Every tile coordinate outside this rectangle is not available on the tile server. regions (object) A mapping from region id to an object. Each region object contains the following properties: name (string) The region name. label_coord (coordinate) The coordinates of the region label. maps (object) A mapping from the map id to an object. Each map object contains the following properties: name (string) The map name. min_level (number) The minimum level of the map. max_level (number) The maximum level of the map. default_floor (number) The default floor of the map. map_rect (rect) The dimensions of the map. continent_rect (rect) The dimensions of the map within the continent coordinate system. points_of_interest (list) A list of points of interest (landmarks, waypoints and vistas) Each points of interest object contains the following properties: poi_id (number) The point of interest id. name (string) The name of the point of interest. type (string) The type. This can be either "landmark" for actual points of interest, "waypoint" for waypoints, or "vista" for vistas. floor (number) The floor of this object. coord (coordinate) The coordinates of this object. tasks (list) A list of renown hearts. Each task object contains the following properties: task_id (number) The renown heart id. objective (string) The objective or name of the heart. level (number) The level of the heart. coord (coordinate) The coordinates where it takes place. skill_challenges (list) A list of skill challenges. Each skill challenge object contains the following properties: coord (coordinate) The coordinates of this skill challenge. sectors (list) A list of areas within the map. Each sector object contains the following properties: sector_id (number) The area id. name (string) The name of the area. level (number) The level of the area. coord (coordinate) The coordinates of this area (this is usually the center position). Special types: Dimension properties are two-element lists of width and height. Coordinate properties are two-element lists of the x and y position. Rect properties are two-element lists of coordinates of the upper-left and lower-right coordinates. """ cache_name = "map_floor.%s-%s.%s.json" % (continent_id, floor, lang) params = {"continent_id": continent_id, "floor": floor, "lang": lang} return get_cached("map_floor.json", cache_name, params=params)
hackedd/gw2api
gw2api/map.py
map_floor
python
def map_floor(continent_id, floor, lang="en"): cache_name = "map_floor.%s-%s.%s.json" % (continent_id, floor, lang) params = {"continent_id": continent_id, "floor": floor, "lang": lang} return get_cached("map_floor.json", cache_name, params=params)
This resource returns details about a map floor, used to populate a world map. All coordinates are map coordinates. The returned data only contains static content. Dynamic content, such as vendors, is not currently available. :param continent_id: The continent. :param floor: The map floor. :param lang: Show localized texts in the specified language. The response is an object with the following properties: texture_dims (dimension) The dimensions of the texture. clamped_view (rect) If present, it represents a rectangle of downloadable textures. Every tile coordinate outside this rectangle is not available on the tile server. regions (object) A mapping from region id to an object. Each region object contains the following properties: name (string) The region name. label_coord (coordinate) The coordinates of the region label. maps (object) A mapping from the map id to an object. Each map object contains the following properties: name (string) The map name. min_level (number) The minimum level of the map. max_level (number) The maximum level of the map. default_floor (number) The default floor of the map. map_rect (rect) The dimensions of the map. continent_rect (rect) The dimensions of the map within the continent coordinate system. points_of_interest (list) A list of points of interest (landmarks, waypoints and vistas) Each points of interest object contains the following properties: poi_id (number) The point of interest id. name (string) The name of the point of interest. type (string) The type. This can be either "landmark" for actual points of interest, "waypoint" for waypoints, or "vista" for vistas. floor (number) The floor of this object. coord (coordinate) The coordinates of this object. tasks (list) A list of renown hearts. Each task object contains the following properties: task_id (number) The renown heart id. objective (string) The objective or name of the heart. level (number) The level of the heart. coord (coordinate) The coordinates where it takes place. skill_challenges (list) A list of skill challenges. Each skill challenge object contains the following properties: coord (coordinate) The coordinates of this skill challenge. sectors (list) A list of areas within the map. Each sector object contains the following properties: sector_id (number) The area id. name (string) The name of the area. level (number) The level of the area. coord (coordinate) The coordinates of this area (this is usually the center position). Special types: Dimension properties are two-element lists of width and height. Coordinate properties are two-element lists of the x and y position. Rect properties are two-element lists of coordinates of the upper-left and lower-right coordinates.
train
https://github.com/hackedd/gw2api/blob/5543a78e6e3ed0573b7e84c142c44004b4779eac/gw2api/map.py#L108-L236
[ "def get_cached(path, cache_name=None, **kwargs):\n \"\"\"Request a resource form the API, first checking if there is a cached\n response available. Returns the parsed JSON data.\n \"\"\"\n if gw2api.cache_dir and gw2api.cache_time and cache_name is not False:\n if cache_name is None:\n cache_name = path\n cache_file = os.path.join(gw2api.cache_dir, cache_name)\n if mtime(cache_file) >= time.time() - gw2api.cache_time:\n with open(cache_file, \"r\") as fp:\n return json.load(fp)\n else:\n cache_file = None\n\n r = gw2api.session.get(gw2api.BASE_URL + path, **kwargs)\n\n if not r.ok:\n try:\n response = r.json()\n except ValueError: # pragma: no cover\n response = None\n if isinstance(response, dict) and \"text\" in response:\n r.reason = response[\"text\"]\n\n r.raise_for_status()\n data = r.json()\n\n if cache_file:\n with open(cache_file, \"w\") as fp:\n json.dump(data, fp, indent=2)\n\n return data\n" ]
from .util import get_cached __all__ = ("continents", "map_names", "maps", "map_floor") def continents(): """This resource returns static information about the continents used with the map_floor resource. The response is a dictionary where the key is the continent id, and the value is a dictionary containing the following properties: name (string): The name of the continent. continent_dims (dimension): The width and height dimensions of the continent. min_zoom (number): The minimal zoom level for use with the map tile service. max_zoom (number): The maximum zoom level for use with the map tile service. floors (list): A list of floors available for this continent. *Note: There are only two continents, Tyria and Mists.* """ return get_cached("continents.json").get("continents") def map_names(lang="en"): """This resource returns an dictionary of the localized map names for the specified language. Only maps with events are listed - if you need a list of all maps, use ``maps.json`` instead. :param lang: The language to query the names for. :return: the response is a dictionary where the key is the map id and the value is the name of the map in the specified language. """ cache_name = "map_names.%s.json" % lang data = get_cached("map_names.json", cache_name, params=dict(lang=lang)) return dict([(item["id"], item["name"]) for item in data]) def maps(map_id=None, lang="en"): """This resource returns details about maps in the game, including details about floor and translation data on how to translate between world coordinates and map coordinates. :param map_id: Only list this map. :param lang: Show localized texts in the specified language. The response is a dictionary where the key is the map id and the value is a dictionary containing the following properties: map_name (string) The map name. min_level (number) The minimal level of this map. max_level (number) The maximum level of this map. default_floor (number) The default floor of this map. floors (list) A list of available floors for this map. region_id (number) The id of the region this map belongs to. region_name (string) The name of the region this map belongs to. continent_id (number) The id of the continent this map belongs to. continent_name (string) The name of the continent this map belongs to. map_rect (rect) The dimensions of the map. continent_rect (rect) The dimensions of the map within the continent coordinate system. If a map_id is given, only the values for that map are returned. """ if map_id: cache_name = "maps.%s.%s.json" % (map_id, lang) params = {"map_id": map_id, "lang": lang} else: cache_name = "maps.%s.json" % lang params = {"lang": lang} data = get_cached("maps.json", cache_name, params=params).get("maps") return data.get(str(map_id)) if map_id else data
NickMonzillo/SmartCloud
SmartCloud/wordplay.py
separate
python
def separate(text): '''Takes text and separates it into a list of words''' alphabet = 'abcdefghijklmnopqrstuvwxyz' words = text.split() standardwords = [] for word in words: newstr = '' for char in word: if char in alphabet or char in alphabet.upper(): newstr += char if newstr != '': standardwords.append(newstr) return map(lambda x: x.lower(),standardwords)
Takes text and separates it into a list of words
train
https://github.com/NickMonzillo/SmartCloud/blob/481d1ef428427b452a8a787999c1d4a8868a3824/SmartCloud/wordplay.py#L1-L13
null
def read_file(filename): '''Reads in a .txt file.''' with open(filename,'r') as f: content = f.read() return content def eliminate_repeats(text): '''Returns a list of words that occur in the text. Eliminates stopwords.''' bannedwords = read_file('stopwords.txt') alphabet = 'abcdefghijklmnopqrstuvwxyz' words = text.split() standardwords = [] for word in words: newstr = '' for char in word: if char in alphabet or char in alphabet.upper(): newstr += char if newstr not in standardwords and newstr != '' and newstr not in bannedwords: standardwords.append(newstr) return map(lambda x: x.lower(),standardwords) def wordcount(text): '''Returns the count of the words in a file.''' bannedwords = read_file('stopwords.txt') wordcount = {} separated = separate(text) for word in separated: if word not in bannedwords: if not wordcount.has_key(word): wordcount[word] = 1 else: wordcount[word] += 1 return wordcount def tuplecount(text): '''Changes a dictionary into a list of tuples.''' worddict = wordcount(text) countlist = [] for key in worddict.keys(): countlist.append((key,worddict[key])) countlist = list(reversed(sorted(countlist,key = lambda x: x[1]))) return countlist
NickMonzillo/SmartCloud
SmartCloud/wordplay.py
eliminate_repeats
python
def eliminate_repeats(text): '''Returns a list of words that occur in the text. Eliminates stopwords.''' bannedwords = read_file('stopwords.txt') alphabet = 'abcdefghijklmnopqrstuvwxyz' words = text.split() standardwords = [] for word in words: newstr = '' for char in word: if char in alphabet or char in alphabet.upper(): newstr += char if newstr not in standardwords and newstr != '' and newstr not in bannedwords: standardwords.append(newstr) return map(lambda x: x.lower(),standardwords)
Returns a list of words that occur in the text. Eliminates stopwords.
train
https://github.com/NickMonzillo/SmartCloud/blob/481d1ef428427b452a8a787999c1d4a8868a3824/SmartCloud/wordplay.py#L21-L36
[ "def read_file(filename):\n '''Reads in a .txt file.'''\n with open(filename,'r') as f:\n content = f.read()\n return content\n" ]
def separate(text): '''Takes text and separates it into a list of words''' alphabet = 'abcdefghijklmnopqrstuvwxyz' words = text.split() standardwords = [] for word in words: newstr = '' for char in word: if char in alphabet or char in alphabet.upper(): newstr += char if newstr != '': standardwords.append(newstr) return map(lambda x: x.lower(),standardwords) def read_file(filename): '''Reads in a .txt file.''' with open(filename,'r') as f: content = f.read() return content def eliminate_repeats(text): '''Returns a list of words that occur in the text. Eliminates stopwords.''' bannedwords = read_file('stopwords.txt') alphabet = 'abcdefghijklmnopqrstuvwxyz' words = text.split() standardwords = [] for word in words: newstr = '' for char in word: if char in alphabet or char in alphabet.upper(): newstr += char if newstr not in standardwords and newstr != '' and newstr not in bannedwords: standardwords.append(newstr) return map(lambda x: x.lower(),standardwords) def wordcount(text): '''Returns the count of the words in a file.''' bannedwords = read_file('stopwords.txt') wordcount = {} separated = separate(text) for word in separated: if word not in bannedwords: if not wordcount.has_key(word): wordcount[word] = 1 else: wordcount[word] += 1 return wordcount def tuplecount(text): '''Changes a dictionary into a list of tuples.''' worddict = wordcount(text) countlist = [] for key in worddict.keys(): countlist.append((key,worddict[key])) countlist = list(reversed(sorted(countlist,key = lambda x: x[1]))) return countlist
NickMonzillo/SmartCloud
SmartCloud/wordplay.py
wordcount
python
def wordcount(text): '''Returns the count of the words in a file.''' bannedwords = read_file('stopwords.txt') wordcount = {} separated = separate(text) for word in separated: if word not in bannedwords: if not wordcount.has_key(word): wordcount[word] = 1 else: wordcount[word] += 1 return wordcount
Returns the count of the words in a file.
train
https://github.com/NickMonzillo/SmartCloud/blob/481d1ef428427b452a8a787999c1d4a8868a3824/SmartCloud/wordplay.py#L38-L49
[ "def separate(text):\n '''Takes text and separates it into a list of words'''\n alphabet = 'abcdefghijklmnopqrstuvwxyz'\n words = text.split()\n standardwords = []\n for word in words:\n newstr = ''\n for char in word:\n if char in alphabet or char in alphabet.upper():\n newstr += char\n if newstr != '':\n standardwords.append(newstr)\n return map(lambda x: x.lower(),standardwords)\n", "def read_file(filename):\n '''Reads in a .txt file.'''\n with open(filename,'r') as f:\n content = f.read()\n return content\n" ]
def separate(text): '''Takes text and separates it into a list of words''' alphabet = 'abcdefghijklmnopqrstuvwxyz' words = text.split() standardwords = [] for word in words: newstr = '' for char in word: if char in alphabet or char in alphabet.upper(): newstr += char if newstr != '': standardwords.append(newstr) return map(lambda x: x.lower(),standardwords) def read_file(filename): '''Reads in a .txt file.''' with open(filename,'r') as f: content = f.read() return content def eliminate_repeats(text): '''Returns a list of words that occur in the text. Eliminates stopwords.''' bannedwords = read_file('stopwords.txt') alphabet = 'abcdefghijklmnopqrstuvwxyz' words = text.split() standardwords = [] for word in words: newstr = '' for char in word: if char in alphabet or char in alphabet.upper(): newstr += char if newstr not in standardwords and newstr != '' and newstr not in bannedwords: standardwords.append(newstr) return map(lambda x: x.lower(),standardwords) def tuplecount(text): '''Changes a dictionary into a list of tuples.''' worddict = wordcount(text) countlist = [] for key in worddict.keys(): countlist.append((key,worddict[key])) countlist = list(reversed(sorted(countlist,key = lambda x: x[1]))) return countlist
NickMonzillo/SmartCloud
SmartCloud/wordplay.py
tuplecount
python
def tuplecount(text): '''Changes a dictionary into a list of tuples.''' worddict = wordcount(text) countlist = [] for key in worddict.keys(): countlist.append((key,worddict[key])) countlist = list(reversed(sorted(countlist,key = lambda x: x[1]))) return countlist
Changes a dictionary into a list of tuples.
train
https://github.com/NickMonzillo/SmartCloud/blob/481d1ef428427b452a8a787999c1d4a8868a3824/SmartCloud/wordplay.py#L51-L58
[ "def wordcount(text):\n '''Returns the count of the words in a file.'''\n bannedwords = read_file('stopwords.txt')\n wordcount = {}\n separated = separate(text)\n for word in separated:\n if word not in bannedwords:\n if not wordcount.has_key(word):\n wordcount[word] = 1\n else:\n wordcount[word] += 1\n return wordcount\n" ]
def separate(text): '''Takes text and separates it into a list of words''' alphabet = 'abcdefghijklmnopqrstuvwxyz' words = text.split() standardwords = [] for word in words: newstr = '' for char in word: if char in alphabet or char in alphabet.upper(): newstr += char if newstr != '': standardwords.append(newstr) return map(lambda x: x.lower(),standardwords) def read_file(filename): '''Reads in a .txt file.''' with open(filename,'r') as f: content = f.read() return content def eliminate_repeats(text): '''Returns a list of words that occur in the text. Eliminates stopwords.''' bannedwords = read_file('stopwords.txt') alphabet = 'abcdefghijklmnopqrstuvwxyz' words = text.split() standardwords = [] for word in words: newstr = '' for char in word: if char in alphabet or char in alphabet.upper(): newstr += char if newstr not in standardwords and newstr != '' and newstr not in bannedwords: standardwords.append(newstr) return map(lambda x: x.lower(),standardwords) def wordcount(text): '''Returns the count of the words in a file.''' bannedwords = read_file('stopwords.txt') wordcount = {} separated = separate(text) for word in separated: if word not in bannedwords: if not wordcount.has_key(word): wordcount[word] = 1 else: wordcount[word] += 1 return wordcount
NickMonzillo/SmartCloud
SmartCloud/__init__.py
Cloud.render_word
python
def render_word(self,word,size,color): '''Creates a surface that contains a word.''' pygame.font.init() font = pygame.font.Font(None,size) self.rendered_word = font.render(word,0,color) self.word_size = font.size(word)
Creates a surface that contains a word.
train
https://github.com/NickMonzillo/SmartCloud/blob/481d1ef428427b452a8a787999c1d4a8868a3824/SmartCloud/__init__.py#L16-L21
null
class Cloud(object): def __init__(self,width=500,height=500): pygame.init() pygame.font.init() self.width = width self.height = height self.cloud = pygame.Surface((width,height)) self.used_pos = [] def plot_word(self,position): '''Blits a rendered word on to the main display surface''' posrectangle = pygame.Rect(position,self.word_size) self.used_pos.append(posrectangle) self.cloud.blit(self.rendered_word,position) def collides(self,position,size): '''Returns True if the word collides with another plotted word.''' word_rect = pygame.Rect(position,self.word_size) if word_rect.collidelistall(self.used_pos) == []: return False else: return True def expand(self,delta_width,delta_height): '''Makes the cloud surface bigger. Maintains all word positions.''' temp_surface = pygame.Surface((self.width + delta_width,self.height + delta_height)) (self.width,self.height) = (self.width + delta_width, self.height + delta_height) temp_surface.blit(self.cloud,(0,0)) self.cloud = temp_surface def smart_cloud(self,input,max_text_size=72,min_text_size=12,exclude_words = True): '''Creates a word cloud using the input. Input can be a file, directory, or text. Set exclude_words to true if you want to eliminate words that only occur once.''' self.exclude_words = exclude_words if isdir(input): self.directory_cloud(input,max_text_size,min_text_size) elif isfile(input): text = read_file(input) self.text_cloud(text,max_text_size,min_text_size) elif isinstance(input, basestring): self.text_cloud(input,max_text_size,min_text_size) else: print 'Input type not supported.' print 'Supported types: String, Directory, .txt file' def directory_cloud(self,directory,max_text_size=72,min_text_size=12,expand_width=50,expand_height=50,max_count=100000): '''Creates a word cloud using files from a directory. The color of the words correspond to the amount of documents the word occurs in.''' worddict = assign_fonts(tuplecount(read_dir(directory)),max_text_size,min_text_size,self.exclude_words) sorted_worddict = list(reversed(sorted(worddict.keys(), key=lambda x: worddict[x]))) colordict = assign_colors(dir_freq(directory)) num_words = 0 for word in sorted_worddict: self.render_word(word,worddict[word],colordict[word]) if self.width < self.word_size[0]: #If the word is bigger than the surface, expand the surface. self.expand(self.word_size[0]-self.width,0) elif self.height < self.word_size[1]: self.expand(0,self.word_size[1]-self.height) position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] #the initial position is determined loopcount = 0 while self.collides(position,self.word_size): if loopcount > max_count: #If it can't find a position for the word, create a bigger cloud. self.expand(expand_width,expand_height) loopcount = 0 position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] loopcount += 1 self.plot_word(position) num_words += 1 def text_cloud(self,text,max_text_size=72,min_text_size=12,expand_width=50,expand_height=50,max_count=100000): '''Creates a word cloud using plain text.''' worddict = assign_fonts(tuplecount(text),max_text_size,min_text_size,self.exclude_words) sorted_worddict = list(reversed(sorted(worddict.keys(), key=lambda x: worddict[x]))) for word in sorted_worddict: self.render_word(word,worddict[word],(randint(0,255),randint(0,255),randint(0,255))) if self.width < self.word_size[0]: #If the word is bigger than the surface, expand the surface. self.expand(self.word_size[0]-self.width,0) elif self.height < self.word_size[1]: self.expand(0,self.word_size[1]-self.height) position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] loopcount = 0 while self.collides(position,self.word_size): if loopcount > max_count: #If it can't find a position for the word, expand the cloud. self.expand(expand_width,expand_height) loopcount = 0 position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] loopcount += 1 self.plot_word(position) def display(self): '''Displays the word cloud to the screen.''' pygame.init() self.display = pygame.display.set_mode((self.width,self.height)) self.display.blit(self.cloud,(0,0)) pygame.display.update() while True: for event in pygame.event.get(): if event.type == pygame.QUIT: pygame.quit() return def save(self,filename): '''Saves the cloud to a file.''' pygame.image.save(self.cloud,filename)
NickMonzillo/SmartCloud
SmartCloud/__init__.py
Cloud.plot_word
python
def plot_word(self,position): '''Blits a rendered word on to the main display surface''' posrectangle = pygame.Rect(position,self.word_size) self.used_pos.append(posrectangle) self.cloud.blit(self.rendered_word,position)
Blits a rendered word on to the main display surface
train
https://github.com/NickMonzillo/SmartCloud/blob/481d1ef428427b452a8a787999c1d4a8868a3824/SmartCloud/__init__.py#L23-L27
null
class Cloud(object): def __init__(self,width=500,height=500): pygame.init() pygame.font.init() self.width = width self.height = height self.cloud = pygame.Surface((width,height)) self.used_pos = [] def render_word(self,word,size,color): '''Creates a surface that contains a word.''' pygame.font.init() font = pygame.font.Font(None,size) self.rendered_word = font.render(word,0,color) self.word_size = font.size(word) def collides(self,position,size): '''Returns True if the word collides with another plotted word.''' word_rect = pygame.Rect(position,self.word_size) if word_rect.collidelistall(self.used_pos) == []: return False else: return True def expand(self,delta_width,delta_height): '''Makes the cloud surface bigger. Maintains all word positions.''' temp_surface = pygame.Surface((self.width + delta_width,self.height + delta_height)) (self.width,self.height) = (self.width + delta_width, self.height + delta_height) temp_surface.blit(self.cloud,(0,0)) self.cloud = temp_surface def smart_cloud(self,input,max_text_size=72,min_text_size=12,exclude_words = True): '''Creates a word cloud using the input. Input can be a file, directory, or text. Set exclude_words to true if you want to eliminate words that only occur once.''' self.exclude_words = exclude_words if isdir(input): self.directory_cloud(input,max_text_size,min_text_size) elif isfile(input): text = read_file(input) self.text_cloud(text,max_text_size,min_text_size) elif isinstance(input, basestring): self.text_cloud(input,max_text_size,min_text_size) else: print 'Input type not supported.' print 'Supported types: String, Directory, .txt file' def directory_cloud(self,directory,max_text_size=72,min_text_size=12,expand_width=50,expand_height=50,max_count=100000): '''Creates a word cloud using files from a directory. The color of the words correspond to the amount of documents the word occurs in.''' worddict = assign_fonts(tuplecount(read_dir(directory)),max_text_size,min_text_size,self.exclude_words) sorted_worddict = list(reversed(sorted(worddict.keys(), key=lambda x: worddict[x]))) colordict = assign_colors(dir_freq(directory)) num_words = 0 for word in sorted_worddict: self.render_word(word,worddict[word],colordict[word]) if self.width < self.word_size[0]: #If the word is bigger than the surface, expand the surface. self.expand(self.word_size[0]-self.width,0) elif self.height < self.word_size[1]: self.expand(0,self.word_size[1]-self.height) position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] #the initial position is determined loopcount = 0 while self.collides(position,self.word_size): if loopcount > max_count: #If it can't find a position for the word, create a bigger cloud. self.expand(expand_width,expand_height) loopcount = 0 position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] loopcount += 1 self.plot_word(position) num_words += 1 def text_cloud(self,text,max_text_size=72,min_text_size=12,expand_width=50,expand_height=50,max_count=100000): '''Creates a word cloud using plain text.''' worddict = assign_fonts(tuplecount(text),max_text_size,min_text_size,self.exclude_words) sorted_worddict = list(reversed(sorted(worddict.keys(), key=lambda x: worddict[x]))) for word in sorted_worddict: self.render_word(word,worddict[word],(randint(0,255),randint(0,255),randint(0,255))) if self.width < self.word_size[0]: #If the word is bigger than the surface, expand the surface. self.expand(self.word_size[0]-self.width,0) elif self.height < self.word_size[1]: self.expand(0,self.word_size[1]-self.height) position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] loopcount = 0 while self.collides(position,self.word_size): if loopcount > max_count: #If it can't find a position for the word, expand the cloud. self.expand(expand_width,expand_height) loopcount = 0 position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] loopcount += 1 self.plot_word(position) def display(self): '''Displays the word cloud to the screen.''' pygame.init() self.display = pygame.display.set_mode((self.width,self.height)) self.display.blit(self.cloud,(0,0)) pygame.display.update() while True: for event in pygame.event.get(): if event.type == pygame.QUIT: pygame.quit() return def save(self,filename): '''Saves the cloud to a file.''' pygame.image.save(self.cloud,filename)
NickMonzillo/SmartCloud
SmartCloud/__init__.py
Cloud.collides
python
def collides(self,position,size): '''Returns True if the word collides with another plotted word.''' word_rect = pygame.Rect(position,self.word_size) if word_rect.collidelistall(self.used_pos) == []: return False else: return True
Returns True if the word collides with another plotted word.
train
https://github.com/NickMonzillo/SmartCloud/blob/481d1ef428427b452a8a787999c1d4a8868a3824/SmartCloud/__init__.py#L29-L35
null
class Cloud(object): def __init__(self,width=500,height=500): pygame.init() pygame.font.init() self.width = width self.height = height self.cloud = pygame.Surface((width,height)) self.used_pos = [] def render_word(self,word,size,color): '''Creates a surface that contains a word.''' pygame.font.init() font = pygame.font.Font(None,size) self.rendered_word = font.render(word,0,color) self.word_size = font.size(word) def plot_word(self,position): '''Blits a rendered word on to the main display surface''' posrectangle = pygame.Rect(position,self.word_size) self.used_pos.append(posrectangle) self.cloud.blit(self.rendered_word,position) def expand(self,delta_width,delta_height): '''Makes the cloud surface bigger. Maintains all word positions.''' temp_surface = pygame.Surface((self.width + delta_width,self.height + delta_height)) (self.width,self.height) = (self.width + delta_width, self.height + delta_height) temp_surface.blit(self.cloud,(0,0)) self.cloud = temp_surface def smart_cloud(self,input,max_text_size=72,min_text_size=12,exclude_words = True): '''Creates a word cloud using the input. Input can be a file, directory, or text. Set exclude_words to true if you want to eliminate words that only occur once.''' self.exclude_words = exclude_words if isdir(input): self.directory_cloud(input,max_text_size,min_text_size) elif isfile(input): text = read_file(input) self.text_cloud(text,max_text_size,min_text_size) elif isinstance(input, basestring): self.text_cloud(input,max_text_size,min_text_size) else: print 'Input type not supported.' print 'Supported types: String, Directory, .txt file' def directory_cloud(self,directory,max_text_size=72,min_text_size=12,expand_width=50,expand_height=50,max_count=100000): '''Creates a word cloud using files from a directory. The color of the words correspond to the amount of documents the word occurs in.''' worddict = assign_fonts(tuplecount(read_dir(directory)),max_text_size,min_text_size,self.exclude_words) sorted_worddict = list(reversed(sorted(worddict.keys(), key=lambda x: worddict[x]))) colordict = assign_colors(dir_freq(directory)) num_words = 0 for word in sorted_worddict: self.render_word(word,worddict[word],colordict[word]) if self.width < self.word_size[0]: #If the word is bigger than the surface, expand the surface. self.expand(self.word_size[0]-self.width,0) elif self.height < self.word_size[1]: self.expand(0,self.word_size[1]-self.height) position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] #the initial position is determined loopcount = 0 while self.collides(position,self.word_size): if loopcount > max_count: #If it can't find a position for the word, create a bigger cloud. self.expand(expand_width,expand_height) loopcount = 0 position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] loopcount += 1 self.plot_word(position) num_words += 1 def text_cloud(self,text,max_text_size=72,min_text_size=12,expand_width=50,expand_height=50,max_count=100000): '''Creates a word cloud using plain text.''' worddict = assign_fonts(tuplecount(text),max_text_size,min_text_size,self.exclude_words) sorted_worddict = list(reversed(sorted(worddict.keys(), key=lambda x: worddict[x]))) for word in sorted_worddict: self.render_word(word,worddict[word],(randint(0,255),randint(0,255),randint(0,255))) if self.width < self.word_size[0]: #If the word is bigger than the surface, expand the surface. self.expand(self.word_size[0]-self.width,0) elif self.height < self.word_size[1]: self.expand(0,self.word_size[1]-self.height) position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] loopcount = 0 while self.collides(position,self.word_size): if loopcount > max_count: #If it can't find a position for the word, expand the cloud. self.expand(expand_width,expand_height) loopcount = 0 position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] loopcount += 1 self.plot_word(position) def display(self): '''Displays the word cloud to the screen.''' pygame.init() self.display = pygame.display.set_mode((self.width,self.height)) self.display.blit(self.cloud,(0,0)) pygame.display.update() while True: for event in pygame.event.get(): if event.type == pygame.QUIT: pygame.quit() return def save(self,filename): '''Saves the cloud to a file.''' pygame.image.save(self.cloud,filename)
NickMonzillo/SmartCloud
SmartCloud/__init__.py
Cloud.expand
python
def expand(self,delta_width,delta_height): '''Makes the cloud surface bigger. Maintains all word positions.''' temp_surface = pygame.Surface((self.width + delta_width,self.height + delta_height)) (self.width,self.height) = (self.width + delta_width, self.height + delta_height) temp_surface.blit(self.cloud,(0,0)) self.cloud = temp_surface
Makes the cloud surface bigger. Maintains all word positions.
train
https://github.com/NickMonzillo/SmartCloud/blob/481d1ef428427b452a8a787999c1d4a8868a3824/SmartCloud/__init__.py#L37-L42
null
class Cloud(object): def __init__(self,width=500,height=500): pygame.init() pygame.font.init() self.width = width self.height = height self.cloud = pygame.Surface((width,height)) self.used_pos = [] def render_word(self,word,size,color): '''Creates a surface that contains a word.''' pygame.font.init() font = pygame.font.Font(None,size) self.rendered_word = font.render(word,0,color) self.word_size = font.size(word) def plot_word(self,position): '''Blits a rendered word on to the main display surface''' posrectangle = pygame.Rect(position,self.word_size) self.used_pos.append(posrectangle) self.cloud.blit(self.rendered_word,position) def collides(self,position,size): '''Returns True if the word collides with another plotted word.''' word_rect = pygame.Rect(position,self.word_size) if word_rect.collidelistall(self.used_pos) == []: return False else: return True def smart_cloud(self,input,max_text_size=72,min_text_size=12,exclude_words = True): '''Creates a word cloud using the input. Input can be a file, directory, or text. Set exclude_words to true if you want to eliminate words that only occur once.''' self.exclude_words = exclude_words if isdir(input): self.directory_cloud(input,max_text_size,min_text_size) elif isfile(input): text = read_file(input) self.text_cloud(text,max_text_size,min_text_size) elif isinstance(input, basestring): self.text_cloud(input,max_text_size,min_text_size) else: print 'Input type not supported.' print 'Supported types: String, Directory, .txt file' def directory_cloud(self,directory,max_text_size=72,min_text_size=12,expand_width=50,expand_height=50,max_count=100000): '''Creates a word cloud using files from a directory. The color of the words correspond to the amount of documents the word occurs in.''' worddict = assign_fonts(tuplecount(read_dir(directory)),max_text_size,min_text_size,self.exclude_words) sorted_worddict = list(reversed(sorted(worddict.keys(), key=lambda x: worddict[x]))) colordict = assign_colors(dir_freq(directory)) num_words = 0 for word in sorted_worddict: self.render_word(word,worddict[word],colordict[word]) if self.width < self.word_size[0]: #If the word is bigger than the surface, expand the surface. self.expand(self.word_size[0]-self.width,0) elif self.height < self.word_size[1]: self.expand(0,self.word_size[1]-self.height) position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] #the initial position is determined loopcount = 0 while self.collides(position,self.word_size): if loopcount > max_count: #If it can't find a position for the word, create a bigger cloud. self.expand(expand_width,expand_height) loopcount = 0 position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] loopcount += 1 self.plot_word(position) num_words += 1 def text_cloud(self,text,max_text_size=72,min_text_size=12,expand_width=50,expand_height=50,max_count=100000): '''Creates a word cloud using plain text.''' worddict = assign_fonts(tuplecount(text),max_text_size,min_text_size,self.exclude_words) sorted_worddict = list(reversed(sorted(worddict.keys(), key=lambda x: worddict[x]))) for word in sorted_worddict: self.render_word(word,worddict[word],(randint(0,255),randint(0,255),randint(0,255))) if self.width < self.word_size[0]: #If the word is bigger than the surface, expand the surface. self.expand(self.word_size[0]-self.width,0) elif self.height < self.word_size[1]: self.expand(0,self.word_size[1]-self.height) position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] loopcount = 0 while self.collides(position,self.word_size): if loopcount > max_count: #If it can't find a position for the word, expand the cloud. self.expand(expand_width,expand_height) loopcount = 0 position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] loopcount += 1 self.plot_word(position) def display(self): '''Displays the word cloud to the screen.''' pygame.init() self.display = pygame.display.set_mode((self.width,self.height)) self.display.blit(self.cloud,(0,0)) pygame.display.update() while True: for event in pygame.event.get(): if event.type == pygame.QUIT: pygame.quit() return def save(self,filename): '''Saves the cloud to a file.''' pygame.image.save(self.cloud,filename)
NickMonzillo/SmartCloud
SmartCloud/__init__.py
Cloud.smart_cloud
python
def smart_cloud(self,input,max_text_size=72,min_text_size=12,exclude_words = True): '''Creates a word cloud using the input. Input can be a file, directory, or text. Set exclude_words to true if you want to eliminate words that only occur once.''' self.exclude_words = exclude_words if isdir(input): self.directory_cloud(input,max_text_size,min_text_size) elif isfile(input): text = read_file(input) self.text_cloud(text,max_text_size,min_text_size) elif isinstance(input, basestring): self.text_cloud(input,max_text_size,min_text_size) else: print 'Input type not supported.' print 'Supported types: String, Directory, .txt file'
Creates a word cloud using the input. Input can be a file, directory, or text. Set exclude_words to true if you want to eliminate words that only occur once.
train
https://github.com/NickMonzillo/SmartCloud/blob/481d1ef428427b452a8a787999c1d4a8868a3824/SmartCloud/__init__.py#L44-L58
[ "def read_file(filename):\n '''Reads in a .txt file.'''\n with open(filename,'r') as f:\n content = f.read()\n return content\n", "def directory_cloud(self,directory,max_text_size=72,min_text_size=12,expand_width=50,expand_height=50,max_count=100000):\n '''Creates a word cloud using files from a directory.\n The color of the words correspond to the amount of documents the word occurs in.'''\n worddict = assign_fonts(tuplecount(read_dir(directory)),max_text_size,min_text_size,self.exclude_words)\n sorted_worddict = list(reversed(sorted(worddict.keys(), key=lambda x: worddict[x])))\n colordict = assign_colors(dir_freq(directory))\n num_words = 0\n for word in sorted_worddict:\n self.render_word(word,worddict[word],colordict[word])\n if self.width < self.word_size[0]:\n #If the word is bigger than the surface, expand the surface.\n self.expand(self.word_size[0]-self.width,0)\n elif self.height < self.word_size[1]:\n self.expand(0,self.word_size[1]-self.height)\n position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])]\n #the initial position is determined\n loopcount = 0\n while self.collides(position,self.word_size):\n if loopcount > max_count:\n #If it can't find a position for the word, create a bigger cloud.\n self.expand(expand_width,expand_height) \n loopcount = 0\n position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])]\n loopcount += 1\n self.plot_word(position)\n num_words += 1\n", "def text_cloud(self,text,max_text_size=72,min_text_size=12,expand_width=50,expand_height=50,max_count=100000):\n '''Creates a word cloud using plain text.'''\n worddict = assign_fonts(tuplecount(text),max_text_size,min_text_size,self.exclude_words)\n sorted_worddict = list(reversed(sorted(worddict.keys(), key=lambda x: worddict[x])))\n for word in sorted_worddict:\n self.render_word(word,worddict[word],(randint(0,255),randint(0,255),randint(0,255)))\n if self.width < self.word_size[0]:\n #If the word is bigger than the surface, expand the surface.\n self.expand(self.word_size[0]-self.width,0)\n elif self.height < self.word_size[1]:\n self.expand(0,self.word_size[1]-self.height)\n position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])]\n loopcount = 0\n while self.collides(position,self.word_size):\n if loopcount > max_count:\n #If it can't find a position for the word, expand the cloud.\n self.expand(expand_width,expand_height)\n loopcount = 0\n position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])]\n loopcount += 1\n self.plot_word(position)\n" ]
class Cloud(object): def __init__(self,width=500,height=500): pygame.init() pygame.font.init() self.width = width self.height = height self.cloud = pygame.Surface((width,height)) self.used_pos = [] def render_word(self,word,size,color): '''Creates a surface that contains a word.''' pygame.font.init() font = pygame.font.Font(None,size) self.rendered_word = font.render(word,0,color) self.word_size = font.size(word) def plot_word(self,position): '''Blits a rendered word on to the main display surface''' posrectangle = pygame.Rect(position,self.word_size) self.used_pos.append(posrectangle) self.cloud.blit(self.rendered_word,position) def collides(self,position,size): '''Returns True if the word collides with another plotted word.''' word_rect = pygame.Rect(position,self.word_size) if word_rect.collidelistall(self.used_pos) == []: return False else: return True def expand(self,delta_width,delta_height): '''Makes the cloud surface bigger. Maintains all word positions.''' temp_surface = pygame.Surface((self.width + delta_width,self.height + delta_height)) (self.width,self.height) = (self.width + delta_width, self.height + delta_height) temp_surface.blit(self.cloud,(0,0)) self.cloud = temp_surface def directory_cloud(self,directory,max_text_size=72,min_text_size=12,expand_width=50,expand_height=50,max_count=100000): '''Creates a word cloud using files from a directory. The color of the words correspond to the amount of documents the word occurs in.''' worddict = assign_fonts(tuplecount(read_dir(directory)),max_text_size,min_text_size,self.exclude_words) sorted_worddict = list(reversed(sorted(worddict.keys(), key=lambda x: worddict[x]))) colordict = assign_colors(dir_freq(directory)) num_words = 0 for word in sorted_worddict: self.render_word(word,worddict[word],colordict[word]) if self.width < self.word_size[0]: #If the word is bigger than the surface, expand the surface. self.expand(self.word_size[0]-self.width,0) elif self.height < self.word_size[1]: self.expand(0,self.word_size[1]-self.height) position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] #the initial position is determined loopcount = 0 while self.collides(position,self.word_size): if loopcount > max_count: #If it can't find a position for the word, create a bigger cloud. self.expand(expand_width,expand_height) loopcount = 0 position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] loopcount += 1 self.plot_word(position) num_words += 1 def text_cloud(self,text,max_text_size=72,min_text_size=12,expand_width=50,expand_height=50,max_count=100000): '''Creates a word cloud using plain text.''' worddict = assign_fonts(tuplecount(text),max_text_size,min_text_size,self.exclude_words) sorted_worddict = list(reversed(sorted(worddict.keys(), key=lambda x: worddict[x]))) for word in sorted_worddict: self.render_word(word,worddict[word],(randint(0,255),randint(0,255),randint(0,255))) if self.width < self.word_size[0]: #If the word is bigger than the surface, expand the surface. self.expand(self.word_size[0]-self.width,0) elif self.height < self.word_size[1]: self.expand(0,self.word_size[1]-self.height) position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] loopcount = 0 while self.collides(position,self.word_size): if loopcount > max_count: #If it can't find a position for the word, expand the cloud. self.expand(expand_width,expand_height) loopcount = 0 position = [randint(0,self.width-self.word_size[0]),randint(0,self.height-self.word_size[1])] loopcount += 1 self.plot_word(position) def display(self): '''Displays the word cloud to the screen.''' pygame.init() self.display = pygame.display.set_mode((self.width,self.height)) self.display.blit(self.cloud,(0,0)) pygame.display.update() while True: for event in pygame.event.get(): if event.type == pygame.QUIT: pygame.quit() return def save(self,filename): '''Saves the cloud to a file.''' pygame.image.save(self.cloud,filename)