import re import uuid from nltk.stem.snowball import SnowballStemmer class TargetStatus: EMPTY = 'EMPTY' NON_EMPTY = 'NON_EMPTY' TRUE = "TRUE" FALSE = "FALSE" MIN = "MIN" MAX = "MAX" MIDDLE = "MIDDLE" EXAMPLE = 'EXAMPLE' class TargetType: ENUM = 'ENUM' NUM = 'NUM' ARRAY = "ARRAY" BOOL = "BOOL" STRING = "STRING" EXAMPLE = 'EXAMPLE' class Parameter: def __init__(self, name, body={}, method_path=""): self.name = name self.raw_body = body self.has_schema = False self.schema = None self.method_path = method_path self.attribute_path_dict = {} self.parameter_id = "h" + uuid.uuid4().__str__() self.parameter_names = set() self.parameter_names.add(name) self.parameter_body_tuple = {} self.parameter_body_tuple[name] = [(name, body)] # record_all_attribute self.attributes = set([name]) # target consists of (type, path, value) self.cover_targets = set() # if name == 'delete_lun_groups_request': # print() self.attribute_path_dict[name] = set([name]) self.parse(body) # extend parameter name by path resource if self.attribute_should_extend(name): self.parameter_names.remove(name) self.parameter_body_tuple[name].remove((name, body)) canonical_name = self.get_canonical_name(name) tokens = self.tokenize_method_path(method_path) for token in tokens: attribute_path = token + '_' + canonical_name self.attribute_path_dict[attribute_path] = set([name]) self.parameter_names.add(attribute_path) self.parameter_body_tuple[attribute_path] = [(attribute_path, body)] def parse(self, body={}): if body.__contains__("schema"): self.has_schema = True self.parse_schema(body["schema"]) else: if body.__contains__("name"): name = body["name"] self.parameter_names.add(name) body_tuple_list = self.parameter_body_tuple.get(name, []) body_tuple_list.append((name, body)) self.parameter_body_tuple[name] = body_tuple_list self.parse_schema(body) elif body.__contains__("properties"): self.has_schema = True self.schema = body self.parse_schema(self.schema["properties"]) else: print("Body without name") def parse_schema(self, schema={}): if isinstance(schema, dict): if schema.__contains__("properties") or schema.__contains__("allOf") or schema.__contains__('name'): queue = [("", schema)] # handle allOf case if schema.__contains__("allOf"): for item in schema["allOf"]: queue.append(("", item)) while len(queue) > 0: elem = queue.pop() prefix = elem[0] elem = elem[1] if elem.__contains__("properties"): properties = elem["properties"] # required_properties = [] # if elem.__contains__('required'): # required_properties = elem['required'] for name in properties.keys(): # if name in required_properties: # properties[name]['required'] = True # check array type is_array = False if properties[name].__contains__("type") and properties[name]["type"] == "array": is_array = True # check prefix if len(prefix) == 0: child_prefix = prefix else: child_prefix = prefix child_prefix += "." self.parameter_names.add(name) # add into attribute set attribute_path = child_prefix + name attribute_set = self.attribute_path_dict.get(name, set()) attribute_set.add(attribute_path) body_tuple_list = self.parameter_body_tuple.get(name, []) body_tuple_list.append((attribute_path, properties[name])) self.parameter_body_tuple[name] = body_tuple_list self.attributes.add(self.name + "." + attribute_path) self.attribute_path_dict[name] = attribute_set # if attribute_path == 'id': # print() # extend property name by attribute path if self.attribute_should_extend(name): self.parameter_names.remove(name) self.parameter_body_tuple[name].remove((attribute_path, properties[name])) is_first_level_attribute = False if len(prefix) != 0: prefix_tokens = [self.tokenize_attribute_path(child_prefix)] else: prefix_tokens = self.tokenize_method_path(self.method_path) is_first_level_attribute = True for prefix_token in prefix_tokens: if is_first_level_attribute: path = prefix_token + '_' + name else: path = prefix_token + '_' + name attribute_path_set = self.attribute_path_dict.get(path, set()) attribute_path_set.add(attribute_path) self.attribute_path_dict[path] = attribute_path_set self.parameter_names.add(path) body_tuple_list = self.parameter_body_tuple.get(path, []) body_tuple_list.append((path, properties[name])) self.parameter_body_tuple[path] = body_tuple_list # traverse inner properties if isinstance(properties[name], dict): if is_array: queue.append((child_prefix + name + "[0]", properties[name])) else: queue.append((child_prefix + name, properties[name])) if elem.__contains__("allOf"): for e in elem["allOf"]: queue.append((prefix, e)) if elem.__contains__("items"): if not prefix.endswith('[0]'): prefix += '[0]' self.parameter_body_tuple[prefix] = [ (prefix, elem["items"])] self.attribute_path_dict[prefix] = [prefix] self.parameter_names.add(prefix) queue.append((prefix, elem["items"])) targets = self.cover_targets # try: # int(self.name + 1) # full_path = prefix # except Exception as ex: # if len(prefix) == 0: # full_path = self.name # else: # full_path = self.name + "." + prefix # self.attributes.add(full_path) if len(prefix) == 0: continue full_path = prefix # handle enum targets if elem.__contains__('enum'): for enum in elem["enum"]: tar = (full_path, TargetType.ENUM, enum) targets.add(tar) elif elem.__contains__("type"): element_type = elem['type'] # handle array type if element_type == 'array': targets.add((full_path.replace('[0]', ''), TargetType.ARRAY, TargetStatus.EMPTY)) targets.add((full_path.replace('[0]', ''), TargetType.ARRAY, TargetStatus.NON_EMPTY)) # handle string type elif element_type == 'string': if elem.__contains__('example'): tar = (full_path, TargetType.EXAMPLE, TargetStatus.EXAMPLE) targets.add(tar) # if elem.__contains__("minLength"): # targets.add((full_path, TargetType.STRING, TargetStatus.MIN)) # if elem.__contains__("maxLength"): # targets.add((full_path, TargetType.STRING, TargetStatus.MAX)) # if elem.__contains__("minLength") and elem.__contains__("maxLength"): # targets.add((full_path, TargetType.STRING, TargetStatus.MIDDLE)) elif element_type == 'boolean': targets.add((full_path, TargetType.BOOL, TargetStatus.TRUE)) targets.add((full_path, TargetType.BOOL, TargetStatus.FALSE)) elif element_type == 'integer': if elem.__contains__('example'): tar = (full_path, TargetType.EXAMPLE, TargetStatus.EXAMPLE) targets.add(tar) # if elem.__contains__("minimum"): # targets.add((full_path, TargetType.NUM, TargetStatus.MIN)) # if elem.__contains__("maximum"): # targets.add((full_path, TargetType.NUM, TargetStatus.MAX)) # if elem.__contains__("minimum") and elem.__contains__("maximum"): # targets.add((full_path, TargetType.NUM, TargetStatus.MIDDLE)) elif element_type == 'number': if elem.__contains__('example'): tar = (full_path, TargetType.EXAMPLE, TargetStatus.EXAMPLE) targets.add(tar) # if elem.__contains__("minimum"): # targets.add((full_path, TargetType.NUM, TargetStatus.MIN)) # if elem.__contains__("maximum"): # targets.add((full_path, TargetType.NUM, TargetStatus.MAX)) # if elem.__contains__("minimum") and elem.__contains__("maximum"): # targets.add((full_path, TargetType.NUM, TargetStatus.MIDDLE)) # elif element_type == 'object': # pass # else: # print() # for property in elem.keys(): # item = elem[property] # if isinstance(item, dict): # queue.append((prefix, item)) else: return def get_canonical_name(self, name): stemmer = SnowballStemmer("english") stemmed_word = stemmer.stem(name) return stemmed_word def tokenize_method_path(self, method_path: str): tokens = method_path.split('/') result = [] version_result = [] has_version = False pattern = re.compile('^v[0-9]+$') for token in tokens: # filter in path parameter if '{' in token: continue # remove empty string if len(token) == 0: continue # check whether has version if pattern.match(token) is not None: has_version = True continue token = self.get_canonical_name(token) # ignore base path for service if has_version: version_result.append(token) else: result.append(token) if has_version and len(version_result) > 0: return version_result return result def tokenize_attribute_path(self, attribute_path: str): if len(attribute_path) == 0: return '' tokens = attribute_path.replace('[0]', '').split('.') result = [] for token in tokens: if len(token) == 0: continue result.append(self.get_canonical_name(token)) return result[-1] def attribute_should_extend(self, attribute_name): targets = ['id', 'name'] name = self.get_canonical_name(attribute_name) return name in targets def __str__(self): return f'({self.name}, {self.schema})'