def clean_word(self, word):
if isinstance(word, str):
return singularize(word.lower()) if singularize(word.lower()) not in self.stop_words else None
elif isinstance(word, list):
return [singularize(w.lower()) for w in word if w not in self.stop_words]
else:
raise ValueError()
def before_request():
if current_user.is_authenticated:
if request.endpoint == 'web.show_core':
if 1 == randint(1, 2):
if not current_user.bio.research_interests or not current_user.bio.research_experience:
flash(Markup(
'Your biography is not up to date. '
'<a href="{0}" class="alert-link">Update my biography</a>.'.format(
url_for('show_user', user_id=current_user.id))), category='warning')
else:
for user_survey in current_user.surveys.values():
if not user_survey.completed_on:
parent_type, parent_id = user_survey.survey.parent_type, user_survey.survey.parent_id
if parent_type and parent_id:
endpoint = 'web.show_{0}'.format(singularize(parent_type))
view_args = {'{0}_id'.format(singularize(parent_type)): parent_id}
url = url_for(endpoint, **view_args)
else:
url = url_for('show_user', user_id=current_user.id)
flash(Markup(
'You have not completed the survey. Please complete the '
'<a href="{href}#survey" class="alert-link">{title}</a>.'.format(
href=url,
title=user_survey.title
)), category='warning')
break
def put(self, id_object, resource=None, **kwargs):
resource = resource or self.resource
url = "{}/{}.json".format(resource, id_object)
data = {singularize(resource): kwargs}
resp = self.base._request(url, 'PUT', **data)
if resp.status_code != 200:
content = safe_get_json(resp)
raise RequestException(resp.status_code, content=content)
return self.class_object(**resp.json().get(singularize(resource)))
def create(self, resource=None, **kwargs):
resource = resource or self.resource
url = "{}.json".format(resource)
singular_resource = singularize(resource)
data = {
singular_resource: kwargs,
}
resp = self.base._request(url, 'POST', **data)
if resp.status_code != 201:
content = resp.json() if getattr(resp, 'json') else {}
raise RequestException(resp.status_code, content=content)
return self.class_object(**resp.json().get(singularize(resource)))
def compute_metadata_score(self, stream_info, query_str):
TYPE = 1
metadata_features = {'title': 0, 'rating': -1, 'description': 0, 'viewcount': -1, 'dislikes':-1, 'likes': -1, 'keywords': 0}
query_words = query_str.split()
#
words = stream_info['ftitle'].split('_')
for q_w in query_words:
for w in words:
w = inflection.singularize(w.lower())
if w.find(q_w) >= 0:
metadata_features['title'] += 1
break
#
metadata_features['rating'] = self.get_type_score(stream_info['rating'], 'rating', TYPE)
#
words = stream_info['description'].split()
for q_w in query_words:
for w in words:
w = inflection.singularize(w.lower())
if w.find(q_w) >= 0:
metadata_features['description'] += 1
break
#
metadata_features['viewcount'] = self.get_type_score(stream_info['viewcount'], 'viewcount', TYPE)
#
metadata_features['dislikes'] = self.get_type_score(stream_info['dislikes'], 'dislikes', TYPE)
#
metadata_features['likes'] = self.get_type_score(stream_info['likes'], 'likes', TYPE)
#
for q_w in query_words:
for w in stream_info['keywords']:
w = inflection.singularize(w.lower())
if w.find(q_w) >= 0:
metadata_features['keywords'] += 1
break
# compute score
score = 0
if metadata_features['title'] or metadata_features['description'] or metadata_features['keywords']:
score = metadata_features['title'] + metadata_features['description'] + metadata_features['keywords']
#for key in metadata_features.keys():
# score += metadata_features[key]
return score, metadata_features['viewcount']
def test_uncountable_word_is_not_greedy():
uncountable_word = "ors"
countable_word = "sponsor"
inflection.UNCOUNTABLES.add(uncountable_word)
try:
assert uncountable_word == inflection.singularize(uncountable_word)
assert uncountable_word == inflection.pluralize(uncountable_word)
assert inflection.pluralize(uncountable_word) == inflection.singularize(uncountable_word)
assert "sponsor" == inflection.singularize(countable_word)
assert "sponsors" == inflection.pluralize(countable_word)
assert "sponsor" == inflection.singularize(inflection.pluralize(countable_word))
finally:
inflection.UNCOUNTABLES.remove(uncountable_word)
def generate_model_name(name):
""" Generate model name.
:param name: String representing a field or route name.
"""
model_name = inflection.camelize(name.strip('/'))
return inflection.singularize(model_name)
def handle(self):
name = self.argument('name')
singular = inflection.singularize(inflection.tableize(name))
directory = self._get_path()
filepath = self._get_path(singular + '.py')
if os.path.exists(filepath):
raise RuntimeError('The model file already exists.')
mkdir_p(directory)
parent = os.path.join(directory, '__init__.py')
if not os.path.exists(parent):
with open(parent, 'w'):
pass
stub = self._get_stub()
stub = self._populate_stub(name, stub)
with open(filepath, 'w') as f:
f.write(stub)
self.info('Model <comment>%s</> successfully created.' % name)
if self.option('migration'):
table = inflection.tableize(name)
self.call(
'make:migration',
[
('name', 'create_%s_table' % table),
('--table', table),
('--create', True)
]
)
def gen_db_model(self, schema_name, table_name):
sql = """ SELECT column_name, column_default, data_type, column_key, is_nullable,
character_maximum_length, column_comment
FROM information_schema.columns
WHERE TABLE_SCHEMA = :schema_name AND
TABLE_NAME = :table_name
ORDER BY ORDINAL_POSITION"""
with self.create_session(self._default_db) as session:
entities = session.execute(sql, dict(schema_name=schema_name, table_name=table_name))
metas = [GenMeta(entity) for entity in entities if entity[0] != 'pkid']
model_meta = DbModelMeta()
model_meta.table_metas = metas
for meta in metas:
if model_meta.has_id is False:
model_meta.has_id = meta.column_name == 'id'
model_meta.is_string_id = meta.column_name == 'id' and \
u'varchar' == meta.data_type
if model_meta.has_created_at is False:
model_meta.has_created_at = meta.column_name == 'created_at'
if model_meta.has_updated_at is False:
model_meta.has_updated_at = meta.column_name == 'updated_at'
if model_meta.has_is_deleted is False:
model_meta.has_is_deleted = meta.column_name == 'is_deleted'
model_meta.class_name = inflection.singularize(inflection.camelize(table_name))
template_path = os.path.join(os.path.dirname(__file__))
t = Loader(template_path, **{}).load('model.tpl')
print t.generate(**model_meta.__dict__)
def _resolve_obj_from_path(self, path_str, params):
"""
:param path_str: stripped (no extra slashes)
:return: new obj
"""
obj = {}
if path_str == '0':
obj['name'] = 'jonathan'
self._set_cached_obj(path_str, '0', obj) # add proper fields to obj and assign path and id globally to it
obj.update(self._get_meta(params, ['name', 'id', 'birthday'], ['photos', 'feed', 'albums'], 'user'))
return obj
m = re.search('[0|me]/(photos|albums|feed)([0-9]+)$', path_str)
if m:
type = singularize(m.group(1))
id = m.group(2)
if int(id) < 100:
raise NoSuchPathException(path_str)
obj['type'] = type
self._set_cached_obj(path_str, str(id), obj)
if type == 'photo':
fields = ['id', 'created_time']
connections = ['comments', 'likes']
elif type == 'album':
fields = ['id', 'cover_photo']
connections = ['comments', 'photos']
elif type == 'feed':
fields = ['id', 'message', 'story']
connections = ['comments', 'attachments']
obj.update(self._get_meta(params, fields, connections, type))
return obj
raise NoSuchPathException(path_str)
def get_response_query_template(service, operation):
"""refers to definition of API in botocore, and autogenerates template
Assume that response format is xml when protocol is query
You can see example of elbv2 from link below.
https://github.com/boto/botocore/blob/develop/botocore/data/elbv2/2015-12-01/service-2.json
"""
client = boto3.client(service)
aws_operation_name = to_upper_camel_case(operation)
op_model = client._service_model.operation_model(aws_operation_name)
result_wrapper = op_model.output_shape.serialization['resultWrapper']
response_wrapper = result_wrapper.replace('Result', 'Response')
metadata = op_model.metadata
xml_namespace = metadata['xmlNamespace']
# build xml tree
t_root = etree.Element(response_wrapper, xmlns=xml_namespace)
# build metadata
t_metadata = etree.Element('ResponseMetadata')
t_request_id = etree.Element('RequestId')
t_request_id.text = '1549581b-12b7-11e3-895e-1334aEXAMPLE'
t_metadata.append(t_request_id)
t_root.append(t_metadata)
# build result
t_result = etree.Element(result_wrapper)
outputs = op_model.output_shape.members
replace_list = []
for output_name, output_shape in outputs.items():
t_result.append(_get_subtree(output_name, output_shape, replace_list))
t_root.append(t_result)
xml_body = etree.tostring(t_root, pretty_print=True).decode('utf-8')
xml_body_lines = xml_body.splitlines()
for replace in replace_list:
name = replace[0]
prefix = replace[1]
singular_name = singularize(name)
start_tag = '<%s>' % name
iter_name = '{}.{}'.format(prefix[-1], name.lower())if prefix else name.lower()
loop_start = '{%% for %s in %s %%}' % (singular_name.lower(), iter_name)
end_tag = '</%s>' % name
loop_end = '{{ endfor }}'
start_tag_indexes = [i for i, l in enumerate(xml_body_lines) if start_tag in l]
if len(start_tag_indexes) != 1:
raise Exception('tag %s not found in response body' % start_tag)
start_tag_index = start_tag_indexes[0]
xml_body_lines.insert(start_tag_index + 1, loop_start)
end_tag_indexes = [i for i, l in enumerate(xml_body_lines) if end_tag in l]
if len(end_tag_indexes) != 1:
raise Exception('tag %s not found in response body' % end_tag)
end_tag_index = end_tag_indexes[0]
xml_body_lines.insert(end_tag_index, loop_end)
xml_body = '\n'.join(xml_body_lines)
body = '\n{}_TEMPLATE = """{}"""'.format(operation.upper(), xml_body)
return body
def __init__(self, resource, **kwargs):
self.save = self.__save
self.resource = resource
self.__dict__ = kwargs
singular_resource = singularize(resource)
name_custom_fields = '{}_fields'.format(singular_resource)
self.CustomFields = CustomFields(**kwargs.pop(name_custom_fields, {}))
map(lambda x: setattr(self, x[0], x[1]), kwargs.items())
def __add_resource(self, resource): """ Add a vanilla resource. """ resource = inflection.singularize(resource) item = getattr(self.cli, 'new_' + resource)() create = getattr(self.dao, 'create_' + resource) create(item)
def longflag(param):
if len(param.name) > 1:
x = param.name.replace('_', '-')
if param.annotation in {list, COUNT}:
y = singularize(x)
else:
y = x
return '--' + y
def reduce_flatten_dto(dto, known_dto, config, suggested_name=None):
if isinstance(dto, list):
elements_type = None
for item in dto:
if isinstance(item, dict):
new_dto = reduce_flatten_dto(item, known_dto, config, suggested_name)
element_type = new_dto.name
else:
element_type = get_type_name(item)
if element_type:
if not elements_type:
elements_type = element_type
elif elements_type != element_type:
elements_type = get_type_name(object)
return elements_type
elif isinstance(dto, dict):
fields = []
dto_name = suggested_name
for key, value in dto.items():
if value is not None:
field = None
if isinstance(value, dict):
element_name = key
# parse special attribute to get dto name
if all(d.name != element_name for d in known_dto):
new_dto = reduce_flatten_dto(value, known_dto, config, element_name)
element_name = new_dto.name
field = FieldDescription(key, element_name)
elif isinstance(value, list):
element_name = inflection.singularize(key)
elements_type = reduce_flatten_dto(value, known_dto, config, element_name)
if not elements_type:
elements_type = inflection.singularize(key)
field = FieldDescription(key, get_type_name(value), elements_type)
else:
if key == "__name__":
dto_name = value
else:
field = FieldDescription(key, get_type_name(value))
if field is not None:
fields.append(field)
new_dto = DtoDescription(dto_name, fields)
update_dtos(known_dto, new_dto)
return new_dto
return known_dto
def metadata_Breed(data, breed_label):
breed_array = np.zeros((len(data), 370))
for i in range(len(data)):
word_list2 = set([
inflection.singularize(x)
for x in data['metadata_annots_top_desc'][i].split(' ')
])
for c, bn in enumerate([
inflection.singularize(x).lower()
for x in breed_label.BreedName.values
]):
breed_array[i, c] = int(bn in word_list2)
metadata_breed = pd.DataFrame(breed_array)
metadata_breed.columns = [
'metadata_breed_' + str(x) for x in range(breed_array.shape[1])
]
data = pd.concat([data, metadata_breed], axis=1)
return data
def inflect_given_cardinality(word, num_items):
"""
Return the singular form of the word if `num_items` is 1. Otherwise,
return the plural form of the word.
"""
if num_items == 1:
return singularize(word)
else:
return pluralize(word)
def create(cls, zendesk, **kwargs):
resource = "{}.json".format(cls.resource)
singular_resource = singularize(cls.resource)
data = {singular_resource: kwargs}
resp = zendesk._request(resource, 'POST', **data)
if resp.status_code != 201:
from exceptions import RequestException
raise RequestException(resp.status_code)
return cls(**resp.json().get(singular_resource))
def singular(self, text: str) -> str:
if text in self.__singular_cache:
return self.__singular_cache[text]
singular_text = inflection.singularize(text)
self.__singular_cache[text] = singular_text
return singular_text
def generate_model_name(raml_resource):
""" Generate model name.
:param raml_resource: Instance of ramlfications.raml.ResourceNode.
"""
resource_uri = get_resource_uri(raml_resource).strip('/')
resource_uri = re.sub('\W', ' ', resource_uri)
model_name = inflection.titleize(resource_uri)
return inflection.singularize(model_name).replace(' ', '')
def generate_model_name(raml_resource):
""" Generate model name.
:param raml_resource: Instance of ramlfications.raml.ResourceNode.
"""
resource_uri = get_resource_uri(raml_resource).strip('/')
resource_uri = re.sub('\W', ' ', resource_uri)
model_name = inflection.titleize(resource_uri)
return inflection.singularize(model_name).replace(' ', '')
def type(string: str):
try:
prefix, resulting_type = Naming.pop_prefix(string)
prefix += Naming.TYPE_PREFIX
except IndexError:
prefix = ""
resulting_type = string
resulting_type = inflection.singularize(resulting_type) if Naming._pluralize(resulting_type) else resulting_type
return prefix + inflection.camelize(resulting_type)
def names(word, format=[]):
ret = {}
ret["table"] = inflection.tableize(word)
ret["model"] = inflection.singularize(ret["table"])
ret["class"] = inflection.camelize(ret["model"])
ret["crud_class"] = inflection.camelize(ret["table"] + "_controller")
if not format:
return ret
ret = [ret[k] for k in format]
return ret[0] if len(ret) == 1 else ret
def __init__(self, data):
self.initDatabaseDictionaries()
self.words = word_tokenize(data.lower())
self.wordsUnFiltered = nltk.pos_tag(self.words)
self.wordsTagged = []
for wt in self.wordsUnFiltered:
if wt[0] not in self.stopWords:
tuple = (singularize(wt[0]), wt[1])
self.wordsTagged.append(tuple)
def check_object(job, activity, obj):
if not obj:
activity.end_date = datetime.datetime.now()
activity.status = 3
job.overall_status = 3
job.status_message = "%s Definition not found" % inflection.singularize(job.activity_type).title()
job.end_date = datetime.datetime.now()
db.session.commit()
return False
return True
def preprocess(self, text):
text = text.encode('ascii', 'ignore').lower()
text = ZEPreprocessor.replace_numeral_words(text)
text = ZEPreprocessor.re_irrelation.sub(' ', text)
text = ZEPreprocessor.re_single_space.sub('', text)
text = ZEPreprocessor.re_phone_number.sub('', text)
text = text.split('\n')
text = [' '.join([inflection.singularize(token)
for token in _.split(' ')]) for _ in text]
return text
def __delete_resource(self, resource, resource_id=None):
"""
Delete a vanilla resource.
"""
resource = inflection.singularize(resource)
if not resource_id:
item = getattr(self.cli, 'search_' + resource)()
resource_id = item.id
delete = getattr(self.dao, 'delete_' + resource)
delete(resource_id)
def serialise(self):
import inflection
import binascii
import simplejson
# 'proposal', 'superblock', etc.
name = self._meta.name
obj_type = inflection.singularize(name)
return binascii.hexlify(simplejson.dumps((obj_type, self.get_dict()), sort_keys=True).encode('utf-8')).decode('utf-8')
def standardize(ingredients):
recipe = []
for ing in ingredients:
temp = ''
ing = ing.lower()
ing = unidecode.unidecode(ing)
ing = ing.split(" or ")[0]
ing = re.sub(r"[\(\[].*?[\)\]]", " ", ing)
ing = re.sub(r"([0-9]*-ounces)+", " ", ing)
ing = re.sub(r"([0-9]*-ounce)+", " ", ing)
ing = re.sub(r"([0-9]*-inches)+", " ", ing)
ing = re.sub(r"([0-9]*-inch)+", " ", ing)
ing = re.sub(r"([0-9])+", " ", ing)
ing = re.sub(r"(-)+", " ", ing)
ing = re.sub(r"(/)+", " ", ing)
ing = ing.split(",")[0]
ing = ing.replace("*", " ")
ing = ing.replace("%", " ")
ing = ing.replace("+", " ")
ing = ing.replace("-", " ")
ing = ing.replace(".", " ")
ing = ing.replace(":", " ")
ing = ing.replace("(", " ")
ing = ing.split("http")[0]
wi = TextBlob(ing)
tagbag = wi.tags
for pos in tagbag:
if (pos[1] in tags) or (singularize(pos[0]) in allow):
if singularize(pos[0]) not in stopWords:
temp = temp + " " + singularize(pos[0])
temp = temp.strip()
if temp:
recipe.append(temp)
ingredients_set.append(temp)
return list(set(recipe))
def register_blueprints(flask, package):
package_dir = path.join(getcwd(), __name__, package)
module_suffix = '_' + singularize(package) + '.py'
module_names = [sub('\.py$', '', c)
for c in listdir(package_dir) if c.endswith(module_suffix)]
for module_name in module_names:
module = import_module(__name__ + '.%s.%s' % (package, module_name))
flask.register_blueprint(module.blueprint)
def standardize(ingredients):
recipe = []
for flag, ing in ingredients:
temp = ''
# print(ing)
# ing = '1 1/4 cups all-purpose flour (about 5 1/2 ounces)'
ing = ing.lower()
ing = unidecode.unidecode(ing)
ing = ing.split(" or ")[0]
ing = re.sub(r" ?\([^)]+\)", "", ing)
ing = re.sub(r"([0-9]*-ounces)+", "", ing)
ing = re.sub(r"([0-9]*-ounce)+", "", ing)
ing = re.sub(r"([0-9]*-inches)+", "", ing)
ing = re.sub(r"([0-9]*-inch)+", "", ing)
ing = re.sub(r"([0-9]*)+", "", ing)
ing = re.sub(r"(-)+", "", ing)
ing = re.sub(r"(/)+", "", ing)
commasplit = re.split(r"\,", ing)
ing = commasplit[0]
ing = ing.replace("*", "")
ing = ing.replace("+", "")
ing = ing.replace("-", "")
ing = ing.replace(".", "")
ing = ing.replace(":", "")
ing = ing.replace("(", "")
ing = ing.split("http")[0]
wi = TextBlob(ing)
tagbag = wi.tags
for pos in tagbag:
if (pos[1] in tags) or (singularize(pos[0]) in allow):
if singularize(pos[0]) not in stopWords:
temp = temp + " " + singularize(pos[0])
temp = re.sub(r"^\s+", "", temp)
temp = re.sub(r"\s+$", "", temp)
if temp != "":
recipe.append((flag, temp))
return recipe
def navigate_to_project_page(self, entity_type=''):
"""Navigates to a project entity query page"""
# clean up the entity_type
entity_type = inflection.singularize(inflection.camelize(entity_type))
# Form the url...
url = '%s/page/project_default?entity_type=%s&project_id=%s' %(self.baseUrl, entity_type, self.project_id)
# Get the url...
self.get(url)
# Ensure the page is loaded...
self.wait_for_page_to_load()
def type(string: str):
try:
prefix, resulting_type = Naming.pop_prefix(string)
prefix += Naming.TYPE_PREFIX
except IndexError:
prefix = ''
resulting_type = string
resulting_type = inflection.singularize(
resulting_type) if Naming._pluralize(
resulting_type) else resulting_type
return prefix + inflection.camelize(resulting_type)
def get_model(self, resource_name):
"""Returns the model :class:`flask.ext.sqlalchemy.Model`
for a resource.
:param resource_name: name of resource
"""
resource_name = singularize(resource_name)
if resource_name not in self._model_for_resources:
raise UnableToProcess('Resource Error',
'Resource [%s] not found' % resource_name)
return self._model_for_resources[resource_name]
def get_cfg(self, resource_name):
"""Returns the :class:`flask_resteasy.configs.APIConfig` for
a resource.
:param resource_name: name of resource
"""
resource_name = singularize(resource_name)
if resource_name not in self._cfg_for_resources:
raise UnableToProcess('Resource Error',
'Resource [%s] not found' % resource_name)
return self._cfg_for_resources[resource_name]
def get_model(self, resource_name):
"""Returns the model :class:`flask.ext.sqlalchemy.Model`
for a resource.
:param resource_name: name of resource
"""
resource_name = singularize(resource_name)
if resource_name not in self._model_for_resources:
raise UnableToProcess('Resource Error',
'Resource [%s] not found' % resource_name)
return self._model_for_resources[resource_name]
def next(self, token):
token = inflection.singularize(token)
# Fixing nonsensical singularization
if token in ['ha', 'contain', 'shoe']:
token += 's'
self.state = self.state.next(token)
if type(self.state) is UnexpectedState:
print "UnexpectedState!"
def get_cfg(self, resource_name):
"""Returns the :class:`flask_resteasy.configs.APIConfig` for
a resource.
:param resource_name: name of resource
"""
resource_name = singularize(resource_name)
if resource_name not in self._cfg_for_resources:
raise UnableToProcess('Resource Error',
'Resource [%s] not found' % resource_name)
return self._cfg_for_resources[resource_name]
def type(string: str):
try:
prefix, resulting_type = Naming.pop_prefix(string)
prefix += Naming.TYPE_PREFIX
except IndexError:
prefix = ''
resulting_type = string
resulting_type = singularize(resulting_type) if Naming._pluralize(
resulting_type) else resulting_type
resulting_type = resulting_type.replace(
'-', '_') # camelize does not convert '-' but '_'
return prefix + camelize(resulting_type)
def to_singular_pascal_case(name: str):
"""Make any string into a singular "PascalCased" string.
Args:
name (str): The string to convert.
Returns:
str: The converted string.
"""
return inflection.singularize(
inflection.camelize(
inflection.parameterize(inflection.titleize(name), separator="_")))
def augment_depends_on_one_get(self, attributes, name):
_get_method = inflection.singularize(name)
_key = inflection.dasherize(
inflection.underscore(
inflection.singularize(self.klass)))
_get_config_method = inflection.pluralize(name)
if 'dependencies' not in attributes:
attributes['dependencies'] = set()
attributes['dependencies'].add(_get_method)
def get(this):
name = inflection.dasherize(inflection.underscore(this.__class__.__name__))
if _key not in this.config:
return None
return next(val for key, val in
this.env.config.get(_get_config_method).iteritems() if
(val.name == this.config[_key]))
attributes[_get_method] = get
return _get_method
def folder_has_items(folder_id, folder_type):
sql = "select count(*) from %s where folder_id = %s" % (
inflection.singularize(folder_type), folder_id)
try:
result = db.engine.execute(sql)
for row in result:
if int(row[0]) > 0:
return True
return False
except Exception as ex:
print ex
return True
def singularize_token(tkn, lookup_table, micro_status):
lemma=tkn
if (tkn.endswith("us") or tkn.endswith("ia") or tkn.endswith(
"ta")): # for inflection exception in general-takes into account both lower and upper case (apart from some inflection-exception list used also in next
lemma = tkn
elif (tkn not in lookup_table[
"inflection_exceptions"]): # Further Inflection Exception list is taken into account
lemma = inflection.singularize(tkn)
if (tkn != lemma): # Only in case when inflection makes some changes in lemma
micro_status.append("Inflection (Plural) Treatment: " + tkn)
return lemma
def censor(input_text):
""" Returns input_text with any profane words censored """
bad_words = self.get_profane_words()
spl = input_text.split(' ')
for index, value in spl:
word = singularize(value)
if word in bad_words:
spl[index] = '*' * len(value)
return ' '.join(spl)
def generatePair(opinionList, featureList):
pair = {}
for (key, group) in featureList.groupby("business_id"):
opinion = opinionList[key]
pair[key] = []
feature = list(group["featureList"])[0]
for f in feature:
attribute = [jj for (nn, jj) in opinion if singularize(nn) == f]
if attribute:
result = [(f, attr) for attr in attribute]
pair[key].extend(result)
return pair
def singularize_rule(self):
item = self.item
if len(item['prioritized_docids']) < 1:
claim_tokens = item['claim_tokens']
# finded_keys = item['prioritized_docids']
claim_tokens = [inflection.singularize(c) for c in claim_tokens]
claim = ' '.join(claim_tokens)
fkd_new, fk_new = self._keyword_match(claim, raw_set=True)
# finded_keys = set(finded_keys) | set(fk_new)
item['prioritized_docids'] = list(fk_new)
item['structured_docids'] = fkd_new
return self
def test_uncountable_word_is_not_greedy():
uncountable_word = "ors"
countable_word = "sponsor"
inflection.UNCOUNTABLES.add(uncountable_word)
try:
assert uncountable_word == inflection.singularize(uncountable_word)
assert uncountable_word == inflection.pluralize(uncountable_word)
assert(
inflection.pluralize(uncountable_word) ==
inflection.singularize(uncountable_word)
)
assert "sponsor" == inflection.singularize(countable_word)
assert "sponsors" == inflection.pluralize(countable_word)
assert (
"sponsor" ==
inflection.singularize(inflection.pluralize(countable_word))
)
finally:
inflection.UNCOUNTABLES.remove(uncountable_word)
def cleaning_data_function_new(text):
text = text.lower()
words = re.split(r'\n', text)
for word in words:
text1 = ProgRoman.sub(' ', word)
split = re.split(r' ', text1)
msg = [
inflection.singularize(text2) for text2 in split
if (text2 not in stoplist) and (len(text2) > 2)
]
new1.append(msg)
return new1
def from_json(cls, name: str, data: str, skip_fields_with_errors=False):
data = json.loads(data)
# TODO: More useful support for lists
# TODO: Bubble up a warning that we ignored everything except the first nonlist item
if isinstance(data, list):
if len(data) == 0:
raise NotImplementedError("Top-level array cannot be an empty list")
items_name = pluralize(name)
if singularize(items_name) == name:
items_name = "items"
return cls.from_dict(name, {items_name: data}, skip_fields_with_errors)
return cls.from_dict(name, data, skip_fields_with_errors)
def sanitize_name(name):
singular = inflection.singularize(name)
plural = name
if singular == name:
plural = inflection.pluralize(name)
num = 2
while os.path.exists(plural + '.py'):
plural = plural + str(num)
num = num + 1
class_name = inflection.camelize(singular)
return singular, plural, class_name
def singularize_rule(self):
"""Singularize words
"""
item = self.item
if len(item['prioritized_docids']) < 1:
claim_tokens = item['claim_tokens']
# finded_keys = item['prioritized_docids']
claim_tokens = [inflection.singularize(c) for c in claim_tokens]
claim = ' '.join(claim_tokens)
fk_new = self._keyword_match(claim)
# finded_keys = set(finded_keys) | set(fk_new)
item['prioritized_docids'] = list(fk_new)
return self
def augment(self, attributes, name):
_get_method = self.augment_depends_on_one_get(attributes, name)
_create = call_parents('create')
if 'create' in attributes:
_create = attributes['create']
def create(self):
item = getattr(self, _get_method)()
if item and (not item.created()):
item.create()
_create(self)
attributes['create'] = create
return {inflection.singularize(name): None}
def __get_raw_data__(self):
if self._raw_data:
return self._raw_data
cls = self.__class__
params = {'id': str(self.code)}
options = Util.merge_options('params', params, **self.options)
path = Util.constructed_path(cls.get_path(), options['params'])
r = Connection.request('get', path, **options)
response_data = r.json()
Util.convert_to_dates(response_data)
self._raw_data = response_data[singularize(cls.lookup_key())]
return self._raw_data
def inflect(that,word):
"""
:param that: some object; if it has a length, we use that
:param word: this word will be made singular or plural depending on the singularity or plurarlity of that
:return: inflected word
"""
try:
cnt = len(that)
except TypeError:
cnt = 1
inflected_word = singularize(word) if cnt == 1 else pluralize(word)
return f"{cnt} {inflected_word}"
def singularize(word):
'''
Return the singular form of the given word.
TODO: Check that word is a noun (or an adjective or at any rate can
be sensibly used as a noun) before calling inflection.singularize?
FIXME BUGS: inflection returns many wrong answers by pattern:
*aves -> *afe
It uses incomplete special case matching (octopus),
and does not recognize many other pairs such as:
(locus, loci)
NB: pattern3.en is not yet functional (2017.07.10)
'''
if word.lower()[-4:] == 'aves':
return word.rstrip('sS')
return inflection.singularize(word)
def __init__(self, klass, other, rel_column=None, backref=None, **kwargs):
self.klass = klass
self.other = other
if rel_column:
prop_name = inflection.pluralize(backref) # phonenumbers
backref_id = rel_column # owner_id
backref = backref_id.split('_id')[0] # parent
else:
# user.projects
prop_name = inflection.pluralize(self.klass.__name__).lower()
# project.user_id
backref_id = '%s_id' % inflection.singularize(self.other.__name__).lower()
# project.user
backref = inflection.singularize(self.other.__name__).lower()
setattr(other, prop_name, self)
self.rel_column = backref_id
column = Column(ObjectId)
column.name = backref_id
klass.__columns__[backref_id] = column
setattr(klass, backref_id, column)
setattr(klass, backref, RelationshipHasOne(klass, other, backref_id))
print ">>>", backref
def name_or_flag(has_keyword_only, param):
def _name_or_flag(p):
name = p.name.replace('_', '-')
if p.default == p.empty or (has_keyword_only and p.kind == p.POSITIONAL_OR_KEYWORD):
return p.name
elif len(name) == 1:
return '-' + name
else:
return '--' + name
try:
if issubclass(param.annotation, (list, Count)):
return singularize(_name_or_flag(param))
except TypeError:
pass
return _name_or_flag(param)
def find_related_fields_to_rename(fields, prefix=[]):
"""
Find all related fields on a Serializer or NestedBoundField.
Args:
fields: an object whose `.__iter__()` method returns instances of
`BoundField` (e.g. a `Serializer` or `NestedBoundField`)
prefix: a list of parent keys that map to the current list of
fields. (e.g. `{'one': {'two': {'three': 'value'}}}` will have
`prefix=['one', 'two'] when `{'three': 'value'})` is passed
into the function.
Returns:
list: list of RelatedFieldRename instances.
"""
ret = []
if isinstance(fields, ListSerializer):
fields = fields.child.__class__()
# this is selfish hack to support something I am doing
# hopefully I can factor this out soon
if isinstance(fields, BoundField):
fields = fields._field
# iterate over each field to determine if it is related
for field in fields:
assert isinstance(field, BoundField), (
'Fields must be an iterator that returns `BoundField` instances.'
)
key = prefix + field.name.split('.')
# iterate over any nested lists
if field._proxy_class is ListSerializer:
ret.extend(
find_related_fields_to_rename(field.child, prefix=key)
)
# any nested dicts should be iterated over
# this could check if `field` is an instance of `NestedBoundField`
# but that would break my hack above
elif hasattr(field, 'fields'):
ret.extend(find_related_fields_to_rename(field, prefix=key))
# the field is a list of pks and should be renamed
elif field._proxy_class is ManyRelatedField:
new_name = singularize(key[-1]) + '_ids'
ret.append(RelatedFieldRename(key, new_name))
# the field is a single pk and should be renamed
elif field._proxy_class is PrimaryKeyRelatedField:
new_name = key[-1] + '_id'
ret.append(RelatedFieldRename(key, new_name))
return ret
def is_match_splabel(labels, label_to_trace, target_list, gt):
# all matched gt words
all_gt_words = []
for ll in gt:
for l in ll:
if not l.find('none') >= 0:
gt_word = inflection.singularize(l.split('-')[-1])
all_gt_words += [gt_word]
for label in labels:
is_correct = is_label_correct(label, label_to_trace, target_list, all_gt_words)
if is_correct:
return True
return False
def stemming(word):
# if word in be_words:
# return 'be'
# try:
# if en.is_noun(word):
# word = en.noun.singular(word)
# elif en.is_adjective(word) or en.is_adverb(word):
# return word
# else:
# word = en.verb.present(word)
# except Exception:
# pass
# return en.spelling.correct(word)
word = inflection.singularize(word)
return word