def NAMES(*args, config, op=None):
if type(args[0]) == list:
initial_list = [resolve_value(arg, config=config)
for arg in flatten(args)]
else:
initial_list = [args[0]]
names = list([" ".join(filter(remove_empty, names))
for names in generate_names(initial_list)])
logger.debug("Generated list of names: {}".format(names))
new_op = ExtendedOp(op)
new_op.case_sensitive_override = True
return "any_of", names, new_op
def PLURALIZE(*args, config, op=None):
"""
For a noun or a list of nouns, it will match any singular or plural word
Usage for a single word, e.g.:
PLURALIZE("car")
Usage for lists, e.g.:
vehicles = {"car", "bicycle", "ship"}
PLURALIZE(vehicles)
Will work even for regex or if the lemmatizer of spaCy is making an error
Has dependency to the Python inflect package https://pypi.org/project/inflect/
"""
if type(args[0]) == list:
initial_list = [
resolve_value(arg, config=config) for arg in flatten(args)
]
else:
initial_list = [args[0]]
return "any_of", pluralizing(initial_list), op
def nested_parse(values, config: "SessionConfig",
op: ExtendedOp) -> SpacyPattern:
from rita.macros import resolve_value
results = rules_to_patterns(
"", [resolve_value(v, config=config) for v in values], config=config)
return results["pattern"]
def nested_parse(values, config, op=None):
from rita.macros import resolve_value
results = rules_to_patterns(
"", [resolve_value(v, config=config) for v in values], config=config)
return results["pattern"]
def TAG(name, config, op=None):
"""
For generating POS/TAG patterns based on a Regex
e.g. TAG("^NN|^JJ") for nouns or adjectives
"""
return "tag", resolve_value(name, config=config), op
def nested_parse(values, config, op=None):
from rita.macros import resolve_value
(_, patterns) = rules_to_patterns(
"", [resolve_value(v, config=config) for v in values], config=config)
return r"(?P<g{}>{})".format(config.new_nested_group_id(),
"".join(patterns))
def nested_parse(values, config: "SessionConfig", op: ExtendedOp) -> str:
from rita.macros import resolve_value
(_, patterns) = rules_to_patterns("", [resolve_value(v, config=config)
for v in values], config=config)
return r"(?P<g{}>{})".format(config.new_nested_group_id(), "".join(patterns))
def FUZZY(name, config, op=None):
initial = resolve_value(name, config=config)
return ("fuzzy", list(premutations(initial.lower())), op)