def test_type_errors():
F = fact('F', ['a'])
RULE = rule(
'a',
eq('1').interpretation(
custom(int)
)
).interpretation(
F.a
)
parser = Parser(RULE)
match = parser.match('a 1')
with pytest.raises(TypeError):
match.fact
F = fact('F', ['a'])
RULE = rule(
'a',
eq('1').interpretation(
custom(int)
)
).interpretation(
custom(str)
)
parser = Parser(RULE)
match = parser.match('a 1')
with pytest.raises(TypeError):
match.fact
def test_nested_facts():
F = fact('F', ['a'])
G = fact('G', ['b'])
RULE = rule(eq('a').interpretation(F.a)).interpretation(F).interpretation(
G.b).interpretation(G)
parser = Parser(RULE)
match = parser.match('a')
record = match.fact
assert record == G(b=F(a='a'))
assert record.spans == [(0, 1)]
assert record.as_json == {'b': {'a': 'a'}}
def test_pipeline_key():
from yargy.pipelines import morph_pipeline
pipeline = morph_pipeline([
'закрытое общество',
'завод'
])
F = fact('F', ['a'])
RULE = pipeline.interpretation(
F.a.normalized()
).interpretation(
F
)
parser = Parser(RULE)
match = parser.match('закрытом обществе')
record = match.fact
assert record == F(a='закрытое общество')
RULE = pipeline.interpretation(
normalized()
)
parser = Parser(RULE)
match = parser.match('заводе')
value = match.fact
assert value == 'завод'
class ConsoleGame:
__game = fact(
'Game',
['name', 'version_number', 'version_name', 'console']
)
__amount_of_games = []
def __init__(self, names: list = [], version_numbers: list = [], version_names: list = [], consoles: list = []):
rules = rule(morph_pipeline(names).interpretation(self.__game.name.const(names[0])),
morph_pipeline(version_numbers).interpretation(self.__game.version_number).optional(),
morph_pipeline(version_names).interpretation(self.__game.version_name).optional(),
morph_pipeline(consoles).interpretation(self.__game.console).optional())
game = or_(rules).interpretation(self.__game)
self.parser = Parser(game)
def matches(self, data):
matches = []
for sent in data.text[:9000]:
for match in self.parser.findall(sent):
matches.append(match.fact)
self.__amount_of_games.append(matches)
for m in matches:
print(m.name, m.version_number, m.version_name, m.console)
print(len(self.__amount_of_games))
def test_type_errors():
F = fact('F', ['a'])
RULE = rule('a', eq('1').interpretation(custom(int))).interpretation(F.a)
parser = Parser(RULE)
match = parser.match('a 1')
with pytest.raises(TypeError):
match.fact
F = fact('F', ['a'])
RULE = rule('a',
eq('1').interpretation(custom(int))).interpretation(
custom(str))
parser = Parser(RULE)
match = parser.match('a 1')
with pytest.raises(TypeError):
match.fact
def test_pipeline_key():
from yargy import or_
from yargy.pipelines import morph_pipeline
pipeline = morph_pipeline([
'закрытое общество',
'завод'
])
F = fact('F', ['a'])
RULE = pipeline.interpretation(
F.a.normalized()
).interpretation(
F
)
parser = Parser(RULE)
match = parser.match('закрытом обществе')
record = match.fact
assert record == F(a='закрытое общество')
RULE = pipeline.interpretation(
normalized()
)
parser = Parser(RULE)
match = parser.match('заводе')
value = match.fact
assert value == 'завод'
def test_name():
Name = fact(
'Name',
['first', 'last']
)
gnc = gnc_relation()
FIRST = gram('Name').interpretation(
Name.first.inflected()
).match(gnc)
LAST = gram('Surn').interpretation(
Name.last.inflected()
).match(gnc)
NAME = rule(
FIRST,
LAST
).interpretation(Name)
parser = Parser(NAME)
match = parser.match('саше иванову')
assert match.fact == Name(first='саша', last='иванов')
match = parser.match('сашу иванову')
assert match.fact == Name(first='саша', last='иванова')
match = parser.match('сашу ивановой')
assert not match
def __new__(mcs, typename, base_classes, class_attr):
if class_attr.get('_ROOT_FACT_DEFINITION', False):
# Checking for attempt to redeclare base definition class
if mcs.BASE_FACT_DEFINITION_CLS is not None:
raise TypeError(
f"Attempt to redeclare base fact definition class "
f"'{mcs.BASE_FACT_DEFINITION_CLS.__name__}' by '{typename}'"
)
# Saving root class to metaclass attributes
mcs.BASE_FACT_DEFINITION_CLS = super().__new__(
mcs, typename, base_classes, class_attr)
return mcs.BASE_FACT_DEFINITION_CLS
# TODO CONSIDER: research ability of mixins support
# TODO CONSIDER: research ability of transitive inheritance from base class
if base_classes != (mcs.BASE_FACT_DEFINITION_CLS, ):
raise TypeError(
f"Class {typename} must be inherited directly from FactDefinition only."
f" Mixins and transitive inheritance are not currently supported"
)
annotations = class_attr.get("__annotations__")
if not annotations:
raise TypeError(
f"No annotations declared in fact definition class '{typename}'"
)
generated_fact_cls = fact(f"{type}AutoGen", list(annotations))
new_base_classes = (generated_fact_cls, )
return super().__new__(mcs, typename, new_base_classes, class_attr)
def test_normalized_custom_attribute():
F = fact('F', ['a'])
MONTHS = {'январь': 1}
RULE = rule('январе').interpretation(F.a.normalized().custom(
MONTHS.get)).interpretation(F)
parser = Parser(RULE)
match = parser.match('январе')
assert match.fact == F(a=1)
def test_inflected_custom_attribute():
F = fact('F', ['a'])
MONTHS = {'январь': 1}
RULE = rule('январе').interpretation(
F.a.inflected({'nomn', 'sing'}).custom(MONTHS.get)).interpretation(F)
parser = Parser(RULE)
match = parser.match('январе')
assert match.fact == F(a=1)
def test_nested_facts():
F = fact('F', ['a'])
G = fact('G', ['b'])
RULE = rule(
eq('a').interpretation(F.a)
).interpretation(
F
).interpretation(
G.b
).interpretation(
G
)
parser = Parser(RULE)
match = parser.match('a')
record = match.fact
assert record == G(b=F(a='a'))
assert record.spans == [(0, 1)]
assert record.as_json == {'b': {'a': 'a'}}
def test_attribute_custom():
F = fact('F', 'a')
RULE = rule('1').interpretation(F.a.custom(int)).interpretation(F)
parser = Parser(RULE)
match = parser.match('1')
record = match.fact
assert record == F(a=1)
assert record.spans == [(0, 1)]
assert record.as_json == {'a': 1}
def test_attribute_custom_custom():
F = fact('F', 'a')
MAPPING = {'a': 1}
RULE = rule('A').interpretation(F.a.custom(str.lower).custom(
MAPPING.get)).interpretation(F)
parser = Parser(RULE)
match = parser.match('A')
record = match.fact
assert record == F(a=1)
def test_attribute_normalized():
F = fact('F', 'a')
RULE = rule('январе').interpretation(F.a.normalized()).interpretation(F)
parser = Parser(RULE)
match = parser.match('январе')
record = match.fact
assert record == F(a='январь')
assert record.spans == [(0, 6)]
assert record.as_json == {'a': 'январь'}
def test_rule_attribute():
F = fact('F', ['a'])
RULE = rule('a', 'A').interpretation(F.a).interpretation(F)
parser = Parser(RULE)
match = parser.match('a A')
record = match.fact
assert record == F(a='a A')
assert record.spans == [(0, 5)]
assert record.as_json == {'a': 'a A'}
def test_attribute():
F = fact('F', 'a')
RULE = rule(
'a'
).interpretation(
F.a
)
parser = Parser(RULE)
match = parser.match('a')
assert match.fact == 'a'
def test_repeatable():
F = fact('F', [attribute('a').repeatable()])
RULE = rule(eq('a').interpretation(F.a),
eq('b').interpretation(F.a)).interpretation(F)
parser = Parser(RULE)
match = parser.match('a b')
record = match.fact
assert record == F(a=['a', 'b'])
assert record.spans == [(0, 1), (2, 3)]
assert record.as_json == {'a': ['a', 'b']}
def test_attribute_const():
F = fact('F', 'a')
RULE = rule(
'январь'
).interpretation(
F.a.const(1)
)
parser = Parser(RULE)
match = parser.match('январь')
assert match.fact == 1
def test_insted_attributes():
F = fact('F', ['a', 'b'])
RULE = rule(eq('a').interpretation(F.a)).interpretation(
F.b).interpretation(F)
parser = Parser(RULE)
match = parser.match('a')
record = match.fact
assert record == F(a=None, b='a')
assert record.spans == [(0, 1)]
assert record.as_json == {'b': 'a'}
def test_attribute_const():
F = fact('F', 'a')
RULE = rule(
'январь'
).interpretation(
F.a.const(1)
)
parser = Parser(RULE)
match = parser.match('январь')
assert match.fact == 1
def test_attribute():
F = fact('F', 'a')
RULE = rule(
'a'
).interpretation(
F.a
)
parser = Parser(RULE)
match = parser.match('a')
assert match.fact == 'a'
def test_attribute_inflected():
F = fact('F', 'a')
RULE = rule('январе').interpretation(F.a.inflected({'nomn', 'plur'
})).interpretation(F)
parser = Parser(RULE)
match = parser.match('январе')
record = match.fact
assert record == F(a='январи')
assert record.spans == [(0, 6)]
assert record.as_json == {'a': 'январи'}
def test_predicate_attribute():
F = fact('F', ['a'])
RULE = rule(
eq('a').interpretation(F.a)
).interpretation(F)
parser = Parser(RULE)
match = parser.match('a')
record = match.fact
assert record == F(a='a')
assert record.spans == [(0, 1)]
assert record.as_json == {'a': 'a'}
def test_merge_facts():
F = fact('F', ['a', 'b'])
A = rule(eq('a').interpretation(F.a)).interpretation(F)
B = rule(eq('b').interpretation(F.b)).interpretation(F)
RULE = rule(A, B).interpretation(F)
parser = Parser(RULE)
match = parser.match('a b')
record = match.fact
assert record == F(a='a', b='b')
assert record.spans == [(0, 1), (2, 3)]
assert record.as_json == {'a': 'a', 'b': 'b'}
def test_rule_attribute_custom():
F = fact('F', ['a'])
RULE = rule(
'1'
).interpretation(
F.a
).interpretation(
custom(int)
)
parser = Parser(RULE)
match = parser.match('1')
assert match.fact == 1
def test_rule_attribute_custom():
F = fact('F', ['a'])
RULE = rule(
'1'
).interpretation(
F.a
).interpretation(
custom(int)
)
parser = Parser(RULE)
match = parser.match('1')
assert match.fact == 1
def test_insted_attributes():
F = fact('F', ['a', 'b'])
RULE = rule(
eq('a').interpretation(F.a)
).interpretation(
F.b
).interpretation(F)
parser = Parser(RULE)
match = parser.match('a')
record = match.fact
assert record == F(a=None, b='a')
assert record.spans == [(0, 1)]
assert record.as_json == {'b': 'a'}
def test_attribute_custom_custom():
F = fact('F', 'a')
MAPPING = {'a': 1}
RULE = rule(
'A'
).interpretation(
F.a.custom(str.lower).custom(MAPPING.get)
).interpretation(
F
)
parser = Parser(RULE)
match = parser.match('A')
record = match.fact
assert record == F(a=1)
def test_repeatable():
F = fact('F', [attribute('a').repeatable()])
RULE = rule(
eq('a').interpretation(F.a),
eq('b').interpretation(F.a)
).interpretation(
F
)
parser = Parser(RULE)
match = parser.match('a b')
record = match.fact
assert record == F(a=['a', 'b'])
assert record.spans == [(0, 1), (2, 3)]
assert record.as_json == {'a': ['a', 'b']}
def test_attribute_normalized():
F = fact('F', 'a')
RULE = rule(
'январе'
).interpretation(
F.a.normalized()
).interpretation(
F
)
parser = Parser(RULE)
match = parser.match('январе')
record = match.fact
assert record == F(a='январь')
assert record.spans == [(0, 6)]
assert record.as_json == {'a': 'январь'}
def test_attribute_inflected():
F = fact('F', 'a')
RULE = rule(
'январе'
).interpretation(
F.a.inflected({'nomn', 'plur'})
).interpretation(
F
)
parser = Parser(RULE)
match = parser.match('январе')
record = match.fact
assert record == F(a='январи')
assert record.spans == [(0, 6)]
assert record.as_json == {'a': 'январи'}
def test_attribute_custom():
F = fact('F', 'a')
RULE = rule(
'1'
).interpretation(
F.a.custom(int)
).interpretation(
F
)
parser = Parser(RULE)
match = parser.match('1')
record = match.fact
assert record == F(a=1)
assert record.spans == [(0, 1)]
assert record.as_json == {'a': 1}
def test_normalized_custom_attribute():
F = fact('F', ['a'])
MONTHS = {
'январь': 1
}
RULE = rule(
'январе'
).interpretation(
F.a.normalized().custom(MONTHS.get)
).interpretation(
F
)
parser = Parser(RULE)
match = parser.match('январе')
assert match.fact == F(a=1)
def test_inflected_custom_attribute():
F = fact('F', ['a'])
MONTHS = {
'январь': 1
}
RULE = rule(
'январе'
).interpretation(
F.a.inflected({'nomn', 'sing'}).custom(MONTHS.get)
).interpretation(
F
)
parser = Parser(RULE)
match = parser.match('январе')
assert match.fact == F(a=1)
def test_merge_facts():
F = fact('F', ['a', 'b'])
A = rule(
eq('a').interpretation(F.a)
).interpretation(F)
B = rule(
eq('b').interpretation(F.b)
).interpretation(F)
RULE = rule(
A, B
).interpretation(F)
parser = Parser(RULE)
match = parser.match('a b')
record = match.fact
assert record == F(a='a', b='b')
assert record.spans == [(0, 1), (2, 3)]
assert record.as_json == {'a': 'a', 'b': 'b'}
def yargy_smart_home(msg):
Do = fact('Entity', ['action', 'object', 'place'])
Actions = dictionary({'Включи', 'Отключи', 'Выключи'})
Objects = dictionary(
{'Лампочку', 'Свет', 'Розетку', 'Видеокамеру', 'Камеру'})
ObjectsList = or_(
rule(Objects),
rule(Objects, Objects),
)
Prep = dictionary({'в', 'на'})
Place = dictionary({
'Гостевой', 'Ванной', 'спальной', 'спальне', 'холле', 'коридоре',
'кухне'
})
Room = {'комната'}
ActionPhrase = or_(
rule(Actions.interpretation(Do.action.normalized()),
Objects.interpretation(Do.object.normalized()), Prep.optional(),
Place.interpretation(Do.place.normalized()),
rule(normalized('комната')).optional()),
rule(Actions.interpretation(Do.action.normalized()),
Objects.interpretation(Do.object.normalized()), Prep.optional(),
Place.interpretation(Do.place.normalized())),
rule(Prep.optional(), Place.interpretation(Do.place.normalized()),
rule(normalized('комната')).optional(),
Actions.interpretation(Do.action.normalized()),
Objects.interpretation(
Do.object.normalized()))).interpretation(Do)
res = []
parser = Parser(ActionPhrase)
for match in parser.findall(msg):
res.append({
'Действие': match.fact.action,
'Объект': match.fact.object,
'Место': match.fact.place,
})
return res
def test_bnf():
from yargy.interpretation import fact
from yargy.relations import gnc_relation
F = fact('F', ['a'])
gnc = gnc_relation()
assert_bnf(rule('a').named('A').interpretation(F), "F -> 'a'")
assert_bnf(
rule('a').interpretation(F.a).interpretation(F), 'F -> F.a',
"F.a -> 'a'")
assert_bnf(rule('a').match(gnc).interpretation(F.a), "F.a^gnc -> 'a'")
assert_bnf(
rule('a').interpretation(F.a).repeatable(), 'R0 -> F.a | F.a R0',
"F.a -> 'a'")
assert_bnf(
rule('a').repeatable().interpretation(F.a), 'F.a -> R1',
"R1 -> 'a' | 'a' R1")
def test_bnf():
from yargy.interpretation import fact
from yargy.relations import gnc_relation
F = fact('F', ['a'])
gnc = gnc_relation()
assert_bnf(
rule('a').named('A').interpretation(F),
"F -> 'a'"
)
assert_bnf(
rule('a').interpretation(F.a).interpretation(F),
'F -> F.a',
"F.a -> 'a'"
)
assert_bnf(
rule('a').match(gnc).interpretation(F.a),
"F.a^gnc -> 'a'"
)
assert_bnf(
rule('a').interpretation(F.a).repeatable(),
'R0 -> F.a R0 | F.a',
"F.a -> 'a'"
)
assert_bnf(
rule('a').repeatable().interpretation(F.a),
'F.a -> R1',
"R1 -> 'a' R1 | 'a'"
)
A = rule('a')
B = A.named('B')
C = A.named('C')
D = rule(B, C).named('D')
assert_bnf(
D,
'D -> B C',
'B -> R0',
'C -> R0',
"R0 -> 'a'"
)
def yargy_get_genre(msg):
Genre = fact('Genre', ['genre'])
GENRES = {
'ужасы', 'ужастики', 'мелодрама', 'комедия', 'боевик', 'триллер',
'мультик', 'мультфильм', 'драма'
}
GENRES_NAME = dictionary(GENRES)
GENRES_WORDS = or_(rule(normalized('жанр')), rule(normalized('раздел')))
GENRE_PHRASE = or_(rule(GENRES_NAME, GENRES_WORDS.optional()),
rule(GENRES_WORDS.optional(),
GENRES_NAME)).interpretation(
Genre.genre.inflected()).interpretation(Genre)
res = []
parser = Parser(GENRE_PHRASE)
for match in parser.findall(msg):
res.append(match.fact.genre)
return res
def yargy_get_channel(msg):
Channel = fact('Channel', ['name'])
CNANNELS = {
'Первый', 'Россия', 'ТВЦ', 'НТВ', 'ТНТ', 'СТС', 'Культура', 'Дождь',
'Спас'
}
CNANNELS_NAME = dictionary(CNANNELS)
CHANNEL_WORDS = or_(rule(normalized('канал')),
rule(normalized('программа')))
CHANNEL_PHRASE = or_(
rule(CHANNEL_WORDS, CNANNELS_NAME),
rule(CNANNELS_NAME, CHANNEL_WORDS.optional())).interpretation(
Channel.name.inflected()).interpretation(Channel)
res = []
parser = Parser(CHANNEL_PHRASE)
for match in parser.findall(msg):
# print(match.fact)
for channel in CNANNELS:
if channel.lower() in match.fact.name:
res.append(channel)
return res
from yargy import Parser, rule, and_, not_, or_
from yargy.interpretation import fact
from yargy.predicates import gram
from yargy.relations import gnc_relation
from yargy.pipelines import morph_pipeline
import pymorphy2
morph = pymorphy2.MorphAnalyzer()
LimbState = fact(
'LimbState',
['state'],
)
Limb = fact(
'limb',
['name'],
)
Disease = fact('Person', ['limb', 'limbstate'])
LIMBSTATE = rule(gram('VERB').interpretation(
LimbState.state)).interpretation(LimbState)
LIMB = rule(gram('NOUN').interpretation(Limb.name)).interpretation(Limb)
DISEASE = or_(
rule(
LIMB.interpretation(Disease.limb),
or_(
from .name import SIMPLE_NAME
from .person import POSITION_NAME
from yargy.rule.transformators import RuleTransformator
class StripInterpretationTransformator(RuleTransformator):
def visit_InterpretationRule(self, item):
return self.visit(item.rule)
NAME = SIMPLE_NAME.transform(StripInterpretationTransformator)
PERSON = POSITION_NAME.transform(StripInterpretationTransformator)
Organisation = fact('Organisation', ['name'])
TYPE = morph_pipeline([
'АО',
'ОАО',
'ООО',
'ЗАО',
'ПАО',
# TODO Check abbrs
# 'ик',
# 'нк',
# 'хк',
# 'ип',
# 'чп',
from yargy import rule, and_, or_, not_
from yargy.predicates import eq, type as _type, normalized, custom
from yargy.pipelines import morph_pipeline
from yargy.interpretation import fact
CityFact = fact('city', ['prefix', 'title'])
CityTitle = morph_pipeline({
'липецк', 'сургут', 'нальчик', 'москва', 'санкт-петербург', 'питер',
'нижний новгород', 'видное'
}).interpretation(CityFact.title.normalized())
CityRule = rule(
normalized('город').optional().interpretation(CityFact.prefix), CityTitle,
eq(';').optional()).interpretation(CityFact)
from yargy import (rule, or_)
from yargy.interpretation import (fact)
from yargy.predicates import (dictionary, is_capitalized, eq, caseless)
from yargy.pipelines import caseless_pipeline, morph_pipeline
from natasha.extractors import Extractor
Position = fact('position', ['level', 'field', 'name'])
LEVEL = rule(
caseless_pipeline([
'junior', 'middle', 'senior', 'lead', 'chief', 'head', 'team lead',
"старший", "младший", "руководитель направления"
]).interpretation(Position.level))
# TODO: нужно учесть жаргонные варианты (датасаентолог, датасатанист и т.д.) Скорее всего, придется парсить регулярками
NAME = rule(
or_(
caseless_pipeline([
'data scientist', 'data engineer', 'engineer', 'analyst',
'data analyst', 'data manager', 'scientist', 'researcher',
"developer", "intern"
]), rule(dictionary(['DS', 'DE']), is_capitalized()),
morph_pipeline(["аналитик", "разработчик",
"стажер"])).interpretation(Position.name.inflected()))
FIELD = rule(
caseless_pipeline([
'ML', 'DL', 'CV', 'computer vision', 'NLP', 'bi', 'machine learning',
'deep learning', 'software', 'research', 'big data', 'python', 'c++',
"scala", "java", 'ios', "android", 'devops', "backend", 'frontend'
# coding: utf-8
from __future__ import unicode_literals
from yargy import (
rule,
and_, or_
)
from yargy.interpretation import fact, attribute
from yargy.predicates import (
eq, gte, lte, length_eq,
dictionary, normalized,
)
Date = fact(
'Date',
['year', 'month', 'day', attribute('current_era', True)]
)
MONTHS = {
'январь': 1,
'февраль': 2,
'март': 3,
'апрель': 4,
'май': 5,
'июнь': 6,
'июль': 7,
'август': 8,
'сентябрь': 9,
'октябрь': 10,
'ноябрь': 11,
from yargy import (
Parser,
rule,
and_, or_
)
from yargy.pipelines import morph_pipeline
from yargy.interpretation import fact, attribute
from yargy.predicates import (
eq, gte, lte, length_eq,
dictionary, normalized, gram
)
COD = fact(
'Codex',
['n0', 'point', 'n1', 'subpoint', 'n2', 'part', 'n3', 'article', 'n4', 'par', 'n5',
'subsection', 'n6', 'section', 'n7', 'chapter', 'n8', 'type', 'codex']
)
COURT_ = fact(
'Court',
['smth', 'type', 'court', 'rf']
)
NUM = and_(gte(1), lte(10000))
NUMBERS = rule(NUM,
rule(eq('.').optional(), NUM).repeatable().optional())
CODEX = rule(
or_(rule(normalized('пункт')),
rule('п', eq('.').optional())
or_, and_
)
from yargy.interpretation import fact, attribute
from yargy.predicates import (
eq, lte, gte, gram, type, tag,
length_eq,
in_, in_caseless, dictionary,
normalized, caseless,
is_title
)
from yargy.pipelines import morph_pipeline
from yargy.tokenizer import QUOTES
Address = fact(
'Address',
[attribute('parts').repeatable()]
)
Index = fact(
'Index',
['value']
)
Country = fact(
'Country',
['name']
)
Region = fact(
'Region',
['name', 'type']
)
Settlement = fact(
'Settlement',
from yargy import (
rule,
or_
)
from yargy.interpretation import fact
from yargy.predicates import gram
from yargy.pipelines import morph_pipeline
from .name import (
NAME,
SIMPLE_NAME
)
Person = fact(
'Person',
['position', 'name']
)
POSITION = morph_pipeline([
'святой',
'патриарх',
'митрополит',
'царь',
'король',
'царица',
'император',
'императрица',
'принц',
'принцесса',
from ipymarkup import show_markup
from yargy import rule, and_, or_
from yargy.interpretation import fact, attribute
from yargy.predicates import dictionary, normalized
ProfileTypeFact = fact('ProfileTypeFact', ['profile'])
PROFILES = {
'модель': 'Model',
'визажист': 'Visagiste',
'фотограф': 'Photographer',
'стилист': 'Stylist',
}
NAME = dictionary(PROFILES).interpretation(
ProfileTypeFact.profile.normalized().custom(PROFILES.__getitem__))
PROFILE_TYPE_PARSER = or_(rule(NAME), ).interpretation(ProfileTypeFact)
eq, length_eq,
gram, tag,
is_single, is_capitalized
)
from yargy.predicates.bank import DictionaryPredicate as dictionary
from yargy.relations import gnc_relation
from natasha.data import load_dict
from yargy.rule.transformators import RuleTransformator
from yargy.rule.constructors import Rule
from yargy.predicates.constructors import AndPredicate
Name = fact(
'Name',
['first', 'middle', 'last', 'nick']
)
FIRST_DICT = set(load_dict('first.txt'))
MAYBE_FIRST_DICT = set(load_dict('maybe_first.txt'))
LAST_DICT = set(load_dict('last.txt'))
##########
#
# COMPONENTS
#
###########
eq, length_eq,
in_, in_caseless,
gram, type,
normalized, caseless, dictionary
)
from natasha.utils import Record
from natasha.dsl import (
Normalizable,
money as dsl
)
Money = fact(
'Money',
['integer', 'fraction', 'multiplier', 'currency', 'coins']
)
class Money(Money, Normalizable):
@property
def normalized(self):
amount = self.integer
if self.fraction:
amount += self.fraction / 100
if self.multiplier:
amount *= self.multiplier
if self.coins:
amount += self.coins / 100
return dsl.Money(amount, self.currency)
from yargy import (
rule,
and_, or_, not_
)
from yargy.interpretation import fact
from yargy.predicates import (
caseless, normalized,
eq, length_eq,
gram, dictionary,
is_single, is_title
)
from yargy.relations import gnc_relation
Location = fact(
'Location',
['name'],
)
gnc = gnc_relation()
REGION = rule(
gram('ADJF').match(gnc),
dictionary({
'край',
'район',
'область',
'губерния',
'уезд',
}).match(gnc),
).interpretation(Location.name.inflected())