def req_deverbal(require_deverbal_noun: str = "?"):
if require_deverbal_noun == "1": ## strictly deverbal noun
return y.and_(yp.gram("NOUN"), yp.in_caseless(deverbal_nouns))
elif require_deverbal_noun == "0": ## strictly regular verb
return y.or_(yp.gram("VERB"), yp.gram("INFN"))
elif require_deverbal_noun == "?": ## anything
return y.or_(
y.and_(yp.gram("NOUN"), yp.in_caseless(deverbal_nouns)),
yp.gram("VERB"),
yp.gram("INFN"),
)
else:
raise ValueError("Incorrect deverbal status")
)
DASH = eq('-')
DOT = eq('.')
ADJF = gram('ADJF')
NOUN = gram('NOUN')
INT = type('INT')
TITLE = is_title()
ANUM = rule(
INT,
DASH.optional(),
in_caseless({
'я', 'й', 'е',
'ое', 'ая', 'ий', 'ой'
})
)
#########
#
# STRANA
#
##########
# TODO
COUNTRY_VALUE = dictionary({
'россия',
'украина'
'сутки',
'смена',
'm',
'month',
'м'
'месяц',
'мес',
'y',
'year',
'г',
'год',
}
PERIOD = dictionary(PERIODS)
PER = or_(eq('/'), in_caseless({'в', 'за', 'per'}))
RATE = rule(PER, PERIOD.interpretation(Money.period))
MONEY = rule(
or_(
in_({
'•', ':', '`', '~', '*', '-', '–', '—', ';', '.', '(', 'от', 'from'
}),
type('RU'),
type('LATIN'),
).optional(),
CURRENCY.interpretation(Money.currency).optional(),
eq('+').optional(),
eq('*').optional(),
INTEGER.interpretation(Money.integer_min.custom(normalize_integer)),
MULTIPLIER = or_(MILLIARD, MILLION, THOUSAND).interpretation(Money.multiplier)
########
#
# NUMERAL
#
#######
NUMR = or_(
gram('NUMR'),
# https://github.com/OpenCorpora/opencorpora/issues/818
dictionary({'ноль', 'один'}),
)
# TODO: можно выпилить дробные части для снижения числа ложных срабатываний, их все равно не бывает в реальных вилках
# Хотя одна вакаха в Tampere University of Technology реально была с дробями
MODIFIER = in_caseless({'целых', 'сотых', 'десятых'})
PART = or_(rule(or_(INT, NUMR, MODIFIER)), MILLIARD, MILLION, THOUSAND,
CURRENCY, COINS_CURRENCY)
# TODO: вот здесь можно поправить, чтобы телефоны не парсились
BOUND = in_('()//')
NUMERAL = rule(BOUND, PART.repeatable(), BOUND)
#######
#
# AMOUNT
#
########
RIGHT_QUOTE = in_("»" + __quotes)
STATION = rule(
STATION_WORD.optional(),
METRO_WORD.optional(),
LEFT_QUOTE.optional(),
STATION_TITLE.interpretation(
meaning.custom(lambda p: p.value)).interpretation(Station.name),
rule(
eq('-').optional(),
LIST_OF_NUMERALS.interpretation(Station.num),
).optional(),
RIGHT_QUOTE.optional(),
).interpretation(Station)
LIST_OF_STATIONS = rule(
STATION.means(Array.element),
rule(
in_caseless('и,-'),
STATION.means(Array.element),
).repeatable().optional(),
).interpretation(Array).interpretation(meaning.custom(lambda p: p.element))
FROM_STATION_TO_STATION = rule(
or_(caseless('с'), caseless('со')),
STATION.means(Array.element),
caseless('на'),
STATION.means(Array.element
), # todo LIST_OF_STATIONS: со спасской на садовую и сенную
).interpretation(Array).interpretation(meaning.custom(lambda p: p.element))
class Building(Building):
value = value('buildingName')
class AddrPart(AddrPart):
@property
def obj(self):
from natasha import obj
part = self.value
return obj.AddrPart(part.value, part.type)
INT = type('INT')
LETTER = in_caseless(set('абвгдежзиклмнопрстуфхшщэюя'))
TYPE_CITY = dictionary({'город'}).interpretation(City.typeCity)
STRUCTURE_TYPE = dictionary({'строение',
'ст'}).interpretation(Structure.structureType)
TYPE_APPART = dictionary({'квартира'}).interpretation(Appart.typeAppart)
BUILDING_TYPE = dictionary({'дом', 'шоссе', 'проспект',
'улица'}).interpretation(Building.buildingType)
VALUE = rule(INT, LETTER.optional())
SEP = in_(r'/\-')
from yargy.tokenizer import QUOTES
INT = type('INT')
DOT = eq('.')
ADJF = gram('ADJF')
NOUN = gram('NOUN')
TITLE = is_title()
DASH = eq('-')
SLASH = eq('/')
QUOTE = in_(QUOTES)
ANUM = rule(
INT,
DASH.optional(),
in_caseless({
'я', 'й', 'е',
'ое', 'ая', 'ий', 'ой'
})
)
def value(key):
@property
def field(self):
return getattr(self, key)
return field
OnlyNameStreet = fact(
'OnlyNameStreet',
['name']
)
class OnlyNameStreet(OnlyNameStreet):
MULTIPLIER = or_(MILLIARD, MILLION, THOUSAND).interpretation(Money.multiplier)
########
#
# NUMERAL
#
#######
NUMR = or_(
gram('NUMR'),
# https://github.com/OpenCorpora/opencorpora/issues/818
dictionary({'ноль', 'один'}),
)
MODIFIER = in_caseless({'целых', 'сотых', 'десятых'})
PART = or_(rule(or_(INT, NUMR, MODIFIER)), MILLIARD, MILLION, THOUSAND,
CURRENCY, COINS_CURRENCY)
BOUND = in_('()//')
NUMERAL = rule(BOUND, PART.repeatable(), BOUND)
#######
#
# AMOUNT
#
########
return obj.AddrPart(part.value, part.type)
DASH = eq('-')
DOT = eq('.')
ADJF = gram('ADJF')
NOUN = gram('NOUN')
INT = type('INT')
TITLE = is_title()
ANUM = rule(
INT,
DASH.optional(),
in_caseless({
'я', 'й', 'е',
'ое', 'ая', 'ий', 'ой'
})
)
#########
#
# STRANA
#
##########
# TODO
COUNTRY_VALUE = dictionary({
'россия',
'украина'
from yargy import or_, rule
from yargy.interpretation import attribute, fact
import yargy.interpretation as meaning
from yargy.predicates import caseless, gram, in_caseless, normalized
from .station import FROM_STATION_TO_STATION, LIST_OF_STATIONS, STATION
Transfer = fact('Transfer', [attribute('to', default=[])])
TRANSFER = rule(
gram('ADJF').optional(), # пешеходный
normalized('переход'),
or_(
FROM_STATION_TO_STATION.interpretation(Transfer.to),
rule(
or_(caseless('на'), caseless('между'), caseless('с')).optional(),
LIST_OF_STATIONS.interpretation(Transfer.to)),
).optional(),
).interpretation(Transfer)
StationAndTransfer = fact('StationAndTransfer', ['station', 'transfer'])
STATION_AND_TRANSFER = rule(
STATION.interpretation(StationAndTransfer.station),
rule(
in_caseless('и,'),
TRANSFER.interpretation(meaning.custom(lambda p: p.to)).interpretation(
StationAndTransfer.transfer),
).optional()).interpretation(StationAndTransfer)
#
#######
NUMR = or_(
gram('NUMR'),
# https://github.com/OpenCorpora/opencorpora/issues/818
dictionary({
'ноль',
'один'
}),
)
MODIFIER = in_caseless({
'целых',
'сотых',
'десятых'
})
PART = or_(
rule(
or_(
INT,
NUMR,
MODIFIER
)
),
MILLIARD,
MILLION,
THOUSAND,
CURRENCY,
Nums = fact('Nums', [attribute('values').repeatable()])
__literals = {
'один': 1,
'два': 2,
'три': 3,
'четыре': 4,
'пять': 5,
'шесть': 6,
'семь': 7,
'восемь': 8,
'девять': 9,
}
LITERAL = dictionary(__literals).means(
interp.normalized().custom(__literals.get))
CONJ_NUMS = in_caseless('-и,')
NUMERAL = or_(*[eq(str(i)) for i in __literals.values()]).means(interp.custom(int))
# вестибюль 1 и 2
LIST_OF_NUMERALS = connect(NUMERAL.means(Nums.values), CONJ_NUMS) \
.means(Nums).means(meaning.custom(lambda p: list(sorted(set(p.values)))))
# первый и второй вестибюли
LIST_OF_LITERALS = connect(LITERAL.means(Nums.values), CONJ_NUMS) \
.means(Nums).means(meaning.custom(lambda p: list(sorted(set(p.values)))))
PREP = gram('PREP')
CONJ = gram('CONJ')
# А творение ниже сделано для критикалов и береговой
SIMPLE_WILDCARD = rule(
NOUN.repeatable().optional(),
ADJF.optional(),
INT.optional()
)
TITLE = is_title()
ANUM = rule(
INT,
DASH.optional(),
in_caseless({
'я', 'й', 'е',
'ое', 'ая', 'ий', 'ой'
})
)
# # # # # # # # # Адрес # # # # # # # # # # #
Address = fact(
'Address',
['City', 'Street',
'House', 'Building',
'Appartment']
)
# # # # # # # # # Города # # # # # # # # # # #
# Сначала беру готовые списки названий городов + добавляю пару своих