class Address(Enum):
# Садовая улица
AdjFull = [
{
'labels': [
gram('ADJF'),
gram_not('Abbr'),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Name,
},
},
{
'labels': [
gram('ADJF'),
gram_not('Abbr'),
gnc_match(-1, solve_disambiguation=True),
],
'repeatable':
True,
'optional':
True,
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Name,
},
},
{
'labels': [
dictionary(STREET_DESCRIPTOR_DICTIONARY),
gnc_match(-1, solve_disambiguation=True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Descriptor,
},
},
]
# улица Садовая
AdjFullReversed = [
{
'labels': [
dictionary(STREET_DESCRIPTOR_DICTIONARY),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Descriptor,
},
},
{
'labels': [
gram('ADJF'),
gram_not('Abbr'),
gnc_match(-1, solve_disambiguation=True),
],
'repeatable':
True,
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Name,
},
},
]
# ул. Садовая
AdjShort = SHORT_STREET_DESCRIPTOR_RULE + AdjFull[:2]
# Садовая ул.
AdjShortReversed = AdjFull[:2] + SHORT_STREET_DESCRIPTOR_RULE
# улица Красных Десантников
AdjNounFull = [AdjFullReversed[0]] + AdjFull[:2] + [{
'labels': [
gram('gent'),
gram_not('Abbr'),
gnc_match(-1, solve_disambiguation=True),
],
'repeatable':
True,
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Name,
},
}]
# ул. Красных Десантников
AdjNounShort = AdjShort + [AdjNounFull[-1]]
# улица Карла Маркса
GentFullReversed = [
AdjFullReversed[0],
{
'labels': [
gram('gent'),
gram_not('Abbr'),
],
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Name,
},
},
{
'labels': [
gram('gent'),
gram_not('Abbr'),
gnc_match(-1, solve_disambiguation=True)
],
'optional':
True,
'repeatable':
True,
'normalization':
NormalizationType.Original,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Name,
},
},
]
# улица К. Маркса
GentFullReversedWithShortcut = [
GentFullReversed[0],
{
'labels': [
gram('Abbr'),
],
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Name,
},
},
{
'labels': [
eq('.'),
],
'normalization': NormalizationType.Original,
},
] + GentFullReversed[1:]
# улица В. В. Ленина
GentFullReversedWithExtendedShortcut = GentFullReversedWithShortcut[:3] + GentFullReversedWithShortcut[
1:3] + GentFullReversedWithShortcut[3:]
# пр. Маршала жукова
GentShortReversed = SHORT_STREET_DESCRIPTOR_RULE + GentFullReversed[1:]
# пр. К. Маркса
GentShortReversedWithShortcut = SHORT_STREET_DESCRIPTOR_RULE + GentFullReversedWithShortcut[
1:]
# пл. В. В. Ленина
GentShortReversedWithExtendedShortcut = SHORT_STREET_DESCRIPTOR_RULE + GentFullReversedWithExtendedShortcut[
1:]
# Николая Ершова улица
GentFull = GentFullReversed[1:] + GentFullReversed[:1]
# Обуховской Обороны пр-кт
GentShort = GentShortReversed[2:] + SHORT_STREET_DESCRIPTOR_RULE
# 1-я новорублевская улица
AdjFullWithNumericPart = NUMERIC_STREET_PART_RULE + AdjFull
# улица 1-я новорублевская
AdjFullReversedWithNumericPart = AdjFullReversed[:
1] + AdjFullWithNumericPart[:
-1]
# 1-я новорублевская ул.
AdjShortWithNumericPart = AdjFullWithNumericPart[:-1] + SHORT_STREET_DESCRIPTOR_RULE
# ул. 1-я промышленная
AdjShortReversedWithNumericPart = SHORT_STREET_DESCRIPTOR_RULE + AdjFullWithNumericPart[:
-1]
# проспект 50 лет октября
GentFullReversedWithNumericPrefix = GentFullReversed[:1] + NUMERIC_STREET_PART_WITHOUT_SUFFIX_RULE + GentFullReversed[
1:2] + GentFullReversed[1:]
# пр-т. 50 лет советской власти
GentShortReversedWithNumericPrefix = GentShortReversed[:2] + NUMERIC_STREET_PART_WITHOUT_SUFFIX_RULE + GentFullReversed[
1:2] + GentFullReversed[1:]
# 2-ой проезд Перова Поля
GentNumericSplittedByFullDescriptor = NUMERIC_STREET_PART_RULE + GentFullReversed
# 7-я ул. текстильщиков
GentNumericSplittedByShortDescriptor = NUMERIC_STREET_PART_RULE + GentShortReversed
'''
Street names with house numbers
'''
# Зеленая улица, дом 7
AdjFullWithHn = AdjFull + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# улица Зеленая, дом 7
AdjFullReversedWithHn = AdjFullReversed + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# ул. Нижняя Красносельская дом 7
AdjShortWithHn = AdjShort + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# Настасьинский пер., дом 2
AdjShortReversedWithHn = AdjShortReversed + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# улица Красной Гвардии, дом 2
AdjNounFullWithHn = AdjNounFull + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# ул. Брянской пролетарской дивизии дом 2
AdjNounShortWithHn = AdjNounShort + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# Николая Ершова улица дом 1
GentFullWithHn = GentFull + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# улица Карла Маркса дом 1
GentFullReversedWithHn = GentFullReversed + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# улица К. Маркса, дом 1
GentFullReversedWithShortcutWithHn = GentFullReversedWithShortcut + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# улица В. И. Ленина, дом 1
GentFullReversedWithExtendedShortcutWithHn = GentFullReversedWithExtendedShortcut + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# Обуховской Обороны пр-кт дом 1
GentShortWithHn = GentShort + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# пр-кт Обуховской Обороны дом 1
GentShortReversedWithHn = GentShortReversed + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# ул. К. Маркса, дом 1
GentShortReversedWithShortcutWithHn = GentShortReversedWithShortcut + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# ул. В. И. Ленина, дом 1
GentShortReversedWithExtendedShortcutWithHn = GentShortReversedWithExtendedShortcut + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# 1-я новорублевская улица дом 1
AdjFullWithNumericPartWithHn = AdjFullWithNumericPart + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# улица 1-я новорублевская, дом 1
AdjFullReversedWithNumericPartWithHn = AdjFullReversedWithNumericPart + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# 1-я новорублевская ул. дом 1
AdjShortWithNumericPartWithHn = AdjShortWithNumericPart + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# ул. 1-я промышленная, дом 1
AdjShortReversedWithNumericPartWithHn = AdjShortReversedWithNumericPart + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# проспект 50 лет октября, дом 1
GentFullReversedWithNumericPrefixWithHn = GentFullReversedWithNumericPrefix + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# пр-т. 50 лет советской власти, дом 1
GentShortReversedWithNumericPrefixWithHn = GentShortReversedWithNumericPrefix + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# 2-ой проезд Перова Поля, дом 1
GentNumericSplittedByFullDescriptorWithHn = GentNumericSplittedByFullDescriptor + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
# 7-я ул. текстильщиков, дом 1
GentNumericSplittedByShortDescriptorWithHn = GentNumericSplittedByShortDescriptor + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR
'''
Street names with house numbers and letters
'''
# Зеленая улица, дом 7, лит А
AdjFullWithHnAndLetter = AdjFull + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# улица Зеленая, дом 7, лит А
AdjFullReversedWithHnAndLetter = AdjFullReversed + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# ул. Нижняя Красносельская дом 7, лит А
AdjShortWithHnAndLetter = AdjShort + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# Настасьинский пер., дом 2
AdjShortReversedWithHnAndLetter = AdjShortReversed + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# улица Красной Гвардии, дом 2
AdjNounFullWithHnAndLetter = AdjNounFull + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# ул. Брянской пролетарской дивизии дом 2
AdjNounShortWithHnAndLetter = AdjNounShort + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# Николая Ершова улица дом 1
GentFullWithHnAndLetter = GentFull + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# улица Карла Маркса дом 1
GentFullReversedWithHnAndLetter = GentFullReversed + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# улица К. Маркса, дом 1
GentFullReversedWithShortcutWithHnAndLetter = GentFullReversedWithShortcut + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# улица В. И. Ленина, дом 1
GentFullReversedWithExtendedShortcutWithHnAndLetter = GentFullReversedWithExtendedShortcut + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# Обуховской Обороны пр-кт дом 1
GentShortWithHnAndLetter = GentShort + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# пр-кт Обуховской Обороны дом 1
GentShortReversedWithHnAndLetter = GentShortReversed + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# ул. К. Маркса, дом 1
GentShortReversedWithShortcutWithHnAndLetter = GentShortReversedWithShortcut + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# ул. В. И. Ленина, дом 1
GentShortReversedWithExtendedShortcutWithHnAndLetter = GentShortReversedWithExtendedShortcut + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# 1-я новорублевская улица дом 1
AdjFullWithNumericPartWithHnAndLetter = AdjFullWithNumericPart + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# улица 1-я новорублевская, дом 1
AdjFullReversedWithNumericPartWithHnAndLetter = AdjFullReversedWithNumericPart + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# 1-я новорублевская ул. дом 1
AdjShortWithNumericPartWithHnAndLetter = AdjShortWithNumericPart + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# ул. 1-я промышленная, дом 1
AdjShortReversedWithNumericPartWithHnAndLetter = AdjShortReversedWithNumericPart + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# проспект 50 лет октября, дом 1
GentFullReversedWithNumericPrefixWithHnAndLetter = GentFullReversedWithNumericPrefix + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# пр-т. 50 лет советской власти, дом 1
GentShortReversedWithNumericPrefixWithHnAndLetter = GentShortReversedWithNumericPrefix + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# 2-ой проезд Перова Поля, дом 1
GentNumericSplittedByFullDescriptorWithHnAndLetter = GentNumericSplittedByFullDescriptor + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
# 7-я ул. текстильщиков, дом 1
GentNumericSplittedByShortDescriptorWithHnAndLetter = GentNumericSplittedByShortDescriptor + OPTIONAL_COMMA_GRAMMAR + HOUSE_NUMBER_GRAMMAR + HOUSE_LETTER_GRAMMAR
class Location(Enum):
FederalDistrict = [
{
'labels': [
gram('ADJF'),
dictionary(FEDERAL_DISTRICT_DICTIONARY),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Name,
},
},
{
'labels': [
gnc_match(-1, solve_disambiguation=True),
dictionary({
'федеральный',
}),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Descriptor,
},
},
{
'labels': [
gnc_match(-1, solve_disambiguation=True),
dictionary({
'округ',
}),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Descriptor,
},
},
]
FederalDistrictAbbr = [
{
'labels': [
gram('ADJF'),
dictionary(FEDERAL_DISTRICT_DICTIONARY),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Name,
},
},
{
'labels': [
eq('ФО'),
],
'interpretation': {
'attribute': LocationObject.Attributes.Descriptor,
},
},
]
AutonomousDistrict = [
{
'labels': [
gram('ADJF'),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Name,
},
},
{
'labels': [
gnc_match(-1, solve_disambiguation=True),
dictionary({
'автономный',
}),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Descriptor,
},
},
{
'labels': [
gnc_match(-1, solve_disambiguation=True),
dictionary({
'округ',
}),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Descriptor,
},
},
]
AutonomousDistrictAbbr = [
{
'labels': [
gram('ADJF'),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Name,
},
},
{
'labels': [
eq('АО'),
],
'interpretation': {
'attribute': LocationObject.Attributes.Descriptor,
},
},
]
Region = [
{
'labels': [
gram('ADJF'),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Name,
},
},
{
'labels': [
dictionary(REGION_TYPE_DICTIONARY),
gnc_match(-1, solve_disambiguation=True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Descriptor,
},
},
]
ComplexObject = [
{
'labels': [
gram('ADJF'),
dictionary(COMPLEX_OBJECT_PREFIX_DICTIONARY),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Name,
},
},
{
'labels': [
gram('NOUN'),
gram('Geox'),
gnc_match(-1, solve_disambiguation=True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Name,
},
},
]
PartialObject = [
{
'labels': [
gram('NOUN'),
dictionary(PARTIAL_OBJECT_PREFIX_DICTIONARY),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Name,
},
},
{
'labels': [
gram('NOUN'),
gram('Geox'),
gnc_match(-1, solve_disambiguation=True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Name,
},
},
]
# Донецкая народная республика / Российская Федерация
AdjfFederation = [
{
'labels': [
gram('ADJF'),
is_capitalized(True),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Name,
},
},
{
'labels': [
gram('ADJF'),
gnc_match(-1, solve_disambiguation=True),
],
'optional': True,
'repeatable': True,
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Name,
},
},
{
'labels': [
gnc_match(0, solve_disambiguation=True),
dictionary({
'федерация',
'республика',
'империя',
}),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Descriptor,
},
},
]
# Соединенные Штаты / Соединенные Штаты Америки
AdjxFederation = [{
'labels': [
gram('Adjx'),
is_capitalized(True),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Name,
},
}, {
'labels': [
gram('Adjx'),
gnc_match(-1, solve_disambiguation=True),
],
'optional':
True,
'repeatable':
True,
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Name,
},
}, {
'labels': [
gnc_match(0, solve_disambiguation=True),
dictionary({
'штат',
'эмират',
}),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Name,
},
}, {
'labels': [
gram('gent'),
],
'optional': True,
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Name,
},
}]
Object = [
{
'labels': [
is_capitalized(True),
gram('Geox'),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': LocationObject.Attributes.Name,
},
},
]
class ProbabilisticPerson(Enum):
'''
This grammars matches words that looks like (but may not to be) person names
Not included in natasha DEFAULT_GRAMMARS, but shows good result on factRuEval-16 testset
'''
FirstnameAndLastname = [
Person.Firstname.value[0],
POSSIBLE_LASTNAME_GRAMMAR,
]
InitialsAndLastname = Person.InitialsAndLastname.value[:4] + [
POSSIBLE_LASTNAME_GRAMMAR,
]
LastnameAndInitials = [POSSIBLE_LASTNAME_GRAMMAR
] + Person.InitialsAndLastname.value[:4]
FirstnameAsInitialsAndLastname = Person.InitialsAndLastname.value[:2] + [
POSSIBLE_LASTNAME_GRAMMAR,
]
LastnameAndfirstnameAsInitials = [
POSSIBLE_LASTNAME_GRAMMAR,
] + Person.InitialsAndLastname.value[:2]
# Джон Х. Доу
FirstnameAndMiddlenameAsInitialsWithLastname = FirstnameAndLastname[:1] + Person.InitialsAndLastname.value[
2:4] + [
POSSIBLE_LASTNAME_GRAMMAR,
]
FirstnameAndLastnameWithNobilityParticle = [
Person.Firstname.value[0],
{
'labels': [
dictionary(NAME_NOBILITY_PARTICLE_DICTIONARY),
],
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': PersonObject.Attributes.Lastname,
}
},
POSSIBLE_LASTNAME_GRAMMAR,
]
FirstnameAndLastnameWithPosition = Person.WithPosition.value[:-1] + [
FirstnameAndLastname[-1]
]
# Эрнесто «Че» Гевара
FirstnameAndLastnameWithQuotedNickname = Person.Firstname.value[:1] + Person.FirstnameAndLastnameWithQuotedNickname.value[
1:-1] + [
POSSIBLE_LASTNAME_GRAMMAR,
]
# С.П. фон Дервиз
InitialsAndLastnameWithNobilityParticle = InitialsAndLastname[:4] + [
FirstnameAndLastnameWithNobilityParticle[1],
POSSIBLE_LASTNAME_GRAMMAR,
]
# John S. Doe
Latin = [
{
'labels': [
gram('LATN'),
is_capitalized(True),
],
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': PersonObject.Attributes.Firstname,
}
},
{
'labels': [
gram('LATN'),
is_capitalized(True),
],
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': PersonObject.Attributes.Middlename,
}
},
{
'labels': [gram('PUNCT'), eq('.')],
},
{
'labels': [
gram('LATN'),
is_capitalized(True),
],
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': PersonObject.Attributes.Lastname,
}
},
]
'дора', # дорога, #88
}),
in_({
'пер', # переулок, #88,
'н', # набережная
}),
)),
},
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Descriptor,
},
},
{
'labels': [
eq('.'),
],
'optional': True,
'normalization': NormalizationType.Original,
}
]
NUMERIC_STREET_PART_RULE = [ # 1-я, 10-й, 100500-ой и т.д.
{
'labels': [
gram('INT'),
gte(1),
],
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Name,
class Person(Enum):
# Иванов Иван Иванович
Full = [
{
'labels': [
gram('Surn'),
gram_not('Abbr'),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Lastname,
},
},
{
'labels': [
gram('Name'),
gram_not('Abbr'),
gnc_match(-1, solve_disambiguation=True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Firstname,
},
},
{
'labels': [
gram('Patr'),
gram_not('Abbr'),
gnc_match(-1, solve_disambiguation=True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Middlename,
},
},
]
# Иван Иванович Иванов
FullReversed = [
{
'labels': [
gram('Name'),
gram_not('Abbr'),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Firstname,
},
},
{
'labels': [
gram('Patr'),
gram_not('Abbr'),
gnc_match(-1, solve_disambiguation=True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Middlename,
},
},
{
'labels': [
gram('Surn'),
gram_not('Abbr'),
gnc_match(-1, solve_disambiguation=True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Lastname,
},
},
]
# Фелипе Родригес Фернандес
# https://www.englishelp.ru/business-english/other/284-patronymic-vs-middle-name.html
FullReversedWithLatinMiddlename = [
{
'labels': [
gram('Name'),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Firstname,
},
},
{
'labels': [
gram('Name'),
gnc_match(-1, solve_disambiguation=True),
],
'repeatable': True,
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Middlename,
},
},
{
'labels': [
gram('Surn'),
gnc_match(0, solve_disambiguation=True),
gnc_match(-1, solve_disambiguation=True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Lastname,
},
},
]
# Л. А. Раневская
InitialsAndLastname = [
{
'labels': [
gram_in(['Name', 'Abbr']),
],
'interpretation': {
'attribute': PersonObject.Attributes.Firstname,
},
'normalization': NormalizationType.Original,
},
{
'labels': [
gram('PUNCT'),
eq('.'),
],
},
{
'labels': [
gram_in(['Patr', 'Abbr']),
gnc_match(0, solve_disambiguation=True),
],
'interpretation': {
'attribute': PersonObject.Attributes.Middlename,
},
'normalization':
NormalizationType.Original,
},
{
'labels': [
gram('PUNCT'),
eq('.'),
],
},
{
'labels': [
gram('Surn'),
gram_not('Abbr'),
gnc_match(0, solve_disambiguation=True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Lastname,
},
},
]
# Иван Иванов
FirstnameAndLastname = [
{
'labels': [
gram('Name'),
gram_not('Abbr'),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Firstname,
},
},
{
'labels': [
gram('Surn'),
gram_not('Abbr'),
gnc_match(-1, solve_disambiguation=True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Lastname,
},
},
]
FirstnameAndLastnameWithQuotedNickname = [
FirstnameAndLastname[0], {
'labels': {
gram('QUOTE'),
gram_any({
'G-QUOTE',
'L-QUOTE',
}),
},
'normalization': NormalizationType.Original,
}, {
'labels': {
gram_not_in({
'QUOTE',
'PUNCT',
}),
},
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': PersonObject.Attributes.Nickname,
},
}, {
'labels': {
gram('QUOTE'),
},
'normalization': NormalizationType.Original,
}, InitialsAndLastname[-1]
]
# Иванов Иван
LastnameAndFirstname = [
{
'labels': [
gram('Surn'),
gram_not('Abbr'),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Lastname,
},
},
{
'labels': [
gram('Name'),
gram_not('Abbr'),
gnc_match(-1, solve_disambiguation=True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Firstname,
},
},
]
# Александр Ф. Скляр
FullReversedWithMiddlenameAsInitials = FullReversed[:
1] + InitialsAndLastname[
3:]
# Раневская Л. А.
LastnameAndInitials = [
LastnameAndFirstname[0],
] + InitialsAndLastname[:4]
# Раневская Л.
LastnameAndFirstnameAsInitials = [
LastnameAndFirstname[0],
] + InitialsAndLastname[:2]
# Л. Раневская
FirstnameAsInitialsAndLastname = InitialsAndLastname[:2] + [
InitialsAndLastname[-1],
]
# Иван Иванович
FirstnameAndMiddlename = [
{
'labels': [
gram('Name'),
gram_not('Abbr'),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Firstname,
},
},
{
'labels': [
gram('Patr'),
gram_not('Abbr'),
gnc_match(-1, solve_disambiguation=True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Middlename,
},
},
]
# Иванов
Lastname = [
{
'labels': [
gram('Surn'),
gram_any({
'sing',
'Stgm',
}),
gram_not('Abbr'),
is_capitalized(True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Lastname,
},
},
]
# Иванович
Middlename = [{
'labels': [
gram('Patr'),
gram_any({
'sing',
'Stgm',
}),
gram_not('Abbr'),
is_capitalized(True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Middlename,
},
}]
# Иван
Firstname = [
{
'labels': [
gram('Name'),
gram_any({
'sing',
'Stgm',
}),
gram_not('Abbr'),
is_capitalized(True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Firstname,
},
},
]
# Отто фон Бисмарк
FirstnameAndLastnameWithNobilityParticle = [
FullReversed[0],
{
'labels': [
dictionary(NAME_NOBILITY_PARTICLE_DICTIONARY),
],
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': PersonObject.Attributes.Lastname,
}
},
{
'labels': [
gram('Surn'),
gram_not('Abbr'),
gnc_match(0, solve_disambiguation=True)
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Lastname,
},
},
]
# Премьер-министр РФ Дмитрий Медведев
WithPosition = [
{
'labels': [
gram('Person/Position'),
],
'interpretation': {
'attribute': PersonObject.Attributes.Descriptor,
},
},
{
'labels': [
or_((
and_((
or_((
gram_any({
'ablt',
'loct',
'gent',
}),
gram('Fixd'),
)),
gram_not_in({
'Name',
'Patr',
'Surn',
}),
)),
gram('Abbr'),
gram('LATN'),
)),
],
'optional':
True,
'repeatable':
True,
'normalization':
NormalizationType.Original,
},
{
'labels': [
gram('Name'),
case_match(0, solve_disambiguation=True),
number_match(0, solve_disambiguation=True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Firstname,
},
},
{
'labels': [
gram('Patr'),
case_match(0, solve_disambiguation=True),
number_match(0, solve_disambiguation=True),
gnc_match(-1, solve_disambiguation=True),
],
'optional':
True,
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Middlename,
},
},
{
'labels': [
gram('Surn'),
case_match(0, solve_disambiguation=True),
number_match(0, solve_disambiguation=True),
gnc_match(-1, solve_disambiguation=True),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': PersonObject.Attributes.Lastname,
},
},
]
# Пресс-секретарь «Роснефти» Михаил Леонтьев
WithPositionAndQuotedOrganisationName = (WithPosition[:2] + [
{
'labels': [
gram('QUOTE'),
],
'normalization': NormalizationType.Original,
},
{
'labels': [
gram_not('QUOTE'),
gram_not_in('END-OF-LINE'),
],
'repeatable': True,
'normalization': NormalizationType.Original,
},
{
'labels': [
gram('QUOTE'),
],
'normalization': NormalizationType.Original,
},
] + WithPosition[2:])
# граф де Кристо
PositionAndNobilitySurname = [{
'labels': [
gram('Person/Position'),
],
}, {
'labels': [
dictionary(NAME_NOBILITY_PARTICLE_DICTIONARY),
],
}, {
'labels': [
gram('Surn'),
gnc_match(0, solve_disambiguation=True),
],
'interpretation': {
'attribute': PersonObject.Attributes.Lastname,
},
}]
# Генрих Восьмой / Карл XII
NameWithNumericPart = [{
'labels': [
gram_in({
'Name',
'sing',
}),
gram_not('Abbr'),
],
'interpretation': {
'attribute': PersonObject.Attributes.Firstname,
},
}, {
'labels': [
or_((
and_((gram_in({
'ADJF',
'Anum',
}), gnc_match(-1, solve_disambiguation=True))),
gram('ROMN'),
))
]
}]
class Organisation(Enum):
OfficialQuoted = [
{
'labels': [
or_((
gram('Orgn/Commercial'),
gram('Orgn/Social'),
gram('Orgn/Abbr'),
)),
],
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': OrganisationObject.Attributes.Descriptor,
},
},
{
'labels': [
is_abbr(True),
],
'optional': True,
'repeatable': True,
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': OrganisationObject.Attributes.Descriptor,
},
},
{
'labels': [
gram('QUOTE'),
],
},
{
'labels': [
gram_not('QUOTE'),
gram_not_in('END-OF-LINE'),
not_eq('.'),
],
'repeatable': True,
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': OrganisationObject.Attributes.Name,
},
},
{
'labels': [
gram('QUOTE'),
],
},
]
Abbr = [
{
'labels': [
gram('Abbr'),
gram('Orgn'),
gram_not('Orgn/Abbr'),
],
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': OrganisationObject.Attributes.Name,
},
},
]
IndividualEntrepreneur = [
{
'labels': [
eq('ИП'),
],
'normalization': NormalizationType.Original,
},
Person.Full.value[0],
Person.Full.value[1],
Person.Full.value[2],
]
SimpleLatin = [
{
'labels': [
gram('Orgn/Commercial'),
],
'normalization': NormalizationType.Normalized,
'interpretation': {
'attribute': OrganisationObject.Attributes.Descriptor,
},
},
{
'labels': [
gram_any({
'LATN',
'NUMBER',
}),
],
'repeatable': True,
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': OrganisationObject.Attributes.Name,
},
},
]
# Санкт-Петербургский Государственный университет
Educational = [{
'labels': [
gram('ADJF'),
is_capitalized(True),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': OrganisationObject.Attributes.Name,
},
}, {
'labels': [
gram('ADJF'),
gnc_match(-1, solve_disambiguation=True),
],
'optional':
True,
'repeatable':
True,
'normalization':
NormalizationType.Inflected,
'interpretation': {
'attribute': OrganisationObject.Attributes.Name,
},
}, {
'labels': [
gram('Orgn/Educational'),
gnc_match(0, solve_disambiguation=True),
gnc_match(-1, solve_disambiguation=True),
],
'normalization':
NormalizationType.Normalized,
'interpretation': {
'attribute': OrganisationObject.Attributes.Descriptor,
},
}, {
'labels': [
gram_any({
'ablt',
'gent',
}),
gram_not_in({
'PREP',
}),
dictionary_not({
'имя',
}),
],
'normalization':
NormalizationType.Original,
'interpretation': {
'attribute': OrganisationObject.Attributes.Name,
},
'optional':
True,
'repeatable':
True,
}]
# Публичная библиотека имени М. Е. Салтыкова-Щедрина
EducationalWithInitials = Educational + NAMED_ORG_INITIALS_PREFIX_RULE + NAMED_ORG_INITIALS_RULE
# Публичная библиотека имени Салтыкова-Щедрина
EducationalWithLastname = Educational + NAMED_ORG_INITIALS_PREFIX_RULE + LASTNAME_GRAMMAR
# Кировский завод
AdjCommercial = Educational[:2] + [
{
'labels': [
gram('Orgn/Commercial'),
gnc_match(0, solve_disambiguation=True),
gnc_match(-1, solve_disambiguation=True),
],
'normalization':
NormalizationType.Normalized,
'interpretation': {
'attribute': OrganisationObject.Attributes.Descriptor,
},
},
Educational[-1],
]
AdjCommercialWithInitials = AdjCommercial + NAMED_ORG_INITIALS_PREFIX_RULE + NAMED_ORG_INITIALS_RULE
AdjCommercialWithLastname = AdjCommercial + NAMED_ORG_INITIALS_PREFIX_RULE + LASTNAME_GRAMMAR
# Общества андрологии и сексуальной медицины
Social = [
{
'labels': [
gram('Orgn/Social'),
gram('sing'),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': OrganisationObject.Attributes.Descriptor,
},
},
{
'labels': [
gram_not_in({
'PREP',
'CONJ',
}),
gram_any({
'accs',
'datv',
'gent',
}),
],
'normalization':
NormalizationType.Original,
'interpretation': {
'attribute': OrganisationObject.Attributes.Name,
},
},
{
'labels': [
gram_any({
'gent',
'accs',
'ablt',
}),
gram_not_in({
'PREP',
'Name',
'Patr',
'Surn',
}),
],
'optional':
True,
'repeatable':
True,
'normalization':
NormalizationType.Original,
'interpretation': {
'attribute': OrganisationObject.Attributes.Name,
},
},
]
SocialWithInitials = Social + NAMED_ORG_INITIALS_PREFIX_RULE + NAMED_ORG_INITIALS_RULE
SocialWithLastname = Social + NAMED_ORG_INITIALS_PREFIX_RULE + LASTNAME_GRAMMAR
AdjSocial = [
{
'labels': [
gram('ADJF'),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': OrganisationObject.Attributes.Name,
},
},
{
'labels': [
gram('ADJF'),
gnc_match(-1, solve_disambiguation=True),
],
'optional': True,
'repeatable': True,
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': OrganisationObject.Attributes.Name,
},
},
{
'labels': [
gram('Orgn/Social'),
gnc_match(0, solve_disambiguation=True),
gnc_match(-1, solve_disambiguation=True),
],
'interpretation': {
'attribute': OrganisationObject.Attributes.Descriptor,
},
},
Social[-1],
]
AdjSocialWithInitials = AdjSocial + NAMED_ORG_INITIALS_PREFIX_RULE + NAMED_ORG_INITIALS_RULE
AdjSocialWithLastname = AdjSocial + NAMED_ORG_INITIALS_PREFIX_RULE + LASTNAME_GRAMMAR
type_required,
string_type,
)
from yargy.compat import string_type
from yargy.parser import OR
from yargy.normalization import NormalizationType
from natasha.grammars import Person
from natasha.grammars.organisation.interpretation import OrganisationObject
NAMED_ORG_INITIALS_PREFIX_RULE = [
OR(
[
{
'labels': [
eq('им'), # имени
],
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': OrganisationObject.Attributes.Name,
},
},
{
'labels': [
eq('.'),
],
'normalization': NormalizationType.Original,
}
],
[{
'labels': [
class Street(Enum):
# Садовая улица
AdjFull = [
{
'labels': [
gram('ADJF'),
gram_not('Abbr'),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Name,
},
},
{
'labels': [
gram('ADJF'),
gram_not('Abbr'),
gnc_match(-1, solve_disambiguation=True),
],
'repeatable': True,
'optional': True,
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Name,
},
},
{
'labels': [
dictionary(STREET_DESCRIPTOR_DICTIONARY),
gnc_match(-1, solve_disambiguation=True),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Descriptor,
},
},
]
# улица Садовая
AdjFullReversed = [
{
'labels': [
dictionary(STREET_DESCRIPTOR_DICTIONARY),
],
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Descriptor,
},
},
{
'labels': [
gram('ADJF'),
gram_not('Abbr'),
gnc_match(-1, solve_disambiguation=True),
],
'repeatable': True,
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Name,
},
},
]
# ул. Садовая
AdjShort = SHORT_STREET_DESCRIPTOR_RULE + AdjFull[:2]
# Садовая ул.
AdjShortReversed = AdjFull[:2] + SHORT_STREET_DESCRIPTOR_RULE
# улица Красных Десантников
AdjNounFull = [AdjFullReversed[0]] + AdjFull[:2] + [
{
'labels': [
gram('gent'),
gram_not('Abbr'),
gnc_match(-1, solve_disambiguation=True),
],
'repeatable': True,
'normalization': NormalizationType.Inflected,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Name,
},
}
]
# ул. Красных Десантников
AdjNounShort = AdjShort + [
AdjNounFull[-1]
]
# улица Карла Маркса
GentFullReversed = [
AdjFullReversed[0],
{
'labels': [
gram('gent'),
gram_not('Abbr'),
],
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Name,
},
},
{
'labels': [
gram('gent'),
gram_not('Abbr'),
gnc_match(-1, solve_disambiguation=True)
],
'optional': True,
'repeatable': True,
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Name,
},
},
]
# улица К. Маркса
GentFullReversedWithShortcut = [
GentFullReversed[0],
{
'labels': [
gram('Abbr'),
],
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Name,
},
},
{
'labels': [
eq('.'),
],
'normalization': NormalizationType.Original,
},
] + GentFullReversed[1:]
# улица В. В. Ленина
GentFullReversedWithExtendedShortcut = GentFullReversedWithShortcut[:3] + GentFullReversedWithShortcut[1:3] + GentFullReversedWithShortcut[3:]
# пр. Маршала жукова
GentShortReversed = SHORT_STREET_DESCRIPTOR_RULE + GentFullReversed[1:]
# пр. К. Маркса
GentShortReversedWithShortcut = SHORT_STREET_DESCRIPTOR_RULE + GentFullReversedWithShortcut[1:]
# пл. В. В. Ленина
GentShortReversedWithExtendedShortcut = SHORT_STREET_DESCRIPTOR_RULE + GentFullReversedWithExtendedShortcut[1:]
# Николая Ершова улица
GentFull = GentFullReversed[1:] + GentFullReversed[:1]
# Обуховской Обороны пр-кт
GentShort = GentShortReversed[2:] + SHORT_STREET_DESCRIPTOR_RULE
# 1-я новорублевская улица
AdjFullWithNumericPart = NUMERIC_STREET_PART_RULE + AdjFull
# улица 1-я новорублевская
AdjFullReversedWithNumericPart = AdjFullReversed[:1] + AdjFullWithNumericPart[:-1]
# 1-я новорублевская ул.
AdjShortWithNumericPart = AdjFullWithNumericPart[:-1] + SHORT_STREET_DESCRIPTOR_RULE
# ул. 1-я промышленная
AdjShortReversedWithNumericPart = SHORT_STREET_DESCRIPTOR_RULE + AdjFullWithNumericPart[:-1]
# проспект 50 лет октября
GentFullReversedWithNumericPrefix = GentFullReversed[:1] + NUMERIC_STREET_PART_WITHOUT_SUFFIX_RULE + GentFullReversed[1:2] + GentFullReversed[1:]
# пр-т. 50 лет советской власти
GentShortReversedWithNumericPrefix = GentShortReversed[:2] + NUMERIC_STREET_PART_WITHOUT_SUFFIX_RULE + GentFullReversed[1:2] + GentFullReversed[1:]
# 2-ой проезд Перова Поля
GentNumericSplittedByFullDescriptor = NUMERIC_STREET_PART_RULE + GentFullReversed
# 7-я ул. текстильщиков
GentNumericSplittedByShortDescriptor = NUMERIC_STREET_PART_RULE + GentShortReversed
'дора', # дорога, #88
}),
in_({
'пер', # переулок, #88,
'н', # набережная
}),
)),
},
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Descriptor,
},
},
{
'labels': [
eq('.'),
],
'optional': True,
'normalization': NormalizationType.Original,
}
]
NUMERIC_STREET_PART_RULE = [ # 1-я, 10-й, 100500-ой и т.д.
{
'labels': [
gram('INT'),
gte(1),
],
'normalization': NormalizationType.Original,
'interpretation': {
'attribute': AddressObject.Attributes.Street_Name,