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_person():
Name = fact(
'Name',
['first', 'last'],
)
Person = fact(
'Person',
['position', 'name']
)
LAST = and_(
gram('Surn'),
not_(gram('Abbr')),
)
FIRST = and_(
gram('Name'),
not_(gram('Abbr')),
)
POSITION = morph_pipeline([
'управляющий директор',
'вице-мэр'
])
gnc = gnc_relation()
NAME = rule(
FIRST.interpretation(
Name.first
).match(gnc),
LAST.interpretation(
Name.last
).match(gnc)
).interpretation(
Name
)
PERSON = rule(
POSITION.interpretation(
Person.position
).match(gnc),
NAME.interpretation(
Person.name
)
).interpretation(
Person
)
parser = Parser(PERSON)
match = parser.match('управляющий директор Иван Ульянов')
assert match
assert match.fact == Person(
position='управляющий директор',
name=Name(
first='Иван',
last='Ульянов'
)
)
def test_pipeline_key():
from parser.parser 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_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_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_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_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_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_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_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_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_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_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 test_bnf():
from parser.parser import fact
from parser.parser 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'")
