spaCy/tests/test_matcher.py

52 lines
1.3 KiB
Python

from __future__ import unicode_literals
import pytest
from spacy.strings import StringStore
from spacy.matcher import *
from spacy.attrs import ORTH
from spacy.tokens.doc import Doc
from spacy.vocab import Vocab
@pytest.fixture
def matcher(EN):
specs = []
for string in ['JavaScript', 'Google Now', 'Java']:
spec = []
for orth_ in string.split():
spec.append([(ORTH, EN.vocab.strings[orth_])])
specs.append((spec, EN.vocab.strings['product']))
return Matcher(specs)
def test_compile(matcher):
assert matcher.n_patterns == 3
def test_no_match(matcher, EN):
tokens = EN('I like cheese')
assert matcher(tokens) == []
def test_match_start(matcher, EN):
tokens = EN('JavaScript is good')
assert matcher(tokens) == [(EN.vocab.strings['product'], 0, 1)]
def test_match_end(matcher, EN):
tokens = EN('I like Java')
assert matcher(tokens) == [(EN.vocab.strings['product'], 2, 3)]
def test_match_middle(matcher, EN):
tokens = EN('I like Google Now best')
assert matcher(tokens) == [(EN.vocab.strings['product'], 2, 4)]
def test_match_multi(matcher, EN):
tokens = EN('I like Google Now and Java best')
assert matcher(tokens) == [(EN.vocab.strings['product'], 2, 4),
(EN.vocab.strings['product'], 5, 6)]
def test_dummy():
pass