mirror of
https://github.com/explosion/spaCy.git
synced 2025-10-31 16:07:41 +03:00
21 lines
652 B
Python
21 lines
652 B
Python
import pytest
|
|
|
|
|
|
from spacy.matcher import Matcher
|
|
|
|
@pytest.mark.xfail
|
|
def test_overlap_issue118(EN):
|
|
'''Test a bug that arose from having overlapping matches'''
|
|
doc = EN.tokenizer(u'how many points did lebron james score against the boston celtics last night')
|
|
ORG = doc.vocab.strings['ORG']
|
|
matcher = Matcher(EN.vocab, {'BostonCeltics': ('ORG', {}, [[{'lower': 'boston'}, {'lower': 'celtics'}], [{'lower': 'celtics'}]])})
|
|
|
|
matches = matcher(doc)
|
|
assert matches == [(ORG, 9, 11)]
|
|
ents = list(doc.ents)
|
|
assert len(ents) == 1
|
|
assert ents[0].label == ORG
|
|
assert ents[0].start == 9
|
|
assert ents[0].end == 11
|
|
|