explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-01-13 10:46:29 +03:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'master' of github.com:spacy-io/spaCy

Browse Source
This commit is contained in:
Henning Peters 2016-04-19 19:50:55 +02:00
commit c356251f45
1 changed files with 28 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
spacy/tests/matcher/test_matcher_bugfixes.py
View File

@ -1,8 +1,11 @@
import pytest import pytest
import numpy import numpy
import os
import spacy
from spacy.matcher import Matcher from spacy.matcher import Matcher
from spacy.attrs import ORTH, LOWER, ENT_IOB, ENT_TYPE from spacy.attrs import ORTH, LOWER, ENT_IOB, ENT_TYPE
from spacy.attrs import ORTH, TAG, LOWER, IS_ALPHA, FLAG63
from spacy.symbols import DATE from spacy.symbols import DATE
@ -31,6 +34,31 @@ def test_overlap_issue118(EN):
assert ents[0].end == 11 assert ents[0].end == 11
def test_overlap_issue242():
'''Test overlapping multi-word phrases.'''
patterns = [
[{LOWER: 'food'}, {LOWER: 'safety'}],
[{LOWER: 'safety'}, {LOWER: 'standards'}],
]
if os.environ.get('SPACY_DATA'):
data_dir = os.environ.get('SPACY_DATA')
else:
data_dir = None
nlp = spacy.en.English(data_dir=data_dir, tagger=False, parser=False, entity=False)
nlp.matcher.add('FOOD', 'FOOD', {}, patterns)
doc = nlp.tokenizer(u'There are different food safety standards in different countries.')
food_safety, safety_standards = nlp.matcher(doc)
assert food_safety[1] == 3
assert food_safety[2] == 5
assert safety_standards[1] == 4
assert safety_standards[2] == 6
def test_overlap_reorder(EN): def test_overlap_reorder(EN):
'''Test order dependence''' '''Test order dependence'''
doc = EN.tokenizer(u'how many points did lebron james score against the boston celtics last night') doc = EN.tokenizer(u'how many points did lebron james score against the boston celtics last night')