spaCy/spacy/tests/regression/test_issue6207.py

19 lines
425 B
Python
Raw Normal View History

# coding: utf8
from __future__ import unicode_literals
from spacy.util import filter_spans
def test_issue6207(en_tokenizer):
doc = en_tokenizer("zero one two three four five six")
# Make spans
s1 = doc[:4]
s2 = doc[3:6] # overlaps with s1
s3 = doc[5:7] # overlaps with s2, not s1
result = filter_spans((s1, s2, s3))
assert s1 in result
assert s2 not in result
assert s3 in result