spaCy/tests/test_contractions.py
2014-07-07 05:07:21 +02:00

45 lines
1.1 KiB
Python

from __future__ import unicode_literals
from spacy.spacy import expand_chunk
from spacy.en import lookup, unhash
from spacy import lex_of
def test_possess():
tokens = expand_chunk(lookup("Mike's"))
assert len(tokens) == 2
assert unhash(lex_of(tokens[0])) == "Mike"
assert unhash(lex_of(tokens[1])) == "'s"
def test_apostrophe():
tokens = expand_chunk(lookup("schools'"))
assert len(tokens) == 2
assert unhash(lex_of(tokens[1])) == "'"
assert unhash(lex_of(tokens[0])) == "schools"
def test_LL():
tokens = expand_chunk(lookup("we'll"))
assert len(tokens) == 2
assert unhash(lex_of(tokens[1])) == "will"
assert unhash(lex_of(tokens[0])) == "we"
def test_aint():
tokens = expand_chunk(lookup("ain't"))
assert len(tokens) == 2
assert unhash(lex_of(tokens[0])) == "are"
assert unhash(lex_of(tokens[1])) == "not"
def test_capitalized():
tokens = expand_chunk(lookup("can't"))
assert len(tokens) == 2
tokens = expand_chunk(lookup("Can't"))
assert len(tokens) == 2
tokens = expand_chunk(lookup("Ain't"))
assert len(tokens) == 2
assert unhash(lex_of(tokens[0])) == "Are"