mirror of
https://github.com/explosion/spaCy.git
synced 2024-12-25 01:16:28 +03:00
15 lines
421 B
Python
15 lines
421 B
Python
# coding: utf-8
|
|
from __future__ import unicode_literals
|
|
|
|
import pytest
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"text,norms,lemmas",
|
|
[("ім.", ["імені"], ["ім'я"]), ("проф.", ["професор"], ["професор"])],
|
|
)
|
|
def test_uk_tokenizer_abbrev_exceptions(uk_tokenizer, text, norms, lemmas):
|
|
tokens = uk_tokenizer(text)
|
|
assert len(tokens) == 1
|
|
assert [token.norm_ for token in tokens] == norms
|