mirror of
https://github.com/explosion/spaCy.git
synced 2024-12-25 09:26:27 +03:00
16 lines
451 B
Python
16 lines
451 B
Python
# coding: utf-8
|
|
from __future__ import unicode_literals
|
|
|
|
import pytest
|
|
|
|
@pytest.mark.parametrize('text', ["ca.", "m.a.o.", "Jan.", "Dec."])
|
|
def test_da_tokenizer_handles_abbr(da_tokenizer, text):
|
|
tokens = da_tokenizer(text)
|
|
assert len(tokens) == 1
|
|
|
|
def test_da_tokenizer_handles_exc_in_text(da_tokenizer):
|
|
text = "Det er bl.a. ikke meningen"
|
|
tokens = da_tokenizer(text)
|
|
assert len(tokens) == 5
|
|
assert tokens[2].text == "bl.a."
|