mirror of
https://github.com/explosion/spaCy.git
synced 2024-12-25 01:16:28 +03:00
23 lines
641 B
Python
23 lines
641 B
Python
# coding: utf-8
|
|
from __future__ import unicode_literals
|
|
|
|
import pytest
|
|
|
|
|
|
@pytest.mark.parametrize("text,length", [("z.B.", 1), ("zb.", 2), ("(z.B.", 2)])
|
|
def test_lb_tokenizer_splits_prefix_interact(lb_tokenizer, text, length):
|
|
tokens = lb_tokenizer(text)
|
|
assert len(tokens) == length
|
|
|
|
|
|
@pytest.mark.parametrize("text", ["z.B.)"])
|
|
def test_lb_tokenizer_splits_suffix_interact(lb_tokenizer, text):
|
|
tokens = lb_tokenizer(text)
|
|
assert len(tokens) == 2
|
|
|
|
|
|
@pytest.mark.parametrize("text", ["(z.B.)"])
|
|
def test_lb_tokenizer_splits_even_wrap_interact(lb_tokenizer, text):
|
|
tokens = lb_tokenizer(text)
|
|
assert len(tokens) == 3
|