mirror of
https://github.com/explosion/spaCy.git
synced 2024-11-14 05:37:03 +03:00
84 lines
2.5 KiB
Python
84 lines
2.5 KiB
Python
# coding: utf-8
|
|
from __future__ import unicode_literals
|
|
from os import path
|
|
|
|
import pytest
|
|
|
|
from ...util import utf8open
|
|
|
|
|
|
def test_tokenizer_handles_no_word(tokenizer):
|
|
tokens = tokenizer("")
|
|
assert len(tokens) == 0
|
|
|
|
|
|
@pytest.mark.parametrize('text', ["lorem"])
|
|
def test_tokenizer_handles_single_word(tokenizer, text):
|
|
tokens = tokenizer(text)
|
|
assert tokens[0].text == text
|
|
|
|
|
|
def test_tokenizer_handles_punct(tokenizer):
|
|
text = "Lorem, ipsum."
|
|
tokens = tokenizer(text)
|
|
assert len(tokens) == 4
|
|
assert tokens[0].text == "Lorem"
|
|
assert tokens[1].text == ","
|
|
assert tokens[2].text == "ipsum"
|
|
assert tokens[1].text != "Lorem"
|
|
|
|
|
|
def test_tokenizer_handles_digits(tokenizer):
|
|
exceptions = ["hu"]
|
|
text = "Lorem ipsum: 1984."
|
|
tokens = tokenizer(text)
|
|
|
|
if tokens[0].lang_ not in exceptions:
|
|
assert len(tokens) == 5
|
|
assert tokens[0].text == "Lorem"
|
|
assert tokens[3].text == "1984"
|
|
|
|
|
|
@pytest.mark.parametrize('text', ["google.com", "python.org", "spacy.io", "explosion.ai"])
|
|
def test_tokenizer_keep_urls(tokenizer, text):
|
|
tokens = tokenizer(text)
|
|
assert len(tokens) == 1
|
|
|
|
|
|
@pytest.mark.parametrize('text', ["hello123@example.com", "hi+there@gmail.it", "matt@explosion.ai"])
|
|
def test_tokenizer_keeps_email(tokenizer, text):
|
|
tokens = tokenizer(text)
|
|
assert len(tokens) == 1
|
|
|
|
|
|
def test_tokenizer_handles_long_text(tokenizer):
|
|
text = """Lorem ipsum dolor sit amet, consectetur adipiscing elit
|
|
|
|
Cras egestas orci non porttitor maximus.
|
|
Maecenas quis odio id dolor rhoncus dignissim. Curabitur sed velit at orci ultrices sagittis. Nulla commodo euismod arcu eget vulputate.
|
|
|
|
Phasellus tincidunt, augue quis porta finibus, massa sapien consectetur augue, non lacinia enim nibh eget ipsum. Vestibulum in bibendum mauris.
|
|
|
|
"Nullam porta fringilla enim, a dictum orci consequat in." Mauris nec malesuada justo."""
|
|
|
|
tokens = tokenizer(text)
|
|
assert len(tokens) > 5
|
|
|
|
|
|
@pytest.mark.parametrize('file_name', ["sun.txt"])
|
|
def test_tokenizer_handle_text_from_file(tokenizer, file_name):
|
|
loc = path.join(path.dirname(__file__), file_name)
|
|
text = utf8open(loc).read()
|
|
assert len(text) != 0
|
|
tokens = tokenizer(text)
|
|
assert len(tokens) > 100
|
|
|
|
|
|
def test_tokenizer_suspected_freeing_strings(tokenizer):
|
|
text1 = "Lorem dolor sit amet, consectetur adipiscing elit."
|
|
text2 = "Lorem ipsum dolor sit amet, consectetur adipiscing elit."
|
|
tokens1 = tokenizer(text1)
|
|
tokens2 = tokenizer(text2)
|
|
assert tokens1[0].text == "Lorem"
|
|
assert tokens2[0].text == "Lorem"
|