explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2024-11-13 05:07:03 +03:00
Code Issues Packages Projects Releases Wiki Activity

Use regex instead of re

Browse Source
This commit is contained in:
Ben Eyal 2017-04-20 02:22:52 +03:00
parent 2bd89e7ade
commit d8098a8be2
4 changed files with 5 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
spacy/fr/tokenizer_exceptions.py
View File

@ -13,7 +13,7 @@ from ..symbols import *
import os import os
import io import io
import re import regex as re
def get_exceptions(): def get_exceptions():

2
spacy/hu/tokenizer_exceptions.py
View File

@ -1,7 +1,7 @@
# coding: utf8 # coding: utf8
from __future__ import unicode_literals from __future__ import unicode_literals
import re import regex as re
from spacy.language_data.punctuation import ALPHA_LOWER, CURRENCY from spacy.language_data.punctuation import ALPHA_LOWER, CURRENCY
from ..language_data.tokenizer_exceptions import _URL_PATTERN from ..language_data.tokenizer_exceptions import _URL_PATTERN

2
spacy/language_data/punctuation.py
View File

@ -16,6 +16,8 @@ A Ä À Á Â Ǎ Æ Ã Å Ā Ă Ą B C Ç Ć Č Ĉ Ċ C̄ D Ð Ď E É È Ê Ë
Î Ï Í Ī Ì Ȉ Ǐ Į Ĩ J K Ķ L Ł Ļ M N Ñ Ń Ň Ņ O Ö Ó Ò Ő Ô Õ Œ Ø Ō Ő Ǒ Ơ P Q R Ř Ŗ S Î Ï Í Ī Ì Ȉ Ǐ Į Ĩ J K Ķ L Ł Ļ M N Ñ Ń Ň Ņ O Ö Ó Ò Ő Ô Õ Œ Ø Ō Ő Ǒ Ơ P Q R Ř Ŗ S
Ś Š Ş Ŝ T Ť U Ú Û Ù Ú Ū Ű Ǔ Ů Ų Ư V W Ŵ X Y Ÿ Ý Ỳ Ŷ Ỹ Z Ź Ž Ż Þ Ś Š Ş Ŝ T Ť U Ú Û Ù Ú Ū Ű Ǔ Ů Ų Ư V W Ŵ X Y Ÿ Ý Ỳ Ŷ Ỹ Z Ź Ž Ż Þ
""" """
import regex as re
re.DEFAULT_VERSION = re.VERSION1
_UNITS = """ _UNITS = """

2
spacy/util.py
View File

@ -2,7 +2,7 @@
from __future__ import unicode_literals, print_function from __future__ import unicode_literals, print_function
import ujson import ujson
import re import regex as re
from pathlib import Path from pathlib import Path
import sys import sys
import textwrap import textwrap