mirror of
https://github.com/explosion/spaCy.git
synced 2024-11-11 12:18:04 +03:00
15 lines
658 B
Python
15 lines
658 B
Python
|
# coding: utf8
|
||
|
|
from __future__ import unicode_literals
|
||
|
|
from ..char_classes import LIST_ELLIPSES, LIST_ICONS
|
||
|
|
from ..char_classes import QUOTES, ALPHA, ALPHA_LOWER, ALPHA_UPPER
|
||
|
|
_quotes = QUOTES.replace("'", '')
|
||
|
|
_infixes = (LIST_ELLIPSES + LIST_ICONS +
|
||
|
|
[r'(?<=[{}])\.(?=[{}])'.format(ALPHA_LOWER, ALPHA_UPPER),
|
||
|
|
r'(?<=[{a}])[,!?](?=[{a}])'.format(a=ALPHA),
|
||
|
|
r'(?<=[{a}"])[:<>=](?=[{a}])'.format(a=ALPHA),
|
||
|
|
r'(?<=[{a}]),(?=[{a}])'.format(a=ALPHA),
|
||
|
|
r'(?<=[{a}])([{q}\)\]\(\[])(?=[\{a}])'.format(a=ALPHA, q=_quotes),
|
||
|
|
r'(?<=[{a}])--(?=[{a}])'.format(a=ALPHA)])
|
||
|
|
|
||
|
|
TOKENIZER_INFIXES = _infixes
|