spaCy/spacy/lang/hu/punctuation.py

52 lines
1.3 KiB
Python
Raw Normal View History

2017-03-12 15:07:28 +03:00
# coding: utf8
from __future__ import unicode_literals
2017-05-30 22:34:46 +03:00
from ..char_classes import LIST_PUNCT, LIST_ELLIPSES, LIST_QUOTES
2017-05-09 01:01:59 +03:00
from ..char_classes import QUOTES, UNITS, ALPHA, ALPHA_LOWER, ALPHA_UPPER
2017-05-08 16:49:56 +03:00
LIST_ICONS = [r"[\p{So}--[°]]"]
2017-05-08 16:49:56 +03:00
_currency = r"\$|¢|£|€|¥|฿"
_quotes = QUOTES.replace("'", "")
2017-05-09 01:01:59 +03:00
_prefixes = (
[r"\+"]
+ LIST_PUNCT
+ LIST_ELLIPSES
+ LIST_QUOTES
+ LIST_ICONS
+ [r"[,.:](?=[{a}])".format(a=ALPHA)]
)
2017-05-09 01:01:59 +03:00
_suffixes = (
LIST_PUNCT
+ LIST_ELLIPSES
+ LIST_QUOTES
+ LIST_ICONS
+ [
r"(?<=[0-9])\+",
r"(?<=°[FfCcKk])\.",
r"(?<=[0-9])(?:{})".format(_currency),
r"(?<=[0-9])(?:{})".format(UNITS),
r"(?<=[{}{}{}(?:{})])\.".format(ALPHA_LOWER, r"%²\-\)\]\+", QUOTES, _currency),
r"(?<=[{})])-e".format(ALPHA_LOWER),
]
)
2017-05-09 01:01:59 +03:00
_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}]),(?=[{a}])".format(a=ALPHA),
r"(?<=[{a}])([{q}\)\]\(\[])(?=[\-{a}])".format(a=ALPHA, q=_quotes),
]
)
2017-05-09 01:01:59 +03:00
TOKENIZER_PREFIXES = _prefixes
TOKENIZER_SUFFIXES = _suffixes
TOKENIZER_INFIXES = _infixes