mirror of
https://github.com/explosion/spaCy.git
synced 2024-09-21 19:39:13 +03:00
f9e603903b
NUM_WORDS and ORDINAL_WORDS are currently not used, but the hard-coded list should be removed from orth.pyx and replaced to use language-specific functions. This will later allow other languages to use their own functions to set those flags. (In English, this is easier because it only needs to be checked against a set – in German for example, this requires a more complex function, as most number words are one word.)
26 lines
829 B
Python
26 lines
829 B
Python
# coding: utf8
|
||
from __future__ import unicode_literals
|
||
|
||
from .. import language_data as base
|
||
from ..language_data import update_exc, strings_to_exc, expand_exc
|
||
from ..symbols import ORTH, LEMMA
|
||
|
||
from .tag_map import TAG_MAP
|
||
from .word_sets import STOP_WORDS, NUM_WORDS
|
||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS, ORTH_ONLY
|
||
from .morph_rules import MORPH_RULES
|
||
|
||
|
||
TAG_MAP = dict(TAG_MAP)
|
||
STOP_WORDS = set(STOP_WORDS)
|
||
|
||
|
||
TOKENIZER_EXCEPTIONS = dict(TOKENIZER_EXCEPTIONS)
|
||
update_exc(TOKENIZER_EXCEPTIONS, strings_to_exc(ORTH_ONLY))
|
||
update_exc(TOKENIZER_EXCEPTIONS, expand_exc(TOKENIZER_EXCEPTIONS, "'", "’"))
|
||
update_exc(TOKENIZER_EXCEPTIONS, strings_to_exc(base.EMOTICONS))
|
||
update_exc(TOKENIZER_EXCEPTIONS, strings_to_exc(base.ABBREVIATIONS))
|
||
|
||
|
||
__all__ = ["TOKENIZER_EXCEPTIONS", "TAG_MAP", "STOP_WORDS", "MORPH_RULES"]
|