2019-02-07 23:07:02 +03:00
|
|
|
# coding: utf8
|
|
|
|
from __future__ import unicode_literals
|
|
|
|
|
2019-01-14 17:32:30 +03:00
|
|
|
from .stop_words import STOP_WORDS
|
|
|
|
from .lex_attrs import LEX_ATTRS
|
|
|
|
|
|
|
|
from ..tokenizer_exceptions import BASE_EXCEPTIONS
|
|
|
|
from ...language import Language
|
|
|
|
from ...attrs import LANG
|
|
|
|
from ...util import update_exc
|
|
|
|
|
2019-02-07 23:07:02 +03:00
|
|
|
|
2019-01-14 17:32:30 +03:00
|
|
|
class TamilDefaults(Language.Defaults):
|
|
|
|
lex_attr_getters = dict(Language.Defaults.lex_attr_getters)
|
2019-02-07 23:07:02 +03:00
|
|
|
lex_attr_getters[LANG] = lambda text: "ta"
|
2019-01-14 17:32:30 +03:00
|
|
|
lex_attr_getters.update(LEX_ATTRS)
|
|
|
|
|
2019-02-07 23:00:04 +03:00
|
|
|
|
2019-01-14 17:32:30 +03:00
|
|
|
class Tamil(Language):
|
2019-02-07 23:07:02 +03:00
|
|
|
lang = "ta"
|
|
|
|
Defaults = TamilDefaults
|
2019-02-07 23:00:04 +03:00
|
|
|
|
2019-01-14 17:32:30 +03:00
|
|
|
|
2019-02-07 23:00:04 +03:00
|
|
|
__all__ = ["Tamil"]
|