Rooba
/
spaCy
mirror of https://github.com/explosion/spaCy.git
1
1
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Add kyrgyz to char_classes

This commit is contained in:
muratjumashev 2021-01-23 21:53:41 +06:00
parent e30bbf5432
commit fe3b5b8ff5
1 changed files with 7 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
spacy/lang/char_classes.py
View File

@ -207,6 +207,10 @@ _tatar_lower = r"әөүҗңһ"
_tatar_upper = r"ӘӨҮҖҢҺ" _tatar_upper = r"ӘӨҮҖҢҺ"
_tatar = r"әөүҗңһӘӨҮҖҢҺ" _tatar = r"әөүҗңһӘӨҮҖҢҺ"
_kyrgyz_lower = r"өңү"
_kyrgyz_upper = r"ӨҢҮ"
_kyrgyz = r"өңүӨҢҮ"
_greek_lower = r"α-ωάέίόώήύ" _greek_lower = r"α-ωάέίόώήύ"
_greek_upper = r"Α-ΩΆΈΊΌΏΉΎ" _greek_upper = r"Α-ΩΆΈΊΌΏΉΎ"
_greek = r"α-ωάέίόώήύΑ-ΩΆΈΊΌΏΉΎ" _greek = r"α-ωάέίόώήύΑ-ΩΆΈΊΌΏΉΎ"
@ -219,8 +223,8 @@ _macedonian_lower = r"ѓѕјљњќѐѝ"
_macedonian_upper = r"ЃЅЈЉЊЌЀЍ" _macedonian_upper = r"ЃЅЈЉЊЌЀЍ"
_macedonian = r"ѓѕјљњќѐѝЃЅЈЉЊЌЀЍ" _macedonian = r"ѓѕјљњќѐѝЃЅЈЉЊЌЀЍ"
_upper = LATIN_UPPER + _russian_upper + _tatar_upper + _greek_upper + _ukrainian_upper + _macedonian_upper _upper = LATIN_UPPER + _russian_upper + _tatar_upper + _kyrgyz_upper + _greek_upper + _ukrainian_upper + _macedonian_upper
_lower = LATIN_LOWER + _russian_lower + _tatar_lower + _greek_lower + _ukrainian_lower + _macedonian_lower _lower = LATIN_LOWER + _russian_lower + _tatar_lower + _kyrgyz_lower + _greek_lower + _ukrainian_lower + _macedonian_lower
_uncased = ( _uncased = (
_ethiopic _ethiopic
@ -236,7 +240,7 @@ _uncased = (
+ _cjk + _cjk
) )
ALPHA = group_chars(LATIN + _russian + _tatar + _greek + _ukrainian + _macedonian + _uncased) ALPHA = group_chars(LATIN + _russian + _tatar + _kyrgyz + _greek + _ukrainian + _macedonian + _uncased)
ALPHA_LOWER = group_chars(_lower + _uncased) ALPHA_LOWER = group_chars(_lower + _uncased)
ALPHA_UPPER = group_chars(_upper + _uncased) ALPHA_UPPER = group_chars(_upper + _uncased)