Download code/ja_ext/jaconv/conv_table.py from IndexTeam/Index-Echo-S2ST-9B: direct link, hf CLI and curl.
- Browser
- Download file 5.67 kB
-
https://huggingface.co/IndexTeam/Index-Echo-S2ST-9B/resolve/main/code/ja_ext/jaconv/conv_table.py
- Command line
-
hf download hf://IndexTeam/Index-Echo-S2ST-9B/code/ja_ext/jaconv/conv_table.py
-
curl -L -o conv_table.py https://huggingface.co/IndexTeam/Index-Echo-S2ST-9B/resolve/main/code/ja_ext/jaconv/conv_table.py
5.67 kB
| # -*- coding: utf-8 -*- | |
| from __future__ import unicode_literals | |
| import re | |
| from .compat import map, zip | |
| HIRAGANA = list( | |
| 'ぁあぃいぅうぇえぉおかがきぎくぐけげこごさざしじすず' | |
| 'せぜそぞただちぢっつづてでとどなにぬねのはばぱひびぴ' | |
| 'ふぶぷへべぺほぼぽまみむめもゃやゅゆょよらりるれろわ' | |
| 'をんーゎゐゑゕゖゔゝゞ・「」。、' | |
| ) | |
| HALF_ASCII = list( | |
| '!"#$%&\'()*+,-./:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ' | |
| '[\\]^_`abcdefghijklmnopqrstuvwxyz{|}~ ' | |
| ) | |
| HALF_DIGIT = list('0123456789') | |
| HALF_KANA_SEION = list( | |
| 'ァアィイゥウェエォオカキクケコサシスセソタチッツテトナニヌネノハヒフヘホマミムメモャヤュユョヨラリルレロワヲンーヮヰヱヵヶヽヾ・「」。、' | |
| ) | |
| HALF_KANA = [ | |
| 'ァ', | |
| 'ア', | |
| 'ィ', | |
| 'イ', | |
| 'ゥ', | |
| 'ウ', | |
| 'ェ', | |
| 'エ', | |
| 'ォ', | |
| 'オ', | |
| 'カ', | |
| 'ガ', | |
| 'キ', | |
| 'ギ', | |
| 'ク', | |
| 'グ', | |
| 'ケ', | |
| 'ゲ', | |
| 'コ', | |
| 'ゴ', | |
| 'サ', | |
| 'ザ', | |
| 'シ', | |
| 'ジ', | |
| 'ス', | |
| 'ズ', | |
| 'セ', | |
| 'ゼ', | |
| 'ソ', | |
| 'ゾ', | |
| 'タ', | |
| 'ダ', | |
| 'チ', | |
| 'ヂ', | |
| 'ッ', | |
| 'ツ', | |
| 'ヅ', | |
| 'テ', | |
| 'デ', | |
| 'ト', | |
| 'ド', | |
| 'ナ', | |
| 'ニ', | |
| 'ヌ', | |
| 'ネ', | |
| 'ノ', | |
| 'ハ', | |
| 'バ', | |
| 'パ', | |
| 'ヒ', | |
| 'ビ', | |
| 'ピ', | |
| 'フ', | |
| 'ブ', | |
| 'プ', | |
| 'ヘ', | |
| 'ベ', | |
| 'ペ', | |
| 'ホ', | |
| 'ボ', | |
| 'ポ', | |
| 'マ', | |
| 'ミ', | |
| 'ム', | |
| 'メ', | |
| 'モ', | |
| 'ャ', | |
| 'ヤ', | |
| 'ュ', | |
| 'ユ', | |
| 'ョ', | |
| 'ヨ', | |
| 'ラ', | |
| 'リ', | |
| 'ル', | |
| 'レ', | |
| 'ロ', | |
| 'ワ', | |
| 'ヲ', | |
| 'ン', | |
| 'ー', | |
| 'ヮ', | |
| 'ヰ', | |
| 'ヱ', | |
| 'ヵ', | |
| 'ヶ', | |
| 'ヴ', | |
| 'ヽ', | |
| 'ヾ', | |
| '・', | |
| '「', | |
| '」', | |
| '。', | |
| '、', | |
| ] | |
| FULL_ASCII = list( | |
| '!"#$%&'()*+,-./:;<=>?@' | |
| 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' | |
| '[\]^_`abcdefghijklmnopqrst' | |
| 'uvwxyz{|}~ ' | |
| ) | |
| FULL_DIGIT = list('0123456789') | |
| FULL_KANA = list( | |
| 'ァアィイゥウェエォオカガキギクグケゲコゴサザシジスズセゼソ' | |
| 'ゾタダチヂッツヅテデトドナニヌネノハバパヒビピフブプヘベペ' | |
| 'ホボポマミムメモャヤュユョヨラリルレロワヲンーヮヰヱヵヶヴ' | |
| 'ヽヾ・「」。、' | |
| ) | |
| FULL_KANA_SEION = list( | |
| 'ァアィイゥウェエォオカキクケコサシスセソタチッツテト' | |
| 'ナニヌネノハヒフヘホマミムメモャヤュユョヨラリルレロ' | |
| 'ワヲンーヮヰヱヵヶヽヾ・「」。、' | |
| ) | |
| HEPBURN = list('aiueoaiueonl~-') | |
| HEPBURN_KANA = list('ぁぃぅぇぉあいうえおんる〜ー') | |
| SMALL_KANA = list('ぁぃぅぇぉゃゅょっァィゥェォヵヶャュョッ') | |
| SMALL_KANA_NORMALIZED = list('あいうえおやゆよつアイウエオカケヤユヨツ') | |
| def _to_ord_list(chars): | |
| return list(map(ord, chars)) | |
| HIRAGANA_ORD = _to_ord_list(HIRAGANA) | |
| FULL_KANA_ORD = _to_ord_list(FULL_KANA) | |
| HALF_ASCII_ORD = _to_ord_list(HALF_ASCII) | |
| FULL_ASCII_ORD = _to_ord_list(FULL_ASCII) | |
| HALF_DIGIT_ORD = _to_ord_list(HALF_DIGIT) | |
| FULL_DIGIT_ORD = _to_ord_list(FULL_DIGIT) | |
| HALF_KANA_SEION_ORD = _to_ord_list(HALF_KANA_SEION) | |
| FULL_KANA_SEION_ORD = _to_ord_list(FULL_KANA_SEION) | |
| SMALL_KANA_ORD = _to_ord_list(SMALL_KANA) | |
| def _to_dict(_from, _to): | |
| return dict(zip(_from, _to)) | |
| H2K_TABLE = _to_dict(HIRAGANA_ORD, FULL_KANA) | |
| H2HK_TABLE = _to_dict(HIRAGANA_ORD, HALF_KANA) | |
| K2H_TABLE = _to_dict(FULL_KANA_ORD, HIRAGANA) | |
| H2Z_A = _to_dict(HALF_ASCII_ORD, FULL_ASCII) | |
| H2Z_AD = _to_dict(HALF_ASCII_ORD + HALF_DIGIT_ORD, FULL_ASCII + FULL_DIGIT) | |
| H2Z_AK = _to_dict(HALF_ASCII_ORD + HALF_KANA_SEION_ORD, FULL_ASCII + FULL_KANA_SEION) | |
| H2Z_D = _to_dict(HALF_DIGIT_ORD, FULL_DIGIT) | |
| H2Z_K = _to_dict(HALF_KANA_SEION_ORD, FULL_KANA_SEION) | |
| H2Z_DK = _to_dict(HALF_DIGIT_ORD + HALF_KANA_SEION_ORD, FULL_DIGIT + FULL_KANA_SEION) | |
| H2Z_ALL = _to_dict( | |
| HALF_ASCII_ORD + HALF_DIGIT_ORD + HALF_KANA_SEION_ORD, | |
| FULL_ASCII + FULL_DIGIT + FULL_KANA_SEION, | |
| ) | |
| Z2H_A = _to_dict(FULL_ASCII_ORD, HALF_ASCII) | |
| Z2H_AD = _to_dict(FULL_ASCII_ORD + FULL_DIGIT_ORD, HALF_ASCII + HALF_DIGIT) | |
| Z2H_AK = _to_dict(FULL_ASCII_ORD + FULL_KANA_ORD, HALF_ASCII + HALF_KANA) | |
| Z2H_D = _to_dict(FULL_DIGIT_ORD, HALF_DIGIT) | |
| Z2H_K = _to_dict(FULL_KANA_ORD, HALF_KANA) | |
| Z2H_DK = _to_dict(FULL_DIGIT_ORD + FULL_KANA_ORD, HALF_DIGIT + HALF_KANA) | |
| Z2H_ALL = _to_dict( | |
| FULL_ASCII_ORD + FULL_DIGIT_ORD + FULL_KANA_ORD, HALF_ASCII + HALF_DIGIT + HALF_KANA | |
| ) | |
| KANA2HEP = _to_dict(_to_ord_list(HEPBURN_KANA), HEPBURN) | |
| HEP2KANA = _to_dict(_to_ord_list(HEPBURN), HEPBURN_KANA) | |
| JULIUS_LONG_VOWEL = tuple( | |
| ( | |
| (re.compile('( a){2,}'), ' a:'), | |
| (re.compile('( i){2,}'), ' i:'), | |
| (re.compile('( u){2,}'), ' u:'), | |
| (re.compile('( e){2,}'), ' e:'), | |
| (re.compile('( o){2,}'), ' o:'), | |
| ) | |
| ) | |
| SMALL_KANA2NORMAL_KANA = _to_dict(SMALL_KANA_ORD, SMALL_KANA_NORMALIZED) | |
| del _to_ord_list | |
| del _to_dict | |
| del HIRAGANA_ORD | |
| del HIRAGANA | |
| del HALF_KANA | |
| del FULL_KANA_ORD | |
| del FULL_KANA | |
| del HALF_ASCII_ORD | |
| del HALF_ASCII | |
| del FULL_ASCII_ORD | |
| del FULL_ASCII | |
| del HALF_DIGIT_ORD | |
| del HALF_DIGIT | |
| del FULL_DIGIT_ORD | |
| del FULL_DIGIT | |
| del HALF_KANA_SEION_ORD | |
| del HALF_KANA_SEION | |
| del FULL_KANA_SEION_ORD | |
| del FULL_KANA_SEION | |
| del HEPBURN | |
| del HEPBURN_KANA | |
| del SMALL_KANA | |
| del SMALL_KANA_ORD | |
| del SMALL_KANA_NORMALIZED | |