diff options
author | Kuuuube <61125188+Kuuuube@users.noreply.github.com> | 2024-06-26 13:05:23 -0400 |
---|---|---|
committer | GitHub <noreply@github.com> | 2024-06-26 17:05:23 +0000 |
commit | dfc3710108fe2617a98fca0f57f5a2f6bb7d1830 (patch) | |
tree | ad402a4028c1b4f678262776c5a8b84d25e7dcf9 /test/language | |
parent | 2a92a0b98c2bb08c2adaca24ff8af3322874ef59 (diff) |
Add normalization of combining dakuten and handakuten to ja preprocessors (#1136)
* Add normalization of combining dakuten and handakuten to ja preprocessors
* Fix typo
* Remove redundant variable assignment
* Fix first character processed incorrectly when it is a character that gets combined
* Add test for combining dakuten and handakuten
Diffstat (limited to 'test/language')
-rw-r--r-- | test/language/japanese-preprocessors.test.js | 202 |
1 files changed, 202 insertions, 0 deletions
diff --git a/test/language/japanese-preprocessors.test.js b/test/language/japanese-preprocessors.test.js new file mode 100644 index 00000000..90313abd --- /dev/null +++ b/test/language/japanese-preprocessors.test.js @@ -0,0 +1,202 @@ +/* + * Copyright (C) 2024 Yomitan Authors + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 3 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <https://www.gnu.org/licenses/>. + */ + +import {describe, expect, test} from 'vitest'; +import {normalizeCombiningCharacters} from '../../ext/js/language/ja/japanese-text-preprocessors.js'; + +const testCasesDakuten = [ + ['か\u3099', 'が'], + ['き\u3099', 'ぎ'], + ['く\u3099', 'ぐ'], + ['け\u3099', 'げ'], + ['こ\u3099', 'ご'], + ['さ\u3099', 'ざ'], + ['し\u3099', 'じ'], + ['す\u3099', 'ず'], + ['せ\u3099', 'ぜ'], + ['そ\u3099', 'ぞ'], + ['た\u3099', 'だ'], + ['ち\u3099', 'ぢ'], + ['つ\u3099', 'づ'], + ['て\u3099', 'で'], + ['と\u3099', 'ど'], + ['は\u3099', 'ば'], + ['ひ\u3099', 'び'], + ['ふ\u3099', 'ぶ'], + ['へ\u3099', 'べ'], + ['ほ\u3099', 'ぼ'], + ['カ\u3099', 'ガ'], + ['キ\u3099', 'ギ'], + ['ク\u3099', 'グ'], + ['ケ\u3099', 'ゲ'], + ['コ\u3099', 'ゴ'], + ['サ\u3099', 'ザ'], + ['シ\u3099', 'ジ'], + ['ス\u3099', 'ズ'], + ['セ\u3099', 'ゼ'], + ['ソ\u3099', 'ゾ'], + ['タ\u3099', 'ダ'], + ['チ\u3099', 'ヂ'], + ['ツ\u3099', 'ヅ'], + ['テ\u3099', 'デ'], + ['ト\u3099', 'ド'], + ['ハ\u3099', 'バ'], + ['ヒ\u3099', 'ビ'], + ['フ\u3099', 'ブ'], + ['ヘ\u3099', 'ベ'], + ['ホ\u3099', 'ボ'], +]; + +const testCasesHandakuten = [ + ['は\u309A', 'ぱ'], + ['ひ\u309A', 'ぴ'], + ['ふ\u309A', 'ぷ'], + ['へ\u309A', 'ぺ'], + ['ほ\u309A', 'ぽ'], + ['ハ\u309A', 'パ'], + ['ヒ\u309A', 'ピ'], + ['フ\u309A', 'プ'], + ['ヘ\u309A', 'ペ'], + ['ホ\u309A', 'ポ'], +]; + +const testCasesIgnored = [ + ['な\u3099', 'な\u3099'], + ['な\u309A', 'な\u309A'], + ['に\u3099', 'に\u3099'], + ['に\u309A', 'に\u309A'], + ['ぬ\u3099', 'ぬ\u3099'], + ['ぬ\u309A', 'ぬ\u309A'], + ['ね\u3099', 'ね\u3099'], + ['ね\u309A', 'ね\u309A'], + ['の\u3099', 'の\u3099'], + ['の\u309A', 'の\u309A'], + ['ま\u3099', 'ま\u3099'], + ['ま\u309A', 'ま\u309A'], + ['み\u3099', 'み\u3099'], + ['み\u309A', 'み\u309A'], + ['む\u3099', 'む\u3099'], + ['む\u309A', 'む\u309A'], + ['め\u3099', 'め\u3099'], + ['め\u309A', 'め\u309A'], + ['も\u3099', 'も\u3099'], + ['も\u309A', 'も\u309A'], + ['ゃ\u3099', 'ゃ\u3099'], + ['ゃ\u309A', 'ゃ\u309A'], + ['や\u3099', 'や\u3099'], + ['や\u309A', 'や\u309A'], + ['ゅ\u3099', 'ゅ\u3099'], + ['ゅ\u309A', 'ゅ\u309A'], + ['ゆ\u3099', 'ゆ\u3099'], + ['ゆ\u309A', 'ゆ\u309A'], + ['ょ\u3099', 'ょ\u3099'], + ['ょ\u309A', 'ょ\u309A'], + ['よ\u3099', 'よ\u3099'], + ['よ\u309A', 'よ\u309A'], + ['ら\u3099', 'ら\u3099'], + ['ら\u309A', 'ら\u309A'], + ['り\u3099', 'り\u3099'], + ['り\u309A', 'り\u309A'], + ['る\u3099', 'る\u3099'], + ['る\u309A', 'る\u309A'], + ['れ\u3099', 'れ\u3099'], + ['れ\u309A', 'れ\u309A'], + ['ろ\u3099', 'ろ\u3099'], + ['ろ\u309A', 'ろ\u309A'], + ['ゎ\u3099', 'ゎ\u3099'], + ['ゎ\u309A', 'ゎ\u309A'], + ['わ\u3099', 'わ\u3099'], + ['わ\u309A', 'わ\u309A'], + ['ゐ\u3099', 'ゐ\u3099'], + ['ゐ\u309A', 'ゐ\u309A'], + ['ゑ\u3099', 'ゑ\u3099'], + ['ゑ\u309A', 'ゑ\u309A'], + ['を\u3099', 'を\u3099'], + ['を\u309A', 'を\u309A'], + ['ん\u3099', 'ん\u3099'], + ['ん\u309A', 'ん\u309A'], + ['ナ\u3099', 'ナ\u3099'], + ['ナ\u309A', 'ナ\u309A'], + ['ニ\u3099', 'ニ\u3099'], + ['ニ\u309A', 'ニ\u309A'], + ['ヌ\u3099', 'ヌ\u3099'], + ['ヌ\u309A', 'ヌ\u309A'], + ['ネ\u3099', 'ネ\u3099'], + ['ネ\u309A', 'ネ\u309A'], + ['ノ\u3099', 'ノ\u3099'], + ['ノ\u309A', 'ノ\u309A'], + ['マ\u3099', 'マ\u3099'], + ['マ\u309A', 'マ\u309A'], + ['ミ\u3099', 'ミ\u3099'], + ['ミ\u309A', 'ミ\u309A'], + ['ム\u3099', 'ム\u3099'], + ['ム\u309A', 'ム\u309A'], + ['メ\u3099', 'メ\u3099'], + ['メ\u309A', 'メ\u309A'], + ['モ\u3099', 'モ\u3099'], + ['モ\u309A', 'モ\u309A'], + ['ャ\u3099', 'ャ\u3099'], + ['ャ\u309A', 'ャ\u309A'], + ['ヤ\u3099', 'ヤ\u3099'], + ['ヤ\u309A', 'ヤ\u309A'], + ['ュ\u3099', 'ュ\u3099'], + ['ュ\u309A', 'ュ\u309A'], + ['ユ\u3099', 'ユ\u3099'], + ['ユ\u309A', 'ユ\u309A'], + ['ョ\u3099', 'ョ\u3099'], + ['ョ\u309A', 'ョ\u309A'], + ['ヨ\u3099', 'ヨ\u3099'], + ['ヨ\u309A', 'ヨ\u309A'], + ['ラ\u3099', 'ラ\u3099'], + ['ラ\u309A', 'ラ\u309A'], + ['リ\u3099', 'リ\u3099'], + ['リ\u309A', 'リ\u309A'], + ['ル\u3099', 'ル\u3099'], + ['ル\u309A', 'ル\u309A'], + ['レ\u3099', 'レ\u3099'], + ['レ\u309A', 'レ\u309A'], + ['ロ\u3099', 'ロ\u3099'], + ['ロ\u309A', 'ロ\u309A'], + ['ヮ\u3099', 'ヮ\u3099'], + ['ヮ\u309A', 'ヮ\u309A'], + ['ワ\u3099', 'ワ\u3099'], + ['ワ\u309A', 'ワ\u309A'], + ['ヰ\u3099', 'ヰ\u3099'], + ['ヰ\u309A', 'ヰ\u309A'], + ['ヱ\u3099', 'ヱ\u3099'], + ['ヱ\u309A', 'ヱ\u309A'], + ['ヲ\u3099', 'ヲ\u3099'], + ['ヲ\u309A', 'ヲ\u309A'], + ['ン\u3099', 'ン\u3099'], + ['ン\u309A', 'ン\u309A'], +]; + +const textCasesMisc = [ + ['', ''], + ['\u3099ハ', '\u3099ハ'], + ['\u309Aハ', '\u309Aハ'], + ['さくらし\u3099また\u3099いこん', 'さくらじまだいこん'], + ['いっほ\u309Aん', 'いっぽん'], +]; + +describe('combining dakuten/handakuten normalization', () => { + const {process} = normalizeCombiningCharacters; + const testCases = [...testCasesDakuten, ...testCasesHandakuten, ...testCasesIgnored, ...textCasesMisc]; + test.each(testCases)('%s normalizes to %s', (input, expected) => { + expect(process(input, true)).toStrictEqual(expected); + }); +}); |