Add Korean Revised Romanization to hangeul IME #716
Conversation
| [ 'b', 'ᄇ' ], | ||
| [ 'ᄇp', 'ᄈ' ], | ||
| [ 'ᄉs', 'ᄊ' ], | ||
| // [ '\'', 'ᄋ'], // Apostrophe can be written to represent silent ᄋ |
| // This version does not support context rules, but we don't need them | ||
| patterns: function(input, context) { | ||
| var patterns, regex, rule, replacement, i, result; | ||
|
|
There was a problem hiding this comment.
Too many empty lines. There should be one.
|
|
||
| // This regex matches jamo that form a syllable so they can be combined | ||
| var jamoRegex = /([ᄀ-ᄒ])([ᅡ-ᅵ])([ᆨ-ᇂ])?([ᄀ-ᄒ]|[\- '])(.*)$/; | ||
| if (jamoRegex.test(result)) { |
There was a problem hiding this comment.
We require spaces inside parentheses.
| // This regex matches jamo that form a syllable so they can be combined | ||
| var jamoRegex = /([ᄀ-ᄒ])([ᅡ-ᅵ])([ᆨ-ᇂ])?([ᄀ-ᄒ]|[\- '])(.*)$/; | ||
| if (jamoRegex.test(result)) { | ||
| return { noop: false, output: result.replace(jamoRegex, combineJamo) }; |
There was a problem hiding this comment.
Here, too, spaces inside parentheses.
| // Conjoining jamo behavior is defined by this Unicode standard | ||
| // https://www.unicode.org/versions/Unicode13.0.0/ch03.pdf#G24646 | ||
| // parameter `final` is optional | ||
| function combineJamo(substring, initial, vowel, final, nextSyllableInitial, otherChars) { |
There was a problem hiding this comment.
Here, too, spaces inside parentheses.
| var syllable = String.fromCharCode(syllableNo); | ||
|
|
||
| const disambig = /[\- ']/; | ||
| if (nextSyllableInitial.match(disambig)) { |
There was a problem hiding this comment.
Here, too, spaces inside parentheses.
| const disambig = /[\- ']/; | ||
| if (nextSyllableInitial.match(disambig)) { | ||
| return syllable; | ||
| } else if (otherChars.match(disambig)) { |
| @@ -1274,6 +1278,10 @@ | |||
| autonym: 'ಕನ್ನಡ', | |||
| inputmethods: [ 'kn-transliteration', 'kn-inscript', 'kn-kgp', 'kn-inscript2' ] | |||
| }, | |||
| kor: { | |||
| autonym: '한국어', | |||
| inputmethods: [ 'kor-rr' ] | |||
There was a problem hiding this comment.
We usually give them an identified that is based on the shortest language code, so it should be "ko" and not "kor".
| description: 'Korean RR test', | ||
| inputmethod: 'kor-rr', | ||
| tests: [ | ||
| // Note that RR is meant to romanize from hangul to latin script, but not |
There was a problem hiding this comment.
Capitalize "Hangul" and "Latin".
| ]; | ||
|
|
||
| var koreanRR = { | ||
| id: 'kor-rr', |
There was a problem hiding this comment.
We usually give them an identified that is based on the shortest language code, so it should be "ko" and not "kor".
There was a problem hiding this comment.
I tested this patch and it works as intended. @annedrewhu Thanks a lot for your contributions and apologies for the delay in reviewing!
Follow-up to wikimedia#716. * Rename the "kor-rr" identifier to "ko-rr" and "kor" to "ko": We use two-letter language codes when they are available. * Whitespace clean-up.
Follow-up to wikimedia#716. * Rename the "kor-rr" identifier to "ko-rr" and "kor" to "ko": We use two-letter language codes when they are available. * Whitespace clean-up.
Follow-up to #716. * Rename the "kor-rr" identifier to "ko-rr" and "kor" to "ko": We use two-letter language codes when they are available. * Whitespace clean-up. Co-authored-by: SrishAkaTux <[email protected]>
Any Korean transliteration keyboard will have a lot of edge cases to check because of differences in romanization between revised romanization, the older McCune–Reischauer, and even a different system included in the RR standard. I've added a modest amount of tests based on this official RR document, but I'm open to adding more before merging.