refactor
This commit is contained in:
parent
2dea82c32b
commit
c1ee7d3289
|
@ -329,6 +329,15 @@ export function addDiacritics({ p, f }: T.PsString, ignoreCommas?: true): T.PsSt
|
||||||
};
|
};
|
||||||
}
|
}
|
||||||
|
|
||||||
|
enum PhonemeStatus {
|
||||||
|
LeadingLongVowel,
|
||||||
|
LeadingConsonantOrShortVowel,
|
||||||
|
DoubleConsonantTashdeed,
|
||||||
|
EndingWithHeyHim,
|
||||||
|
DirectMatch,
|
||||||
|
ShortVowel,
|
||||||
|
}
|
||||||
|
|
||||||
function processPhoneme(
|
function processPhoneme(
|
||||||
acc: DiacriticsAccumulator,
|
acc: DiacriticsAccumulator,
|
||||||
phoneme: Phoneme,
|
phoneme: Phoneme,
|
||||||
|
@ -346,64 +355,46 @@ function processPhoneme(
|
||||||
|
|
||||||
const {
|
const {
|
||||||
phonemeInfo,
|
phonemeInfo,
|
||||||
isBeginningOfWord,
|
|
||||||
currentPLetter,
|
|
||||||
needsTashdeed,
|
|
||||||
sukunOrDiacritic,
|
sukunOrDiacritic,
|
||||||
nextPLetter,
|
phonemeState,
|
||||||
isEndOfWord,
|
|
||||||
} = stateInfo({ state, i, phoneme, phonemes });
|
} = stateInfo({ state, i, phoneme, phonemes });
|
||||||
|
|
||||||
// if it's not an exception (TODO)
|
const p = phonemeState
|
||||||
// it must be one of the following 5 possibilities
|
|
||||||
// 1. beginning a word with a long vowel
|
return (p === PhonemeStatus.LeadingLongVowel) ?
|
||||||
if (isBeginningOfWord && (phonemeInfo.longVowel && !phonemeInfo.endingOnly)) {
|
pipe(
|
||||||
if (phoneme !== "aa" && currentPLetter !== "ا" && !phonemeInfo.matches?.includes(nextPLetter)) {
|
|
||||||
throw Error("phonetics error - needs alef prefix");
|
|
||||||
}
|
|
||||||
return pipe(
|
|
||||||
advanceP,
|
advanceP,
|
||||||
addP(phonemeInfo.diacritic),
|
addP(phonemeInfo.diacritic),
|
||||||
advanceP,
|
advanceP,
|
||||||
)(state);
|
)(state)
|
||||||
// 2. beginning a word with something else
|
: (p === PhonemeStatus.LeadingConsonantOrShortVowel) ?
|
||||||
} else if (isBeginningOfWord && (phonemeInfo.beginningMatches?.includes(currentPLetter) || phonemeInfo.matches?.includes(currentPLetter))) {
|
pipe(
|
||||||
return pipe(
|
|
||||||
advanceP,
|
advanceP,
|
||||||
addP(sukunOrDiacritic),
|
addP(sukunOrDiacritic),
|
||||||
advanceForAin,
|
advanceForAin,
|
||||||
)(state);
|
)(state)
|
||||||
// 3. double consonant to be marked with tashdeed
|
: (p === PhonemeStatus.DoubleConsonantTashdeed) ?
|
||||||
} else if (needsTashdeed) {
|
pipe(
|
||||||
return pipe(
|
|
||||||
addP(tashdeed)
|
addP(tashdeed)
|
||||||
)(state);
|
)(state)
|
||||||
// 4. special ه ending
|
: (p === PhonemeStatus.EndingWithHeyHim) ?
|
||||||
} else if (isEndOfWord && ((phoneme === "u" && currentPLetter === "ه") || (phoneme === "h" && ["ه", "ح"].includes(currentPLetter)))) {
|
pipe(
|
||||||
return pipe(
|
|
||||||
advanceP,
|
advanceP,
|
||||||
addP(phoneme === "u" ? hamzaAbove : sukun),
|
addP(phoneme === "u" ? hamzaAbove : sukun),
|
||||||
)(state);
|
)(state)
|
||||||
// 5. direct match of phoneme / P letter
|
: (p === PhonemeStatus.DirectMatch) ?
|
||||||
} else if (phonemeInfo.matches?.includes(currentPLetter) || (isEndOfWord && phonemeInfo.endingMatches?.includes(currentPLetter)) || (phoneme === "m" && currentPLetter === "ن" && nextPLetter === "ب")) {
|
pipe(
|
||||||
return pipe(
|
|
||||||
addP(sukunOrDiacritic),
|
addP(sukunOrDiacritic),
|
||||||
advanceP,
|
advanceP,
|
||||||
)(state);
|
)(state)
|
||||||
// 6. just a diacritic for short vowel
|
: // p === PhonemeState.ShortVowel
|
||||||
} else if (phonemeInfo.diacritic && !phonemeInfo.longVowel) {
|
pipe(
|
||||||
return pipe(
|
|
||||||
advanceForHamzaMid,
|
advanceForHamzaMid,
|
||||||
addP(phonemeInfo.diacritic),
|
addP(phonemeInfo.diacritic),
|
||||||
advanceForAinOrHamza,
|
advanceForAinOrHamza,
|
||||||
)(state);
|
)(state);
|
||||||
}
|
|
||||||
// anything that gets to this point is a failure/error
|
|
||||||
throw new Error("phonetics error");
|
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
function stateInfo({ state, i, phonemes, phoneme }: {
|
function stateInfo({ state, i, phonemes, phoneme }: {
|
||||||
state: DiacriticsAccumulator,
|
state: DiacriticsAccumulator,
|
||||||
i: number,
|
i: number,
|
||||||
|
@ -425,8 +416,36 @@ function stateInfo({ state, i, phonemes, phoneme }: {
|
||||||
const needsSukun = doubleConsonant && ((previousPhoneme !== phoneme) || phonemeInfo.matches?.includes(currentPLetter));
|
const needsSukun = doubleConsonant && ((previousPhoneme !== phoneme) || phonemeInfo.matches?.includes(currentPLetter));
|
||||||
const diacritic = isEndOfWord ? ((!phonemeInfo.longVowel || phonemeInfo.useEndingDiacritic) ? phonemeInfo.diacritic : undefined) : phonemeInfo.diacritic;
|
const diacritic = isEndOfWord ? ((!phonemeInfo.longVowel || phonemeInfo.useEndingDiacritic) ? phonemeInfo.diacritic : undefined) : phonemeInfo.diacritic;
|
||||||
const sukunOrDiacritic = (needsSukun ? sukun : diacritic);
|
const sukunOrDiacritic = (needsSukun ? sukun : diacritic);
|
||||||
|
|
||||||
|
function getPhonemeState(): PhonemeStatus {
|
||||||
|
if (isBeginningOfWord && (phonemeInfo.longVowel && !phonemeInfo.endingOnly)) {
|
||||||
|
if (phoneme !== "aa" && currentPLetter !== "ا" && !phonemeInfo.matches?.includes(nextPLetter)) {
|
||||||
|
throw Error("phonetics error - needs alef prefix");
|
||||||
|
}
|
||||||
|
return PhonemeStatus.LeadingLongVowel;
|
||||||
|
}
|
||||||
|
if (isBeginningOfWord && (phonemeInfo.beginningMatches?.includes(currentPLetter) || phonemeInfo.matches?.includes(currentPLetter))) {
|
||||||
|
return PhonemeStatus.LeadingConsonantOrShortVowel;
|
||||||
|
}
|
||||||
|
if (needsTashdeed) {
|
||||||
|
return PhonemeStatus.DoubleConsonantTashdeed;
|
||||||
|
}
|
||||||
|
if (isEndOfWord && ((phoneme === "u" && currentPLetter === "ه") || (phoneme === "h" && ["ه", "ح"].includes(currentPLetter)))) {
|
||||||
|
return PhonemeStatus.EndingWithHeyHim;
|
||||||
|
}
|
||||||
|
if ((phonemeInfo.matches?.includes(currentPLetter) || (isEndOfWord && phonemeInfo.endingMatches?.includes(currentPLetter)) || (phoneme === "m" && currentPLetter === "ن" && nextPLetter === "ب"))) {
|
||||||
|
return PhonemeStatus.DirectMatch;
|
||||||
|
}
|
||||||
|
if (phonemeInfo.diacritic && !phonemeInfo.longVowel) {
|
||||||
|
return PhonemeStatus.ShortVowel;
|
||||||
|
}
|
||||||
|
throw new Error("phonetics error - no status found for phoneme: " + phoneme);
|
||||||
|
}
|
||||||
|
|
||||||
|
const phonemeState = getPhonemeState();
|
||||||
|
|
||||||
return {
|
return {
|
||||||
phonemeInfo, isBeginningOfWord, currentPLetter, needsTashdeed, sukunOrDiacritic, nextPLetter, isEndOfWord,
|
phonemeState, phonemeInfo, sukunOrDiacritic,
|
||||||
};
|
};
|
||||||
};
|
};
|
||||||
|
|
||||||
|
|
Loading…
Reference in New Issue