@kensio/pinyinjs 1.9.9 → 1.10.0

package/dist/dictionary/taiwan-reading.js

@@ -1,38 +1,107 @@
 /**
  * The 國語 reading an entry carries, where a source states one.
  *
- * Both sources hedge it. CC-CEDICT hangs `Taiwan pr.` on one sense of a
- * headword and the others know nothing about it, and a note can sit on the
- * right sense and still be about that sense rather than the character. What
- * survives both tests, and is not simply another 普通话 sense of the word, is
- * the delta.
+ * 教育部's dictionaries are the standard and are asked first. For a word they
+ * list, what they say settles the entry, including when they say the word
+ * reads as it does in 普通话. A word they leave out falls back to CC-CEDICT's
+ * `Taiwan pr.` notes and Unihan's `kMandarin` pairs, which between them mark
+ * some 640 items.
+ *
+ * A character is settled the other way round. 教育部 lists a polyphone's
+ * readings in the order its entries are written, which leads with the literary
+ * reading as often as the current one: 們 is `mén` before `men` and 著 `zhù`
+ * before `zhe`. The order says nothing about which a character on its own
+ * means, so the dictionary is used as a check on the notes, not as a default.
+ * Both dictionaries are consulted for a character, since the question is only
+ * whether a reading is listed at all.
+ *
+ * Every source hedges in its own way. CC-CEDICT hangs `Taiwan pr.` on one
+ * sense of a headword and the others know nothing about it. What survives the
+ * sense tests, and is not simply another 普通话 sense of the word, is the
+ * delta. A reading that is the locale's but only for one sense is kept as an
+ * alternate, so a caller can still see it.
  */
 import { isSingleCharacter } from "../script/characters.js";
-import { isOwnSense, isSenseScopedNote } from "./cedict-senses.js";
-import { characterSyllable } from "./character-defaults.js";
+import { isOwnSense } from "./cedict-senses.js";
 import { isSameReading } from "./entry.js";
-import { readDictionaryReading } from "./reading.js";
+import { notedReadingOf } from "./taiwan-notes.js";
+import { alternatesOf, isRetoned, standardReadingFor, standardReadingsOf, } from "./standard-readings.js";
 /**
- * Settle it for one word, or undefined where no source states one.
+ * What a word is held to: 教育部 where it lists the word, the notes where not.
+ *
+ * A word only the revised dictionary lists may move its tones and nothing
+ * else. That dictionary records classical words beside current ones under the
+ * same characters, and where the concise one is silent there is nothing to say
+ * which a headword is: it lists 五百 only as `wǔ bó`, which is not how
+ * Taipei reads the number. A tone is the shift 國語 makes across whole classes
+ * of words, and a change of sound is more often a different word written the
+ * same way.
  */
-export function taiwanReadingOf(word, senses, senseReadings, reading, unihan) {
-    const taiwanSense = senses.find((entry) => entry.taiwanReadings !== undefined);
-    const taiwanTokens = isSenseScopedNote(word, taiwanSense)
-        ? undefined
-        : taiwanSense?.taiwanReadings;
-    const unihanTaiwan = unihan?.taiwanReading;
-    let taiwan;
-    if (taiwanTokens !== undefined) {
-        taiwan = readDictionaryReading(word, taiwanTokens);
+function wordReadingOf(standard, noted, reading, isDelta) {
+    if (standard.length === 0) {
+        return noted !== undefined && isDelta(noted.reading)
+            ? { reading: noted.reading, alternates: [], isStandard: false }
+            : { alternates: [], isStandard: false };
     }
-    else if (isSingleCharacter(word) && unihanTaiwan !== undefined) {
-        const syllable = characterSyllable(word, unihanTaiwan);
-        taiwan = syllable === undefined ? undefined : [syllable];
+    const readings = standard.map((listed) => listed.reading);
+    const isCurrent = standard.some((listed) => listed.dictionary === "concise");
+    const chosen = standardReadingFor(readings, reading, isDelta);
+    const taiwan = chosen !== undefined && (isCurrent || isRetoned(chosen, reading))
+        ? chosen
+        : undefined;
+    return {
+        ...(taiwan !== undefined && { reading: taiwan }),
+        // The readings it could not be told apart from, which are other words
+        // written alike.
+        alternates: isCurrent ? alternatesOf(readings, taiwan, isDelta) : [],
+        isStandard: true,
+    };
+}
+/**
+ * What a character is held to: the notes, as far as 教育部 bears them out.
+ *
+ * A note 教育部 lists the reading for stands, as the default if it is about
+ * the whole character and as an alternate if it is about one sense: 和's `hàn`
+ * is listed, and is the conjunction only. A note it lists no such reading for
+ * is dropped, which is what the standard says of 俄 `è` and 法 `fà`.
+ *
+ * 教育部 never supplies a character's reading by itself. Its order is the
+ * order its entries are written in, and a character it does not list in its
+ * 普通话 reading is as often a gap in the concise dictionary as a reading that
+ * has moved: it has no 差 `chà`, and the one 差 it lists that 普通话 has no
+ * sense for is `cuō`.
+ */
+function characterReadingOf(standard, noted, isDelta) {
+    const isStandard = standard.length > 0;
+    if (noted === undefined || !isDelta(noted.reading)) {
+        return { alternates: [], isStandard };
+    }
+    const isListed = standard.some((listed) => isSameReading(listed.reading, noted.reading));
+    if (isStandard && !isListed) {
+        return { alternates: [], isStandard };
     }
-    if (taiwan !== undefined &&
-        (isSameReading(taiwan, reading) || isOwnSense(taiwan, senseReadings))) {
-        return undefined;
+    return noted.isSenseScoped
+        ? { alternates: [noted.reading], isStandard }
+        : { reading: noted.reading, alternates: [], isStandard };
+}
+/**
+ * Settle it for one word.
+ *
+ * `spellings` are the word's 繁體 forms, which is how 教育部 keys it.
+ */
+export function taiwanReadingOf(word, spellings, senses, senseReadings, reading, unihan, moe) {
+    // A reading the word already has in 普通话 is a sense, not a locale.
+    const isDelta = (taiwan) => taiwan !== undefined &&
+        !isSameReading(taiwan, reading) &&
+        !isOwnSense(taiwan, senseReadings);
+    const noted = notedReadingOf(word, senses, unihan);
+    if (!isSingleCharacter(word)) {
+        return wordReadingOf(standardReadingsOf(spellings, moe), noted, reading, isDelta);
     }
-    return taiwan;
+    // Every spelling, because a 简体 character can be a 繁體 one in its own
+    // right with readings of its own: 教育部 lists 台 only as `yí`, and the
+    // `tái` is under 臺.
+    const listed = spellings.flatMap((spelling) => standardReadingsOf([spelling], moe));
+    return characterReadingOf(listed, noted, isDelta);
 }
 //# sourceMappingURL=taiwan-reading.js.map

package/dist/dictionary/taiwan-reading.js.map

@@ -1 +1 @@
-{"version":3,"file":"taiwan-reading.js","sourceRoot":"","sources":["../../src/dictionary/taiwan-reading.ts"],"names":[],"mappings":"AAAA;;;;;;;;GAQG;AACH,OAAO,EAAE,iBAAiB,EAAE,MAAM,yBAAyB,CAAC;AAI5D,OAAO,EAAE,UAAU,EAAE,iBAAiB,EAAE,MAAM,oBAAoB,CAAC;AACnE,OAAO,EAAE,iBAAiB,EAAE,MAAM,yBAAyB,CAAC;AAC5D,OAAO,EAAE,aAAa,EAAE,MAAM,YAAY,CAAC;AAC3C,OAAO,EAAE,qBAAqB,EAAE,MAAM,cAAc,CAAC;AAErD;;GAEG;AACH,MAAM,UAAU,eAAe,CAC7B,IAAY,EACZ,MAA8B,EAC9B,aAA+C,EAC/C,OAA4B,EAC5B,MAAkC;IAElC,MAAM,WAAW,GAAG,MAAM,CAAC,IAAI,CAC7B,CAAC,KAAK,EAAE,EAAE,CAAC,KAAK,CAAC,cAAc,KAAK,SAAS,CAC9C,CAAC;IACF,MAAM,YAAY,GAAG,iBAAiB,CAAC,IAAI,EAAE,WAAW,CAAC;QACvD,CAAC,CAAC,SAAS;QACX,CAAC,CAAC,WAAW,EAAE,cAAc,CAAC;IAChC,MAAM,YAAY,GAAG,MAAM,EAAE,aAAa,CAAC;IAC3C,IAAI,MAAuC,CAAC;IAC5C,IAAI,YAAY,KAAK,SAAS,EAAE,CAAC;QAC/B,MAAM,GAAG,qBAAqB,CAAC,IAAI,EAAE,YAAY,CAAC,CAAC;IACrD,CAAC;SAAM,IAAI,iBAAiB,CAAC,IAAI,CAAC,IAAI,YAAY,KAAK,SAAS,EAAE,CAAC;QACjE,MAAM,QAAQ,GAAG,iBAAiB,CAAC,IAAI,EAAE,YAAY,CAAC,CAAC;QACvD,MAAM,GAAG,QAAQ,KAAK,SAAS,CAAC,CAAC,CAAC,SAAS,CAAC,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAC;IAC3D,CAAC;IACD,IACE,MAAM,KAAK,SAAS;QACpB,CAAC,aAAa,CAAC,MAAM,EAAE,OAAO,CAAC,IAAI,UAAU,CAAC,MAAM,EAAE,aAAa,CAAC,CAAC,EACrE,CAAC;QACD,OAAO,SAAS,CAAC;IACnB,CAAC;IACD,OAAO,MAAM,CAAC;AAChB,CAAC"}
+{"version":3,"file":"taiwan-reading.js","sourceRoot":"","sources":["../../src/dictionary/taiwan-reading.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;GAsBG;AACH,OAAO,EAAE,iBAAiB,EAAE,MAAM,yBAAyB,CAAC;AAK5D,OAAO,EAAE,UAAU,EAAE,MAAM,oBAAoB,CAAC;AAChD,OAAO,EAAE,aAAa,EAAE,MAAM,YAAY,CAAC;AAC3C,OAAO,EAAqB,cAAc,EAAE,MAAM,mBAAmB,CAAC;AACtE,OAAO,EACL,YAAY,EACZ,SAAS,EAET,kBAAkB,EAClB,kBAAkB,GACnB,MAAM,wBAAwB,CAAC;AAoBhC;;;;;;;;;;GAUG;AACH,SAAS,aAAa,CACpB,QAAoC,EACpC,KAA+B,EAC/B,OAA4B,EAC5B,OAA6D;IAE7D,IAAI,QAAQ,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QAC1B,OAAO,KAAK,KAAK,SAAS,IAAI,OAAO,CAAC,KAAK,CAAC,OAAO,CAAC;YAClD,CAAC,CAAC,EAAE,OAAO,EAAE,KAAK,CAAC,OAAO,EAAE,UAAU,EAAE,EAAE,EAAE,UAAU,EAAE,KAAK,EAAE;YAC/D,CAAC,CAAC,EAAE,UAAU,EAAE,EAAE,EAAE,UAAU,EAAE,KAAK,EAAE,CAAC;IAC5C,CAAC;IACD,MAAM,QAAQ,GAAG,QAAQ,CAAC,GAAG,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC;IAC1D,MAAM,SAAS,GAAG,QAAQ,CAAC,IAAI,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,MAAM,CAAC,UAAU,KAAK,SAAS,CAAC,CAAC;IAC7E,MAAM,MAAM,GAAG,kBAAkB,CAAC,QAAQ,EAAE,OAAO,EAAE,OAAO,CAAC,CAAC;IAC9D,MAAM,MAAM,GACV,MAAM,KAAK,SAAS,IAAI,CAAC,SAAS,IAAI,SAAS,CAAC,MAAM,EAAE,OAAO,CAAC,CAAC;QAC/D,CAAC,CAAC,MAAM;QACR,CAAC,CAAC,SAAS,CAAC;IAChB,OAAO;QACL,GAAG,CAAC,MAAM,KAAK,SAAS,IAAI,EAAE,OAAO,EAAE,MAAM,EAAE,CAAC;QAChD,sEAAsE;QACtE,iBAAiB;QACjB,UAAU,EAAE,SAAS,CAAC,CAAC,CAAC,YAAY,CAAC,QAAQ,EAAE,MAAM,EAAE,OAAO,CAAC,CAAC,CAAC,CAAC,EAAE;QACpE,UAAU,EAAE,IAAI;KACjB,CAAC;AACJ,CAAC;AAED;;;;;;;;;;;;;GAaG;AACH,SAAS,kBAAkB,CACzB,QAAoC,EACpC,KAA+B,EAC/B,OAA6D;IAE7D,MAAM,UAAU,GAAG,QAAQ,CAAC,MAAM,GAAG,CAAC,CAAC;IACvC,IAAI,KAAK,KAAK,SAAS,IAAI,CAAC,OAAO,CAAC,KAAK,CAAC,OAAO,CAAC,EAAE,CAAC;QACnD,OAAO,EAAE,UAAU,EAAE,EAAE,EAAE,UAAU,EAAE,CAAC;IACxC,CAAC;IACD,MAAM,QAAQ,GAAG,QAAQ,CAAC,IAAI,CAAC,CAAC,MAAM,EAAE,EAAE,CACxC,aAAa,CAAC,MAAM,CAAC,OAAO,EAAE,KAAK,CAAC,OAAO,CAAC,CAC7C,CAAC;IACF,IAAI,UAAU,IAAI,CAAC,QAAQ,EAAE,CAAC;QAC5B,OAAO,EAAE,UAAU,EAAE,EAAE,EAAE,UAAU,EAAE,CAAC;IACxC,CAAC;IACD,OAAO,KAAK,CAAC,aAAa;QACxB,CAAC,CAAC,EAAE,UAAU,EAAE,CAAC,KAAK,CAAC,OAAO,CAAC,EAAE,UAAU,EAAE;QAC7C,CAAC,CAAC,EAAE,OAAO,EAAE,KAAK,CAAC,OAAO,EAAE,UAAU,EAAE,EAAE,EAAE,UAAU,EAAE,CAAC;AAC7D,CAAC;AAED;;;;GAIG;AACH,MAAM,UAAU,eAAe,CAC7B,IAAY,EACZ,SAA4B,EAC5B,MAA8B,EAC9B,aAA+C,EAC/C,OAA4B,EAC5B,MAAkC,EAClC,GAA+C;IAE/C,kEAAkE;IAClE,MAAM,OAAO,GAAG,CAAC,MAAuC,EAAW,EAAE,CACnE,MAAM,KAAK,SAAS;QACpB,CAAC,aAAa,CAAC,MAAM,EAAE,OAAO,CAAC;QAC/B,CAAC,UAAU,CAAC,MAAM,EAAE,aAAa,CAAC,CAAC;IAErC,MAAM,KAAK,GAAG,cAAc,CAAC,IAAI,EAAE,MAAM,EAAE,MAAM,CAAC,CAAC;IACnD,IAAI,CAAC,iBAAiB,CAAC,IAAI,CAAC,EAAE,CAAC;QAC7B,OAAO,aAAa,CAClB,kBAAkB,CAAC,SAAS,EAAE,GAAG,CAAC,EAClC,KAAK,EACL,OAAO,EACP,OAAO,CACR,CAAC;IACJ,CAAC;IACD,oEAAoE;IACpE,oEAAoE;IACpE,oBAAoB;IACpB,MAAM,MAAM,GAAG,SAAS,CAAC,OAAO,CAAC,CAAC,QAAQ,EAAE,EAAE,CAC5C,kBAAkB,CAAC,CAAC,QAAQ,CAAC,EAAE,GAAG,CAAC,CACpC,CAAC;IACF,OAAO,kBAAkB,CAAC,MAAM,EAAE,KAAK,EAAE,OAAO,CAAC,CAAC;AACpD,CAAC"}

package/dist/dictionary/taiwan-settle.d.ts

@@ -0,0 +1,22 @@
+import type { Syllable } from "../syllable/syllable.js";
+import type { EntryReadings } from "./entry.js";
+import type { WordSources } from "./merge-types.js";
+import type { TraditionalForm } from "./traditional-form.js";
+import { type TaiwanReading } from "./taiwan-reading.js";
+/**
+ * What a settled 國語 reading adds to the build's counts.
+ */
+declare function taiwanTallyOf(taiwan: TaiwanReading): {
+    readonly taiwanReadings: number;
+    readonly standardTaiwanReadings: number;
+};
+/**
+ * Settle a word's 國語 reading, as the fields it fills on the entry and what
+ * it adds to the build's counts.
+ */
+export declare function settleTaiwanReading(word: string, form: TraditionalForm, senseReadings: readonly (readonly Syllable[])[], reading: readonly Syllable[], sources: WordSources): {
+    readonly readings: Pick<EntryReadings, "tw" | "twAlternates">;
+    readonly tally: ReturnType<typeof taiwanTallyOf>;
+};
+export {};
+//# sourceMappingURL=taiwan-settle.d.ts.map

package/dist/dictionary/taiwan-settle.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"taiwan-settle.d.ts","sourceRoot":"","sources":["../../src/dictionary/taiwan-settle.ts"],"names":[],"mappings":"AAKA,OAAO,KAAK,EAAE,QAAQ,EAAE,MAAM,yBAAyB,CAAC;AACxD,OAAO,KAAK,EAAE,aAAa,EAAE,MAAM,YAAY,CAAC;AAChD,OAAO,KAAK,EAAE,WAAW,EAAE,MAAM,kBAAkB,CAAC;AACpD,OAAO,KAAK,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAC;AAC7D,OAAO,EAAE,KAAK,aAAa,EAAmB,MAAM,qBAAqB,CAAC;AAc1E;;GAEG;AACH,iBAAS,aAAa,CAAC,MAAM,EAAE,aAAa,GAAG;IAC7C,QAAQ,CAAC,cAAc,EAAE,MAAM,CAAC;IAChC,QAAQ,CAAC,sBAAsB,EAAE,MAAM,CAAC;CACzC,CAMA;AAoBD;;;GAGG;AACH,wBAAgB,mBAAmB,CACjC,IAAI,EAAE,MAAM,EACZ,IAAI,EAAE,eAAe,EACrB,aAAa,EAAE,SAAS,CAAC,SAAS,QAAQ,EAAE,CAAC,EAAE,EAC/C,OAAO,EAAE,SAAS,QAAQ,EAAE,EAC5B,OAAO,EAAE,WAAW,GACnB;IACD,QAAQ,CAAC,QAAQ,EAAE,IAAI,CAAC,aAAa,EAAE,IAAI,GAAG,cAAc,CAAC,CAAC;IAC9D,QAAQ,CAAC,KAAK,EAAE,UAAU,CAAC,OAAO,aAAa,CAAC,CAAC;CAClD,CAWA"}

package/dist/dictionary/taiwan-settle.js

@@ -0,0 +1,47 @@
+/**
+ * A word's settled 國語 reading in the shape the merge stores it: the fields
+ * it fills on the entry, and what it adds to the build's counts.
+ */
+import { isSingleCharacter } from "../script/characters.js";
+import { taiwanReadingOf } from "./taiwan-reading.js";
+/**
+ * The fields of an entry's readings that a settled 國語 reading fills.
+ */
+function taiwanColumnsOf(taiwan) {
+    return {
+        ...(taiwan.reading !== undefined && { tw: taiwan.reading }),
+        ...(taiwan.alternates.length > 0 && { twAlternates: taiwan.alternates }),
+    };
+}
+/**
+ * What a settled 國語 reading adds to the build's counts.
+ */
+function taiwanTallyOf(taiwan) {
+    const isDelta = taiwan.reading !== undefined;
+    return {
+        taiwanReadings: isDelta ? 1 : 0,
+        standardTaiwanReadings: isDelta && taiwan.isStandard ? 1 : 0,
+    };
+}
+/**
+ * The spellings 教育部 is asked for a word under, which is how it keys it.
+ *
+ * A character's other 繁體 forms are other characters, which 教育部 reads as
+ * themselves: 最's include 冣, which it lists as `jù`. A word's are the same
+ * word, and its 简体 spelling is only asked where it is also the 繁體 one:
+ * 教育部 lists 听听, written with the 繁體 character 听, as `yǐn yǐn`.
+ */
+function spellingsOf(word, hant, hantVariants) {
+    return isSingleCharacter(word)
+        ? [...new Set([hant, word])]
+        : [hant, ...hantVariants];
+}
+/**
+ * Settle a word's 國語 reading, as the fields it fills on the entry and what
+ * it adds to the build's counts.
+ */
+export function settleTaiwanReading(word, form, senseReadings, reading, sources) {
+    const taiwan = taiwanReadingOf(word, spellingsOf(word, form.hant, form.hantVariants), form.senses, senseReadings, reading, sources.unihanReadings.get(word), sources.moe);
+    return { readings: taiwanColumnsOf(taiwan), tally: taiwanTallyOf(taiwan) };
+}
+//# sourceMappingURL=taiwan-settle.js.map

package/dist/dictionary/taiwan-settle.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"taiwan-settle.js","sourceRoot":"","sources":["../../src/dictionary/taiwan-settle.ts"],"names":[],"mappings":"AAAA;;;GAGG;AACH,OAAO,EAAE,iBAAiB,EAAE,MAAM,yBAAyB,CAAC;AAK5D,OAAO,EAAsB,eAAe,EAAE,MAAM,qBAAqB,CAAC;AAE1E;;GAEG;AACH,SAAS,eAAe,CACtB,MAAqB;IAErB,OAAO;QACL,GAAG,CAAC,MAAM,CAAC,OAAO,KAAK,SAAS,IAAI,EAAE,EAAE,EAAE,MAAM,CAAC,OAAO,EAAE,CAAC;QAC3D,GAAG,CAAC,MAAM,CAAC,UAAU,CAAC,MAAM,GAAG,CAAC,IAAI,EAAE,YAAY,EAAE,MAAM,CAAC,UAAU,EAAE,CAAC;KACzE,CAAC;AACJ,CAAC;AAED;;GAEG;AACH,SAAS,aAAa,CAAC,MAAqB;IAI1C,MAAM,OAAO,GAAG,MAAM,CAAC,OAAO,KAAK,SAAS,CAAC;IAC7C,OAAO;QACL,cAAc,EAAE,OAAO,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;QAC/B,sBAAsB,EAAE,OAAO,IAAI,MAAM,CAAC,UAAU,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;KAC7D,CAAC;AACJ,CAAC;AAED;;;;;;;GAOG;AACH,SAAS,WAAW,CAClB,IAAY,EACZ,IAAY,EACZ,YAA+B;IAE/B,OAAO,iBAAiB,CAAC,IAAI,CAAC;QAC5B,CAAC,CAAC,CAAC,GAAG,IAAI,GAAG,CAAC,CAAC,IAAI,EAAE,IAAI,CAAC,CAAC,CAAC;QAC5B,CAAC,CAAC,CAAC,IAAI,EAAE,GAAG,YAAY,CAAC,CAAC;AAC9B,CAAC;AAED;;;GAGG;AACH,MAAM,UAAU,mBAAmB,CACjC,IAAY,EACZ,IAAqB,EACrB,aAA+C,EAC/C,OAA4B,EAC5B,OAAoB;IAKpB,MAAM,MAAM,GAAG,eAAe,CAC5B,IAAI,EACJ,WAAW,CAAC,IAAI,EAAE,IAAI,CAAC,IAAI,EAAE,IAAI,CAAC,YAAY,CAAC,EAC/C,IAAI,CAAC,MAAM,EACX,aAAa,EACb,OAAO,EACP,OAAO,CAAC,cAAc,CAAC,GAAG,CAAC,IAAI,CAAC,EAChC,OAAO,CAAC,GAAG,CACZ,CAAC;IACF,OAAO,EAAE,QAAQ,EAAE,eAAe,CAAC,MAAM,CAAC,EAAE,KAAK,EAAE,aAAa,CAAC,MAAM,CAAC,EAAE,CAAC;AAC7E,CAAC"}

package/dist/search/match-readings.d.ts

@@ -18,8 +18,8 @@ export declare class CharacterReadings {
      * glyph form is matched by the readings of the Taiwan-standard key it folds
      * to rather than by nothing.
      *
-     * The 國語 reading goes on the end where it differs, so that 圾 answers to
-     * `se` as well as to `ji`. A search box is not the place to insist on one
+     * The 國語 readings go on the end where they differ, so that 圾 answers to
+     * `se` as well as to `ji` and 和 to `han` as well as to `he`. A search box is not the place to insist on one
      * standard: somebody typing what they say is not typing it wrongly, and the
      * decoder's own reading is what ranks the two afterwards.
      */

package/dist/search/match-readings.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"match-readings.d.ts","sourceRoot":"","sources":["../../src/search/match-readings.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AACH,OAAO,KAAK,EAAE,UAAU,EAAE,MAAM,6BAA6B,CAAC;AAI9D,OAAO,KAAK,EAAE,QAAQ,EAAE,MAAM,yBAAyB,CAAC;AACxD,OAAO,KAAK,EAAE,SAAS,EAAE,MAAM,oBAAoB,CAAC;AAEpD,qBAAa,iBAAiB;;IAK5B,YAAY,UAAU,EAAE,UAAU,EAAE,UAAU,EAAE,SAAS,MAAM,EAAE,EAGhE;IAED;;;;;;;;;;;OAWG;IACH,EAAE,CAAC,EAAE,EAAE,MAAM,GAAG,SAAS,CAAC,SAAS,QAAQ,EAAE,CAAC,EAAE,CAgB/C;IAED;;;;;;;;;;;;;OAaG;IACH,YAAY,CAAC,EAAE,EAAE,MAAM,GAAG,SAAS,SAAS,EAAE,CA2B7C;CACF"}
+{"version":3,"file":"match-readings.d.ts","sourceRoot":"","sources":["../../src/search/match-readings.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AACH,OAAO,KAAK,EAAE,UAAU,EAAE,MAAM,6BAA6B,CAAC;AAI9D,OAAO,KAAK,EAAE,QAAQ,EAAE,MAAM,yBAAyB,CAAC;AACxD,OAAO,KAAK,EAAE,SAAS,EAAE,MAAM,oBAAoB,CAAC;AAEpD,qBAAa,iBAAiB;;IAK5B,YAAY,UAAU,EAAE,UAAU,EAAE,UAAU,EAAE,SAAS,MAAM,EAAE,EAGhE;IAED;;;;;;;;;;;OAWG;IACH,EAAE,CAAC,EAAE,EAAE,MAAM,GAAG,SAAS,CAAC,SAAS,QAAQ,EAAE,CAAC,EAAE,CAoB/C;IAED;;;;;;;;;;;;;OAaG;IACH,YAAY,CAAC,EAAE,EAAE,MAAM,GAAG,SAAS,SAAS,EAAE,CA2B7C;CACF"}

package/dist/search/match-readings.js

@@ -16,8 +16,8 @@ export class CharacterReadings {
      * glyph form is matched by the readings of the Taiwan-standard key it folds
      * to rather than by nothing.
      *
-     * The 國語 reading goes on the end where it differs, so that 圾 answers to
-     * `se` as well as to `ji`. A search box is not the place to insist on one
+     * The 國語 readings go on the end where they differ, so that 圾 answers to
+     * `se` as well as to `ji` and 和 to `han` as well as to `he`. A search box is not the place to insist on one
      * standard: somebody typing what they say is not typing it wrongly, and the
      * decoder's own reading is what ranks the two afterwards.
      */
@@ -29,11 +29,16 @@ export class CharacterReadings {
         /* c8 ignore next -- only ever asked about a character that is there */
         const character = toCanonicalGlyphs(this.#characters[at] ?? "");
         const found = this.#dictionary.readingsOf(character);
-        const taiwan = this.#dictionary.lookup(character)?.taiwanReading;
-        const readings = taiwan === undefined ||
-            found.some((reading) => isSameReading(reading, taiwan))
-            ? found
-            : [...found, taiwan];
+        const entry = this.#dictionary.lookup(character);
+        const readings = [...found];
+        for (const taiwan of [
+            ...(entry?.taiwanReading === undefined ? [] : [entry.taiwanReading]),
+            ...(entry?.taiwanAlternates ?? []),
+        ]) {
+            if (!readings.some((reading) => isSameReading(reading, taiwan))) {
+                readings.push(taiwan);
+            }
+        }
         this.#readings[at] = readings;
         return readings;
     }

package/dist/search/match-readings.js.map

@@ -1 +1 @@
-{"version":3,"file":"match-readings.js","sourceRoot":"","sources":["../../src/search/match-readings.ts"],"names":[],"mappings":"AAQA,OAAO,EAAE,aAAa,EAAE,MAAM,wBAAwB,CAAC;AACvD,OAAO,EAAE,aAAa,EAAE,SAAS,EAAE,MAAM,wBAAwB,CAAC;AAClE,OAAO,EAAE,iBAAiB,EAAE,MAAM,qBAAqB,CAAC;AAIxD,MAAM,OAAO,iBAAiB;IACnB,WAAW,CAAa;IACxB,WAAW,CAAoB;IAC/B,SAAS,GAAyC,EAAE,CAAC;IAE9D,YAAY,UAAsB,EAAE,UAA6B;QAC/D,IAAI,CAAC,WAAW,GAAG,UAAU,CAAC;QAC9B,IAAI,CAAC,WAAW,GAAG,UAAU,CAAC;IAChC,CAAC;IAED;;;;;;;;;;;OAWG;IACH,EAAE,CAAC,EAAU;QACX,MAAM,IAAI,GAAG,IAAI,CAAC,SAAS,CAAC,EAAE,CAAC,CAAC;QAChC,IAAI,IAAI,KAAK,SAAS,EAAE,CAAC;YACvB,OAAO,IAAI,CAAC;QACd,CAAC;QACD,uEAAuE;QACvE,MAAM,SAAS,GAAG,iBAAiB,CAAC,IAAI,CAAC,WAAW,CAAC,EAAE,CAAC,IAAI,EAAE,CAAC,CAAC;QAChE,MAAM,KAAK,GAAG,IAAI,CAAC,WAAW,CAAC,UAAU,CAAC,SAAS,CAAC,CAAC;QACrD,MAAM,MAAM,GAAG,IAAI,CAAC,WAAW,CAAC,MAAM,CAAC,SAAS,CAAC,EAAE,aAAa,CAAC;QACjE,MAAM,QAAQ,GACZ,MAAM,KAAK,SAAS;YACpB,KAAK,CAAC,IAAI,CAAC,CAAC,OAAO,EAAE,EAAE,CAAC,aAAa,CAAC,OAAO,EAAE,MAAM,CAAC,CAAC;YACrD,CAAC,CAAC,KAAK;YACP,CAAC,CAAC,CAAC,GAAG,KAAK,EAAE,MAAM,CAAC,CAAC;QACzB,IAAI,CAAC,SAAS,CAAC,EAAE,CAAC,GAAG,QAAQ,CAAC;QAC9B,OAAO,QAAQ,CAAC;IAClB,CAAC;IAED;;;;;;;;;;;;;OAaG;IACH,YAAY,CAAC,EAAU;QACrB,MAAM,QAAQ,GAAG,IAAI,CAAC,EAAE,CAAC,EAAE,CAAC,CAAC;QAC7B,MAAM,KAAK,GAAG,QAAQ,CAAC,GAAG,CAAC,CAAC,OAAO,EAAE,IAAI,EAAE,EAAE,CAAC,CAAC;YAC7C,OAAO;YACP,IAAI;YACJ,UAAU,EAAE,CAAC;SACd,CAAC,CAAC,CAAC;QACJ,IAAI,CAAC,aAAa,CAAC,IAAI,CAAC,WAAW,CAAC,EAAE,GAAG,CAAC,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC;YACnD,OAAO,KAAK,CAAC;QACf,CAAC;QACD,OAAO;YACL,GAAG,KAAK;YACR,GAAG,QAAQ,CAAC,OAAO,CAAC,CAAC,OAAO,EAAE,IAAI,EAAE,EAAE;gBACpC,MAAM,IAAI,GAAG,OAAO,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC;gBAC5B,yEAAyE;gBACzE,IAAI,IAAI,KAAK,SAAS,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,EAAE,CAAC;oBAC9C,OAAO,EAAE,CAAC;gBACZ,CAAC;gBACD,OAAO;oBACL;wBACE,OAAO,EAAE,CAAC,GAAG,OAAO,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,EAAE,SAAS,CAAC,IAAI,CAAC,CAAC;wBACnD,IAAI;wBACJ,UAAU,EAAE,CAAC;qBACd;iBACF,CAAC;YACJ,CAAC,CAAC;SACH,CAAC;IACJ,CAAC;CACF"}
+{"version":3,"file":"match-readings.js","sourceRoot":"","sources":["../../src/search/match-readings.ts"],"names":[],"mappings":"AAQA,OAAO,EAAE,aAAa,EAAE,MAAM,wBAAwB,CAAC;AACvD,OAAO,EAAE,aAAa,EAAE,SAAS,EAAE,MAAM,wBAAwB,CAAC;AAClE,OAAO,EAAE,iBAAiB,EAAE,MAAM,qBAAqB,CAAC;AAIxD,MAAM,OAAO,iBAAiB;IACnB,WAAW,CAAa;IACxB,WAAW,CAAoB;IAC/B,SAAS,GAAyC,EAAE,CAAC;IAE9D,YAAY,UAAsB,EAAE,UAA6B;QAC/D,IAAI,CAAC,WAAW,GAAG,UAAU,CAAC;QAC9B,IAAI,CAAC,WAAW,GAAG,UAAU,CAAC;IAChC,CAAC;IAED;;;;;;;;;;;OAWG;IACH,EAAE,CAAC,EAAU;QACX,MAAM,IAAI,GAAG,IAAI,CAAC,SAAS,CAAC,EAAE,CAAC,CAAC;QAChC,IAAI,IAAI,KAAK,SAAS,EAAE,CAAC;YACvB,OAAO,IAAI,CAAC;QACd,CAAC;QACD,uEAAuE;QACvE,MAAM,SAAS,GAAG,iBAAiB,CAAC,IAAI,CAAC,WAAW,CAAC,EAAE,CAAC,IAAI,EAAE,CAAC,CAAC;QAChE,MAAM,KAAK,GAAG,IAAI,CAAC,WAAW,CAAC,UAAU,CAAC,SAAS,CAAC,CAAC;QACrD,MAAM,KAAK,GAAG,IAAI,CAAC,WAAW,CAAC,MAAM,CAAC,SAAS,CAAC,CAAC;QACjD,MAAM,QAAQ,GAAG,CAAC,GAAG,KAAK,CAAC,CAAC;QAC5B,KAAK,MAAM,MAAM,IAAI;YACnB,GAAG,CAAC,KAAK,EAAE,aAAa,KAAK,SAAS,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,aAAa,CAAC,CAAC;YACpE,GAAG,CAAC,KAAK,EAAE,gBAAgB,IAAI,EAAE,CAAC;SACnC,EAAE,CAAC;YACF,IAAI,CAAC,QAAQ,CAAC,IAAI,CAAC,CAAC,OAAO,EAAE,EAAE,CAAC,aAAa,CAAC,OAAO,EAAE,MAAM,CAAC,CAAC,EAAE,CAAC;gBAChE,QAAQ,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC;YACxB,CAAC;QACH,CAAC;QACD,IAAI,CAAC,SAAS,CAAC,EAAE,CAAC,GAAG,QAAQ,CAAC;QAC9B,OAAO,QAAQ,CAAC;IAClB,CAAC;IAED;;;;;;;;;;;;;OAaG;IACH,YAAY,CAAC,EAAU;QACrB,MAAM,QAAQ,GAAG,IAAI,CAAC,EAAE,CAAC,EAAE,CAAC,CAAC;QAC7B,MAAM,KAAK,GAAG,QAAQ,CAAC,GAAG,CAAC,CAAC,OAAO,EAAE,IAAI,EAAE,EAAE,CAAC,CAAC;YAC7C,OAAO;YACP,IAAI;YACJ,UAAU,EAAE,CAAC;SACd,CAAC,CAAC,CAAC;QACJ,IAAI,CAAC,aAAa,CAAC,IAAI,CAAC,WAAW,CAAC,EAAE,GAAG,CAAC,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC;YACnD,OAAO,KAAK,CAAC;QACf,CAAC;QACD,OAAO;YACL,GAAG,KAAK;YACR,GAAG,QAAQ,CAAC,OAAO,CAAC,CAAC,OAAO,EAAE,IAAI,EAAE,EAAE;gBACpC,MAAM,IAAI,GAAG,OAAO,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC;gBAC5B,yEAAyE;gBACzE,IAAI,IAAI,KAAK,SAAS,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,EAAE,CAAC;oBAC9C,OAAO,EAAE,CAAC;gBACZ,CAAC;gBACD,OAAO;oBACL;wBACE,OAAO,EAAE,CAAC,GAAG,OAAO,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,EAAE,SAAS,CAAC,IAAI,CAAC,CAAC;wBACnD,IAAI;wBACJ,UAAU,EAAE,CAAC;qBACd;iBACF,CAAC;YACJ,CAAC,CAAC;SACH,CAAC;IACJ,CAAC;CACF"}

package/dist/sources/moe-workbook.d.ts

@@ -0,0 +1,10 @@
+/**
+ * Turn a dictionary's worksheet into a table of headword, order and reading.
+ *
+ * Done once, when the download is cached, so the build reads a small table and
+ * never the 125 MB of XML the spreadsheet is made of. The output is one
+ * tab-separated line per reading, with a fourth column where the entry gives
+ * a variant reading, and the header row is dropped.
+ */
+export declare function tabulateMoeWorkbook(sharedStringsXml: string, sheetXml: string): string;
+//# sourceMappingURL=moe-workbook.d.ts.map

package/dist/sources/moe-workbook.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"moe-workbook.d.ts","sourceRoot":"","sources":["../../src/sources/moe-workbook.ts"],"names":[],"mappings":"AA+BA;;;;;;;GAOG;AACH,wBAAgB,mBAAmB,CACjC,gBAAgB,EAAE,MAAM,EACxB,QAAQ,EAAE,MAAM,GACf,MAAM,CA4CR"}

package/dist/sources/moe-workbook.js

@@ -0,0 +1,89 @@
+/**
+ * Taking the readings out of a 教育部 dictionary's workbook.
+ *
+ * The workbook has seventeen columns, and only the headword, the order of its
+ * readings, its 漢語拼音 and its variant reading are read. Definitions,
+ * examples, synonyms and the rest are never parsed. See `moe.ts` for what is
+ * done with the readings.
+ */
+import { HEADER_ROW, readRows, readSharedStrings } from "./xlsx.js";
+/**
+ * The columns this package reads, by the name the header row gives them.
+ *
+ * Found by name rather than by letter, because the two dictionaries lay their
+ * columns out differently: 《重編國語辭典修訂本》 has 漢語拼音 in column L and
+ * 《國語辭典簡編本》 in column J. A release that drops one fails the build
+ * rather than reading the wrong column.
+ */
+const HEADERS = {
+    word: "字詞名",
+    order: "多音排序",
+    pinyin: "漢語拼音",
+    variantKind: "變體類型",
+    variant: "變體漢語拼音",
+};
+/**
+ * The kinds of variant reading that are readings of the word itself.
+ */
+const VARIANT_KINDS = new Set(["又音", "語音", "讀音", "或讀"]);
+/**
+ * Turn a dictionary's worksheet into a table of headword, order and reading.
+ *
+ * Done once, when the download is cached, so the build reads a small table and
+ * never the 125 MB of XML the spreadsheet is made of. The output is one
+ * tab-separated line per reading, with a fourth column where the entry gives
+ * a variant reading, and the header row is dropped.
+ */
+export function tabulateMoeWorkbook(sharedStringsXml, sheetXml) {
+    const strings = readSharedStrings(sharedStringsXml);
+    const header = readRows(HEADER_ROW.exec(sheetXml)?.[0] ?? "", strings).get(1);
+    // 變體類型 carries its legend in the header, and the two dictionaries'
+    // legends differ, so a header is matched on how it starts.
+    const columnOf = (name) => {
+        const found = [...(header ?? [])].find(([, text]) => text.startsWith(name))?.[0];
+        if (found === undefined) {
+            throw new Error(`教育部 dictionary has no ${name} column`);
+        }
+        return found;
+    };
+    const columns = {
+        word: columnOf(HEADERS.word),
+        order: columnOf(HEADERS.order),
+        pinyin: columnOf(HEADERS.pinyin),
+        variantKind: columnOf(HEADERS.variantKind),
+        variant: columnOf(HEADERS.variant),
+    };
+    const rows = readRows(sheetXml, strings, new Set(Object.values(columns)));
+    const lines = [];
+    for (const [row, cells] of rows) {
+        const word = cells.get(columns.word) ?? "";
+        const pinyin = cells.get(columns.pinyin) ?? "";
+        if (row === 1 || word === "" || pinyin === "") {
+            continue;
+        }
+        const kind = cells.get(columns.variantKind) ?? "";
+        const variant = VARIANT_KINDS.has(kind)
+            ? (cells.get(columns.variant) ?? "")
+            : "";
+        lines.push([
+            word,
+            orderOf(cells.get(columns.order)),
+            spaced(pinyin),
+            ...(variant === "" ? [] : [spaced(variant)]),
+        ].join("\t"));
+    }
+    return `${lines.join("\n")}\n`;
+}
+/**
+ * A headword's 多音排序, which is blank for a headword with one reading.
+ */
+function orderOf(order) {
+    return order === undefined || order === "" ? "0" : order;
+}
+/**
+ * A reading with its syllables separated by single spaces.
+ */
+function spaced(pinyin) {
+    return pinyin.replaceAll(/\s+/gu, " ");
+}
+//# sourceMappingURL=moe-workbook.js.map

package/dist/sources/moe-workbook.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"moe-workbook.js","sourceRoot":"","sources":["../../src/sources/moe-workbook.ts"],"names":[],"mappings":"AAAA;;;;;;;GAOG;AACH,OAAO,EAAE,UAAU,EAAE,QAAQ,EAAE,iBAAiB,EAAE,MAAM,WAAW,CAAC;AAEpE;;;;;;;GAOG;AACH,MAAM,OAAO,GAAG;IACd,IAAI,EAAE,KAAK;IACX,KAAK,EAAE,MAAM;IACb,MAAM,EAAE,MAAM;IACd,WAAW,EAAE,MAAM;IACnB,OAAO,EAAE,QAAQ;CACT,CAAC;AAEX;;GAEG;AACH,MAAM,aAAa,GAAG,IAAI,GAAG,CAAC,CAAC,IAAI,EAAE,IAAI,EAAE,IAAI,EAAE,IAAI,CAAC,CAAC,CAAC;AAExD;;;;;;;GAOG;AACH,MAAM,UAAU,mBAAmB,CACjC,gBAAwB,EACxB,QAAgB;IAEhB,MAAM,OAAO,GAAG,iBAAiB,CAAC,gBAAgB,CAAC,CAAC;IACpD,MAAM,MAAM,GAAG,QAAQ,CAAC,UAAU,CAAC,IAAI,CAAC,QAAQ,CAAC,EAAE,CAAC,CAAC,CAAC,IAAI,EAAE,EAAE,OAAO,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC;IAC9E,mEAAmE;IACnE,2DAA2D;IAC3D,MAAM,QAAQ,GAAG,CAAC,IAAY,EAAU,EAAE;QACxC,MAAM,KAAK,GAAG,CAAC,GAAG,CAAC,MAAM,IAAI,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,IAAI,CAAC,EAAE,EAAE,CAClD,IAAI,CAAC,UAAU,CAAC,IAAI,CAAC,CACtB,EAAE,CAAC,CAAC,CAAC,CAAC;QACP,IAAI,KAAK,KAAK,SAAS,EAAE,CAAC;YACxB,MAAM,IAAI,KAAK,CAAC,yBAAyB,IAAI,SAAS,CAAC,CAAC;QAC1D,CAAC;QACD,OAAO,KAAK,CAAC;IACf,CAAC,CAAC;IACF,MAAM,OAAO,GAAG;QACd,IAAI,EAAE,QAAQ,CAAC,OAAO,CAAC,IAAI,CAAC;QAC5B,KAAK,EAAE,QAAQ,CAAC,OAAO,CAAC,KAAK,CAAC;QAC9B,MAAM,EAAE,QAAQ,CAAC,OAAO,CAAC,MAAM,CAAC;QAChC,WAAW,EAAE,QAAQ,CAAC,OAAO,CAAC,WAAW,CAAC;QAC1C,OAAO,EAAE,QAAQ,CAAC,OAAO,CAAC,OAAO,CAAC;KACnC,CAAC;IACF,MAAM,IAAI,GAAG,QAAQ,CAAC,QAAQ,EAAE,OAAO,EAAE,IAAI,GAAG,CAAC,MAAM,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC;IAE1E,MAAM,KAAK,GAAa,EAAE,CAAC;IAC3B,KAAK,MAAM,CAAC,GAAG,EAAE,KAAK,CAAC,IAAI,IAAI,EAAE,CAAC;QAChC,MAAM,IAAI,GAAG,KAAK,CAAC,GAAG,CAAC,OAAO,CAAC,IAAI,CAAC,IAAI,EAAE,CAAC;QAC3C,MAAM,MAAM,GAAG,KAAK,CAAC,GAAG,CAAC,OAAO,CAAC,MAAM,CAAC,IAAI,EAAE,CAAC;QAC/C,IAAI,GAAG,KAAK,CAAC,IAAI,IAAI,KAAK,EAAE,IAAI,MAAM,KAAK,EAAE,EAAE,CAAC;YAC9C,SAAS;QACX,CAAC;QACD,MAAM,IAAI,GAAG,KAAK,CAAC,GAAG,CAAC,OAAO,CAAC,WAAW,CAAC,IAAI,EAAE,CAAC;QAClD,MAAM,OAAO,GAAG,aAAa,CAAC,GAAG,CAAC,IAAI,CAAC;YACrC,CAAC,CAAC,CAAC,KAAK,CAAC,GAAG,CAAC,OAAO,CAAC,OAAO,CAAC,IAAI,EAAE,CAAC;YACpC,CAAC,CAAC,EAAE,CAAC;QACP,KAAK,CAAC,IAAI,CACR;YACE,IAAI;YACJ,OAAO,CAAC,KAAK,CAAC,GAAG,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC;YACjC,MAAM,CAAC,MAAM,CAAC;YACd,GAAG,CAAC,OAAO,KAAK,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC,CAAC;SAC7C,CAAC,IAAI,CAAC,IAAI,CAAC,CACb,CAAC;IACJ,CAAC;IACD,OAAO,GAAG,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,IAAI,CAAC;AACjC,CAAC;AAED;;GAEG;AACH,SAAS,OAAO,CAAC,KAAyB;IACxC,OAAO,KAAK,KAAK,SAAS,IAAI,KAAK,KAAK,EAAE,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,KAAK,CAAC;AAC3D,CAAC;AAED;;GAEG;AACH,SAAS,MAAM,CAAC,MAAc;IAC5B,OAAO,MAAM,CAAC,UAAU,CAAC,OAAO,EAAE,GAAG,CAAC,CAAC;AACzC,CAAC"}

package/dist/sources/moe.d.ts

@@ -0,0 +1,58 @@
+/**
+ * Which of 教育部's two dictionaries a reading comes from.
+ *
+ * `concise` is 《國語辭典簡編本》, which describes current usage, and
+ * `revised` is 《重編國語辭典修訂本》, which adds the literary and historical.
+ */
+export type MoeDictionary = "concise" | "revised";
+/**
+ * One reading 教育部 gives a headword.
+ */
+export interface MoeReading {
+    readonly dictionary: MoeDictionary;
+    /**
+     * Where the reading stands among the headword's readings.
+     *
+     * 0 for a headword with one reading. A headword with several numbers them
+     * from 1, and 1 is the reading the dictionary lists first: 和 is `hé` at 1
+     * and `hàn` at 5, 期 `qí` at 1 and `qī` at 2.
+     */
+    readonly order: number;
+    /**
+     * The reading as tokens {@link import("../dictionary/reading.js").readAlignedReading}
+     * accepts, one per syllable, with 儿化 split off as its own `r5` token.
+     */
+    readonly readings: readonly string[];
+    /**
+     * A second reading the dictionary gives the same entry, where it gives one.
+     *
+     * Only the kinds that are readings of the word: 又音, an alternative, and
+     * the 語音 and 讀音 pair of colloquial and literary readings. 變, which
+     * records a tone change in connected speech such as 一 before a fourth tone,
+     * is left out, since the merge stores tones before sandhi.
+     */
+    readonly variant?: readonly string[];
+}
+/**
+ * Read the table {@link tabulateMoeWorkbook} writes, keyed by headword.
+ *
+ * Headwords are 繁體, as Taiwan writes them. A headword's readings come back
+ * in the dictionary's own order.
+ */
+export declare function parseMoeReadings(text: string, dictionary: MoeDictionary): ReadonlyMap<string, readonly MoeReading[]>;
+/**
+ * Two dictionaries' readings as one.
+ *
+ * 《國語辭典簡編本》 is the one that describes current usage, and for a word
+ * both list it wins. 《重編國語辭典修訂本》 is four times its size and records
+ * the literary and historical readings too, so where the two list a word it is
+ * often the older one that the larger dictionary leads with: 玩 is `wàn` there
+ * before `wán`, 鹽 `yàn` before `yán`, and 它們 is `tuō men`. Only the words
+ * the concise dictionary leaves out are taken from it.
+ *
+ * A character keeps the readings of both, the concise dictionary's first.
+ * The merge only ever asks whether a character's reading is listed, and the
+ * concise dictionary leaves out some current ones: it has no 差 `chà`.
+ */
+export declare function preferReadings(first: ReadonlyMap<string, readonly MoeReading[]>, second: ReadonlyMap<string, readonly MoeReading[]>): ReadonlyMap<string, readonly MoeReading[]>;
+//# sourceMappingURL=moe.d.ts.map

package/dist/sources/moe.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"moe.d.ts","sourceRoot":"","sources":["../../src/sources/moe.ts"],"names":[],"mappings":"AAaA;;;;;GAKG;AACH,MAAM,MAAM,aAAa,GAAG,SAAS,GAAG,SAAS,CAAC;AAElD;;GAEG;AACH,MAAM,WAAW,UAAU;IACzB,QAAQ,CAAC,UAAU,EAAE,aAAa,CAAC;IACnC;;;;;;OAMG;IACH,QAAQ,CAAC,KAAK,EAAE,MAAM,CAAC;IACvB;;;OAGG;IACH,QAAQ,CAAC,QAAQ,EAAE,SAAS,MAAM,EAAE,CAAC;IACrC;;;;;;;OAOG;IACH,QAAQ,CAAC,OAAO,CAAC,EAAE,SAAS,MAAM,EAAE,CAAC;CACtC;AAqBD;;;;;GAKG;AACH,wBAAgB,gBAAgB,CAC9B,IAAI,EAAE,MAAM,EACZ,UAAU,EAAE,aAAa,GACxB,WAAW,CAAC,MAAM,EAAE,SAAS,UAAU,EAAE,CAAC,CAqB5C;AAED;;;;;;;;;;;;;GAaG;AACH,wBAAgB,cAAc,CAC5B,KAAK,EAAE,WAAW,CAAC,MAAM,EAAE,SAAS,UAAU,EAAE,CAAC,EACjD,MAAM,EAAE,WAAW,CAAC,MAAM,EAAE,SAAS,UAAU,EAAE,CAAC,GACjD,WAAW,CAAC,MAAM,EAAE,SAAS,UAAU,EAAE,CAAC,CAO5C"}

package/dist/sources/moe.js

@@ -0,0 +1,79 @@
+/**
+ * The readings of 教育部's two dictionaries of the 國語 standard,
+ * 《國語辭典簡編本》 and 《重編國語辭典修訂本》.
+ *
+ * Read from the table `moe-workbook.ts` takes out of each download, which holds
+ * a headword, the order of its readings, its 漢語拼音 and any variant reading,
+ * and nothing of the dictionaries' own text. See docs/dictionaries for why the
+ * readings alone are taken and what the licence says about them.
+ */
+import { isSingleCharacter } from "../script/characters.js";
+import { readSyllable } from "../syllable/syllable.js";
+import { ERHUA_TOKEN } from "../dictionary/reading.js";
+/**
+ * Split a reading into the tokens the merge aligns against a headword.
+ *
+ * 教育部 writes 儿化 on the syllable, as `yī huǐr` for 一會兒, where the merge
+ * expects CC-CEDICT's shape of a syllable followed by `r5`. A syllable that
+ * reads as 儿化 is split back into that shape, which leaves 兒 with a token to
+ * account for it; `ér` and `èr` are syllables of their own and stay whole.
+ */
+function tokensOf(pinyin) {
+    return pinyin
+        .split(" ")
+        .filter((token) => token !== "")
+        .flatMap((token) => readSyllable(token)?.erhua === true
+        ? [token.slice(0, -1), ERHUA_TOKEN]
+        : [token]);
+}
+/**
+ * Read the table {@link tabulateMoeWorkbook} writes, keyed by headword.
+ *
+ * Headwords are 繁體, as Taiwan writes them. A headword's readings come back
+ * in the dictionary's own order.
+ */
+export function parseMoeReadings(text, dictionary) {
+    const readings = new Map();
+    for (const line of text.split("\n")) {
+        const [word = "", order = "", pinyin = "", variant = ""] = line.split("\t");
+        if (word === "" || pinyin === "") {
+            continue;
+        }
+        const headword = word.normalize("NFC");
+        const listed = readings.get(headword) ?? [];
+        listed.push({
+            dictionary,
+            order: Number(order) || 0,
+            readings: tokensOf(pinyin),
+            ...(variant !== "" && { variant: tokensOf(variant) }),
+        });
+        readings.set(headword, listed);
+    }
+    for (const listed of readings.values()) {
+        listed.sort((left, right) => left.order - right.order);
+    }
+    return readings;
+}
+/**
+ * Two dictionaries' readings as one.
+ *
+ * 《國語辭典簡編本》 is the one that describes current usage, and for a word
+ * both list it wins. 《重編國語辭典修訂本》 is four times its size and records
+ * the literary and historical readings too, so where the two list a word it is
+ * often the older one that the larger dictionary leads with: 玩 is `wàn` there
+ * before `wán`, 鹽 `yàn` before `yán`, and 它們 is `tuō men`. Only the words
+ * the concise dictionary leaves out are taken from it.
+ *
+ * A character keeps the readings of both, the concise dictionary's first.
+ * The merge only ever asks whether a character's reading is listed, and the
+ * concise dictionary leaves out some current ones: it has no 差 `chà`.
+ */
+export function preferReadings(first, second) {
+    const merged = new Map(second);
+    for (const [word, readings] of first) {
+        const extra = isSingleCharacter(word) ? (second.get(word) ?? []) : [];
+        merged.set(word, [...readings, ...extra]);
+    }
+    return merged;
+}
+//# sourceMappingURL=moe.js.map

package/dist/sources/moe.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"moe.js","sourceRoot":"","sources":["../../src/sources/moe.ts"],"names":[],"mappings":"AAAA;;;;;;;;GAQG;AACH,OAAO,EAAE,iBAAiB,EAAE,MAAM,yBAAyB,CAAC;AAC5D,OAAO,EAAE,YAAY,EAAE,MAAM,yBAAyB,CAAC;AACvD,OAAO,EAAE,WAAW,EAAE,MAAM,0BAA0B,CAAC;AAuCvD;;;;;;;GAOG;AACH,SAAS,QAAQ,CAAC,MAAc;IAC9B,OAAO,MAAM;SACV,KAAK,CAAC,GAAG,CAAC;SACV,MAAM,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,KAAK,KAAK,EAAE,CAAC;SAC/B,OAAO,CAAC,CAAC,KAAK,EAAE,EAAE,CACjB,YAAY,CAAC,KAAK,CAAC,EAAE,KAAK,KAAK,IAAI;QACjC,CAAC,CAAC,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,EAAE,WAAW,CAAC;QACnC,CAAC,CAAC,CAAC,KAAK,CAAC,CACZ,CAAC;AACN,CAAC;AAED;;;;;GAKG;AACH,MAAM,UAAU,gBAAgB,CAC9B,IAAY,EACZ,UAAyB;IAEzB,MAAM,QAAQ,GAAG,IAAI,GAAG,EAAwB,CAAC;IACjD,KAAK,MAAM,IAAI,IAAI,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,EAAE,CAAC;QACpC,MAAM,CAAC,IAAI,GAAG,EAAE,EAAE,KAAK,GAAG,EAAE,EAAE,MAAM,GAAG,EAAE,EAAE,OAAO,GAAG,EAAE,CAAC,GAAG,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;QAC5E,IAAI,IAAI,KAAK,EAAE,IAAI,MAAM,KAAK,EAAE,EAAE,CAAC;YACjC,SAAS;QACX,CAAC;QACD,MAAM,QAAQ,GAAG,IAAI,CAAC,SAAS,CAAC,KAAK,CAAC,CAAC;QACvC,MAAM,MAAM,GAAG,QAAQ,CAAC,GAAG,CAAC,QAAQ,CAAC,IAAI,EAAE,CAAC;QAC5C,MAAM,CAAC,IAAI,CAAC;YACV,UAAU;YACV,KAAK,EAAE,MAAM,CAAC,KAAK,CAAC,IAAI,CAAC;YACzB,QAAQ,EAAE,QAAQ,CAAC,MAAM,CAAC;YAC1B,GAAG,CAAC,OAAO,KAAK,EAAE,IAAI,EAAE,OAAO,EAAE,QAAQ,CAAC,OAAO,CAAC,EAAE,CAAC;SACtD,CAAC,CAAC;QACH,QAAQ,CAAC,GAAG,CAAC,QAAQ,EAAE,MAAM,CAAC,CAAC;IACjC,CAAC;IACD,KAAK,MAAM,MAAM,IAAI,QAAQ,CAAC,MAAM,EAAE,EAAE,CAAC;QACvC,MAAM,CAAC,IAAI,CAAC,CAAC,IAAI,EAAE,KAAK,EAAE,EAAE,CAAC,IAAI,CAAC,KAAK,GAAG,KAAK,CAAC,KAAK,CAAC,CAAC;IACzD,CAAC;IACD,OAAO,QAAQ,CAAC;AAClB,CAAC;AAED;;;;;;;;;;;;;GAaG;AACH,MAAM,UAAU,cAAc,CAC5B,KAAiD,EACjD,MAAkD;IAElD,MAAM,MAAM,GAAG,IAAI,GAAG,CAAC,MAAM,CAAC,CAAC;IAC/B,KAAK,MAAM,CAAC,IAAI,EAAE,QAAQ,CAAC,IAAI,KAAK,EAAE,CAAC;QACrC,MAAM,KAAK,GAAG,iBAAiB,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC,GAAG,CAAC,IAAI,CAAC,IAAI,EAAE,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC;QACtE,MAAM,CAAC,GAAG,CAAC,IAAI,EAAE,CAAC,GAAG,QAAQ,EAAE,GAAG,KAAK,CAAC,CAAC,CAAC;IAC5C,CAAC;IACD,OAAO,MAAM,CAAC;AAChB,CAAC"}

package/dist/sources/xlsx.d.ts

@@ -0,0 +1,20 @@
+/**
+ * Reading cells out of a spreadsheet's XML, as an `.xlsx` workbook stores
+ * them: a worksheet of cells, and a table of the strings they share.
+ *
+ * Only as much of the format as 教育部's downloads use. Formulas, styles and
+ * the other sheets of a workbook are never read.
+ */
+/**
+ * The header row of a worksheet.
+ */
+export declare const HEADER_ROW: RegExp;
+/**
+ * Read the workbook's shared string table.
+ */
+export declare function readSharedStrings(xml: string): readonly string[];
+/**
+ * Read a worksheet's cells into rows, keeping only the columns asked for.
+ */
+export declare function readRows(sheetXml: string, strings: readonly string[], wanted?: ReadonlySet<string>): Map<number, Map<string, string>>;
+//# sourceMappingURL=xlsx.d.ts.map

package/dist/sources/xlsx.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"xlsx.d.ts","sourceRoot":"","sources":["../../src/sources/xlsx.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAEH;;GAEG;AACH,eAAO,MAAM,UAAU,QAA+B,CAAC;AAsCvD;;GAEG;AACH,wBAAgB,iBAAiB,CAAC,GAAG,EAAE,MAAM,GAAG,SAAS,MAAM,EAAE,CAIhE;AAED;;GAEG;AACH,wBAAgB,QAAQ,CACtB,QAAQ,EAAE,MAAM,EAChB,OAAO,EAAE,SAAS,MAAM,EAAE,EAC1B,MAAM,CAAC,EAAE,WAAW,CAAC,MAAM,CAAC,GAC3B,GAAG,CAAC,MAAM,EAAE,GAAG,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC,CA0BlC"}