@kensio/pinyinjs 1.8.6 → 1.9.0

package/data/standard.keys

@@ -67108,6 +67108,7 @@ B
 網關
 網頁
 網點
+綳
 綴
 綵
 綷
@@ -96968,7 +96969,6 @@ B
 麩
 麫
 麮
-麯
 麰
 麳
 麴

package/dist/dictionary/character-defaults.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"character-defaults.d.ts","sourceRoot":"","sources":["../../src/dictionary/character-defaults.ts"],"names":[],"mappings":"AAcA,OAAO,KAAK,EAAE,WAAW,EAAE,MAAM,sBAAsB,CAAC;AACxD,OAAO,KAAK,EAAE,QAAQ,EAAE,MAAM,yBAAyB,CAAC;AAExD,OAAO,EAGL,gBAAgB,EACjB,MAAM,kBAAkB,CAAC;AAC1B,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,kBAAkB,CAAC;AAErD;;;;;;GAMG;AACH,wBAAgB,iBAAiB,CAC/B,SAAS,EAAE,MAAM,EACjB,OAAO,EAAE,MAAM,GACd,QAAQ,GAAG,SAAS,CAEtB;AAED;;;GAGG;AACH,MAAM,WAAW,iBAAiB;IAChC,QAAQ,CAAC,WAAW,EAAE,gBAAgB,CAAC;IACvC,QAAQ,CAAC,QAAQ,EAAE,WAAW,CAAC,MAAM,EAAE,SAAS,QAAQ,EAAE,CAAC,CAAC;IAC5D,QAAQ,CAAC,eAAe,EAAE,MAAM,CAAC;CAClC;AAED;;GAEG;AACH,wBAAgB,sBAAsB,CACpC,OAAO,EAAE,YAAY,EACrB,YAAY,EAAE,WAAW,CAAC,MAAM,EAAE,SAAS,WAAW,EAAE,CAAC,EACzD,YAAY,EAAE,WAAW,CAAC,MAAM,EAAE,SAAS,WAAW,EAAE,CAAC,GACxD,iBAAiB,CAiDnB"}
+{"version":3,"file":"character-defaults.d.ts","sourceRoot":"","sources":["../../src/dictionary/character-defaults.ts"],"names":[],"mappings":"AAeA,OAAO,KAAK,EAAE,WAAW,EAAE,MAAM,sBAAsB,CAAC;AACxD,OAAO,KAAK,EAAE,QAAQ,EAAE,MAAM,yBAAyB,CAAC;AAExD,OAAO,EAGL,gBAAgB,EACjB,MAAM,kBAAkB,CAAC;AAC1B,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,kBAAkB,CAAC;AAErD;;;;;;GAMG;AACH,wBAAgB,iBAAiB,CAC/B,SAAS,EAAE,MAAM,EACjB,OAAO,EAAE,MAAM,GACd,QAAQ,GAAG,SAAS,CAEtB;AAED;;;GAGG;AACH,MAAM,WAAW,iBAAiB;IAChC,QAAQ,CAAC,WAAW,EAAE,gBAAgB,CAAC;IACvC,QAAQ,CAAC,QAAQ,EAAE,WAAW,CAAC,MAAM,EAAE,SAAS,QAAQ,EAAE,CAAC,CAAC;IAC5D,QAAQ,CAAC,eAAe,EAAE,MAAM,CAAC;CAClC;AAED;;GAEG;AACH,wBAAgB,sBAAsB,CACpC,OAAO,EAAE,YAAY,EACrB,YAAY,EAAE,WAAW,CAAC,MAAM,EAAE,SAAS,WAAW,EAAE,CAAC,EACzD,YAAY,EAAE,WAAW,CAAC,MAAM,EAAE,SAAS,WAAW,EAAE,CAAC,GACxD,iBAAiB,CAsDnB"}

package/dist/dictionary/character-defaults.js

@@ -8,6 +8,7 @@
  */
 import { countCorpusMass, rankByCorpusMass } from "./corpus-mass.js";
 import { demoteReducedNeutrals, isNeutralAlone, resolveFrequencyTones, } from "./frequency-tones.js";
+import { leadWithFreeReading } from "./free-readings.js";
 import { readAlignedReading, readDictionaryReading } from "./reading.js";
 import { pairScripts, TraditionalTable, } from "./traditional.js";
 /**
@@ -39,6 +40,8 @@ export function buildCharacterDefaults(sources, cedictByWord, cedictByHant) {
     // Unihan ranks the readings and the corpus re-ranks the ones it has seen: a
     // character's default is whichever of its readings the dictionary's own words
     // spend the most of jieba's corpus on, and Unihan's order decides the rest.
+    // A character listed in FREE_READINGS then leads with the reading it takes
+    // standing alone, which neither ranking can see.
     const mass = countCorpusMass({ phrase, cedict, jieba }, traditional);
     const defaults = new Map();
     let reducedNeutrals = 0;
@@ -59,7 +62,7 @@ export function buildCharacterDefaults(sources, cedictByWord, cedictByHant) {
             .map((reading) => characterSyllable(character, reading))
             .filter((syllable) => syllable !== undefined);
         if (parsed.length > 0) {
-            defaults.set(character, rankByCorpusMass(character, parsed, readings.fields, mass));
+            defaults.set(character, leadWithFreeReading(character, rankByCorpusMass(character, parsed, readings.fields, mass)));
         }
     }
     return { traditional, defaults, reducedNeutrals };

package/dist/dictionary/character-defaults.js.map

@@ -1 +1 @@
-{"version":3,"file":"character-defaults.js","sourceRoot":"","sources":["../../src/dictionary/character-defaults.ts"],"names":[],"mappings":"AAAA;;;;;;;GAOG;AACH,OAAO,EAAE,eAAe,EAAE,gBAAgB,EAAE,MAAM,kBAAkB,CAAC;AACrE,OAAO,EACL,qBAAqB,EACrB,cAAc,EACd,qBAAqB,GACtB,MAAM,sBAAsB,CAAC;AAG9B,OAAO,EAAE,kBAAkB,EAAE,qBAAqB,EAAE,MAAM,cAAc,CAAC;AACzE,OAAO,EACL,WAAW,EAEX,gBAAgB,GACjB,MAAM,kBAAkB,CAAC;AAG1B;;;;;;GAMG;AACH,MAAM,UAAU,iBAAiB,CAC/B,SAAiB,EACjB,OAAe;IAEf,OAAO,qBAAqB,CAAC,SAAS,EAAE,CAAC,OAAO,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC;AAC1D,CAAC;AAYD;;GAEG;AACH,MAAM,UAAU,sBAAsB,CACpC,OAAqB,EACrB,YAAyD,EACzD,YAAyD;IAEzD,MAAM,EAAE,cAAc,EAAE,cAAc,EAAE,MAAM,EAAE,MAAM,EAAE,KAAK,EAAE,GAAG,OAAO,CAAC;IAC1E,6DAA6D;IAC7D,6EAA6E;IAC7E,wEAAwE;IACxE,uEAAuE;IACvE,MAAM,QAAQ,GAAoB,EAAE,CAAC;IACrC,KAAK,MAAM,KAAK,IAAI,MAAM,EAAE,CAAC;QAC3B,MAAM,OAAO,GAAG,kBAAkB,CAAC,KAAK,CAAC,UAAU,EAAE,KAAK,CAAC,QAAQ,CAAC,CAAC;QACrE,QAAQ,CAAC,IAAI,CAAC,GAAG,WAAW,CAAC,KAAK,CAAC,UAAU,EAAE,KAAK,CAAC,WAAW,EAAE,OAAO,CAAC,CAAC,CAAC;IAC9E,CAAC;IACD,MAAM,WAAW,GAAG,gBAAgB,CAAC,KAAK,CACxC,QAAQ,EACR,cAAc,EACd,cAAc,CACf,CAAC;IAEF,8DAA8D;IAC9D,4EAA4E;IAC5E,8EAA8E;IAC9E,4EAA4E;IAC5E,MAAM,IAAI,GAAG,eAAe,CAAC,EAAE,MAAM,EAAE,MAAM,EAAE,KAAK,EAAE,EAAE,WAAW,CAAC,CAAC;IACrE,MAAM,QAAQ,GAAG,IAAI,GAAG,EAA+B,CAAC;IACxD,IAAI,eAAe,GAAG,CAAC,CAAC;IACxB,KAAK,MAAM,CAAC,SAAS,EAAE,QAAQ,CAAC,IAAI,cAAc,EAAE,CAAC;QACnD,MAAM,QAAQ,GAAG,qBAAqB,CAAC,QAAQ,CAAC,CAAC;QACjD,qEAAqE;QACrE,4EAA4E;QAC5E,MAAM,MAAM,GAAG,cAAc,CAAC,SAAS,EAAE;YACvC,GAAG,CAAC,YAAY,CAAC,GAAG,CAAC,SAAS,CAAC,IAAI,EAAE,CAAC;YACtC,GAAG,CAAC,YAAY,CAAC,GAAG,CAAC,SAAS,CAAC,IAAI,EAAE,CAAC;SACvC,CAAC;YACA,CAAC,CAAC,QAAQ;YACV,CAAC,CAAC,qBAAqB,CAAC,QAAQ,EAAE,QAAQ,CAAC,CAAC;QAC9C,IAAI,MAAM,KAAK,QAAQ,IAAI,MAAM,CAAC,CAAC,CAAC,KAAK,QAAQ,CAAC,CAAC,CAAC,EAAE,CAAC;YACrD,eAAe,EAAE,CAAC;QACpB,CAAC;QACD,MAAM,MAAM,GAAG,MAAM;aAClB,GAAG,CAAC,CAAC,OAAO,EAAE,EAAE,CAAC,iBAAiB,CAAC,SAAS,EAAE,OAAO,CAAC,CAAC;aACvD,MAAM,CAAC,CAAC,QAAQ,EAAE,EAAE,CAAC,QAAQ,KAAK,SAAS,CAAC,CAAC;QAChD,IAAI,MAAM,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YACtB,QAAQ,CAAC,GAAG,CACV,SAAS,EACT,gBAAgB,CAAC,SAAS,EAAE,MAAM,EAAE,QAAQ,CAAC,MAAM,EAAE,IAAI,CAAC,CAC3D,CAAC;QACJ,CAAC;IACH,CAAC;IAED,OAAO,EAAE,WAAW,EAAE,QAAQ,EAAE,eAAe,EAAE,CAAC;AACpD,CAAC"}
+{"version":3,"file":"character-defaults.js","sourceRoot":"","sources":["../../src/dictionary/character-defaults.ts"],"names":[],"mappings":"AAAA;;;;;;;GAOG;AACH,OAAO,EAAE,eAAe,EAAE,gBAAgB,EAAE,MAAM,kBAAkB,CAAC;AACrE,OAAO,EACL,qBAAqB,EACrB,cAAc,EACd,qBAAqB,GACtB,MAAM,sBAAsB,CAAC;AAC9B,OAAO,EAAE,mBAAmB,EAAE,MAAM,oBAAoB,CAAC;AAGzD,OAAO,EAAE,kBAAkB,EAAE,qBAAqB,EAAE,MAAM,cAAc,CAAC;AACzE,OAAO,EACL,WAAW,EAEX,gBAAgB,GACjB,MAAM,kBAAkB,CAAC;AAG1B;;;;;;GAMG;AACH,MAAM,UAAU,iBAAiB,CAC/B,SAAiB,EACjB,OAAe;IAEf,OAAO,qBAAqB,CAAC,SAAS,EAAE,CAAC,OAAO,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC;AAC1D,CAAC;AAYD;;GAEG;AACH,MAAM,UAAU,sBAAsB,CACpC,OAAqB,EACrB,YAAyD,EACzD,YAAyD;IAEzD,MAAM,EAAE,cAAc,EAAE,cAAc,EAAE,MAAM,EAAE,MAAM,EAAE,KAAK,EAAE,GAAG,OAAO,CAAC;IAC1E,6DAA6D;IAC7D,6EAA6E;IAC7E,wEAAwE;IACxE,uEAAuE;IACvE,MAAM,QAAQ,GAAoB,EAAE,CAAC;IACrC,KAAK,MAAM,KAAK,IAAI,MAAM,EAAE,CAAC;QAC3B,MAAM,OAAO,GAAG,kBAAkB,CAAC,KAAK,CAAC,UAAU,EAAE,KAAK,CAAC,QAAQ,CAAC,CAAC;QACrE,QAAQ,CAAC,IAAI,CAAC,GAAG,WAAW,CAAC,KAAK,CAAC,UAAU,EAAE,KAAK,CAAC,WAAW,EAAE,OAAO,CAAC,CAAC,CAAC;IAC9E,CAAC;IACD,MAAM,WAAW,GAAG,gBAAgB,CAAC,KAAK,CACxC,QAAQ,EACR,cAAc,EACd,cAAc,CACf,CAAC;IAEF,8DAA8D;IAC9D,4EAA4E;IAC5E,8EAA8E;IAC9E,4EAA4E;IAC5E,2EAA2E;IAC3E,iDAAiD;IACjD,MAAM,IAAI,GAAG,eAAe,CAAC,EAAE,MAAM,EAAE,MAAM,EAAE,KAAK,EAAE,EAAE,WAAW,CAAC,CAAC;IACrE,MAAM,QAAQ,GAAG,IAAI,GAAG,EAA+B,CAAC;IACxD,IAAI,eAAe,GAAG,CAAC,CAAC;IACxB,KAAK,MAAM,CAAC,SAAS,EAAE,QAAQ,CAAC,IAAI,cAAc,EAAE,CAAC;QACnD,MAAM,QAAQ,GAAG,qBAAqB,CAAC,QAAQ,CAAC,CAAC;QACjD,qEAAqE;QACrE,4EAA4E;QAC5E,MAAM,MAAM,GAAG,cAAc,CAAC,SAAS,EAAE;YACvC,GAAG,CAAC,YAAY,CAAC,GAAG,CAAC,SAAS,CAAC,IAAI,EAAE,CAAC;YACtC,GAAG,CAAC,YAAY,CAAC,GAAG,CAAC,SAAS,CAAC,IAAI,EAAE,CAAC;SACvC,CAAC;YACA,CAAC,CAAC,QAAQ;YACV,CAAC,CAAC,qBAAqB,CAAC,QAAQ,EAAE,QAAQ,CAAC,CAAC;QAC9C,IAAI,MAAM,KAAK,QAAQ,IAAI,MAAM,CAAC,CAAC,CAAC,KAAK,QAAQ,CAAC,CAAC,CAAC,EAAE,CAAC;YACrD,eAAe,EAAE,CAAC;QACpB,CAAC;QACD,MAAM,MAAM,GAAG,MAAM;aAClB,GAAG,CAAC,CAAC,OAAO,EAAE,EAAE,CAAC,iBAAiB,CAAC,SAAS,EAAE,OAAO,CAAC,CAAC;aACvD,MAAM,CAAC,CAAC,QAAQ,EAAE,EAAE,CAAC,QAAQ,KAAK,SAAS,CAAC,CAAC;QAChD,IAAI,MAAM,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YACtB,QAAQ,CAAC,GAAG,CACV,SAAS,EACT,mBAAmB,CACjB,SAAS,EACT,gBAAgB,CAAC,SAAS,EAAE,MAAM,EAAE,QAAQ,CAAC,MAAM,EAAE,IAAI,CAAC,CAC3D,CACF,CAAC;QACJ,CAAC;IACH,CAAC;IAED,OAAO,EAAE,WAAW,EAAE,QAAQ,EAAE,eAAe,EAAE,CAAC;AACpD,CAAC"}

package/dist/dictionary/character-readings.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"character-readings.d.ts","sourceRoot":"","sources":["../../src/dictionary/character-readings.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AACH,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,uBAAuB,CAAC;AAG5D,eAAO,MAAM,kBAAkB,EAAE,SAAS,cAAc,EAsBvD,CAAC"}
+{"version":3,"file":"character-readings.d.ts","sourceRoot":"","sources":["../../src/dictionary/character-readings.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AACH,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,uBAAuB,CAAC;AAG5D,eAAO,MAAM,kBAAkB,EAAE,SAAS,cAAc,EA+BvD,CAAC"}

package/dist/dictionary/character-readings.js

@@ -21,5 +21,14 @@ export const CHARACTER_READINGS = [
     reads("子", "zǐ", "and a suffix sense is not a claim about the character"),
     reads("夫", "fū", "nor is 大夫's and 丈夫's reduction"),
     reads("吗", "ma", "but a 语气词 CC-CEDICT reads 轻声 alone keeps it"),
+    // FREE_READINGS: the free verb or classifier leads, and the compounds that
+    // ranked the other reading first keep it.
+    reads("剥", "bāo", "the free verb, not 剥削's bō"),
+    reads("剥削", "bō xuē", "and the compounds keep the literary reading"),
+    reads("切", "qiē", "kHanyuPinlu's qiè is nearly all 一切"),
+    reads("一切", "yī qiè", "which keeps it"),
+    reads("幢", "zhuàng", "the classifier, not 经幢's banner"),
+    reads("脏", "zāng", "dirty, not the organ"),
+    reads("心脏", "xīn zàng", "which 心脏 keeps"),
 ];
 //# sourceMappingURL=character-readings.js.map

package/dist/dictionary/character-readings.js.map

@@ -1 +1 @@
-{"version":3,"file":"character-readings.js","sourceRoot":"","sources":["../../src/dictionary/character-readings.ts"],"names":[],"mappings":"AAQA,OAAO,EAAE,KAAK,EAAE,MAAM,oBAAoB,CAAC;AAE3C,MAAM,CAAC,MAAM,kBAAkB,GAA8B;IAC3D,oEAAoE;IACpE,wEAAwE;IACxE,0EAA0E;IAC1E,yEAAyE;IACzE,gEAAgE;IAChE,KAAK,CAAC,IAAI,EAAE,UAAU,EAAE,uCAAuC,CAAC;IAChE,KAAK,CAAC,IAAI,EAAE,SAAS,EAAE,6CAA6C,CAAC;IACrE,KAAK,CAAC,IAAI,EAAE,QAAQ,EAAE,gDAAgD,CAAC;IACvE,0EAA0E;IAC1E,wEAAwE;IACxE,0EAA0E;IAC1E,4EAA4E;IAC5E,uDAAuD;IACvD,KAAK,CAAC,GAAG,EAAE,IAAI,EAAE,yCAAyC,CAAC;IAC3D,KAAK,CAAC,GAAG,EAAE,KAAK,EAAE,+CAA+C,CAAC;IAClE,KAAK,CAAC,GAAG,EAAE,IAAI,EAAE,4CAA4C,CAAC;IAC9D,KAAK,CAAC,IAAI,EAAE,QAAQ,EAAE,mDAAmD,CAAC;IAC1E,KAAK,CAAC,GAAG,EAAE,IAAI,EAAE,8CAA8C,CAAC;IAChE,KAAK,CAAC,GAAG,EAAE,IAAI,EAAE,uDAAuD,CAAC;IACzE,KAAK,CAAC,GAAG,EAAE,IAAI,EAAE,gCAAgC,CAAC;IAClD,KAAK,CAAC,GAAG,EAAE,IAAI,EAAE,6CAA6C,CAAC;CAChE,CAAC"}
+{"version":3,"file":"character-readings.js","sourceRoot":"","sources":["../../src/dictionary/character-readings.ts"],"names":[],"mappings":"AAQA,OAAO,EAAE,KAAK,EAAE,MAAM,oBAAoB,CAAC;AAE3C,MAAM,CAAC,MAAM,kBAAkB,GAA8B;IAC3D,oEAAoE;IACpE,wEAAwE;IACxE,0EAA0E;IAC1E,yEAAyE;IACzE,gEAAgE;IAChE,KAAK,CAAC,IAAI,EAAE,UAAU,EAAE,uCAAuC,CAAC;IAChE,KAAK,CAAC,IAAI,EAAE,SAAS,EAAE,6CAA6C,CAAC;IACrE,KAAK,CAAC,IAAI,EAAE,QAAQ,EAAE,gDAAgD,CAAC;IACvE,0EAA0E;IAC1E,wEAAwE;IACxE,0EAA0E;IAC1E,4EAA4E;IAC5E,uDAAuD;IACvD,KAAK,CAAC,GAAG,EAAE,IAAI,EAAE,yCAAyC,CAAC;IAC3D,KAAK,CAAC,GAAG,EAAE,KAAK,EAAE,+CAA+C,CAAC;IAClE,KAAK,CAAC,GAAG,EAAE,IAAI,EAAE,4CAA4C,CAAC;IAC9D,KAAK,CAAC,IAAI,EAAE,QAAQ,EAAE,mDAAmD,CAAC;IAC1E,KAAK,CAAC,GAAG,EAAE,IAAI,EAAE,8CAA8C,CAAC;IAChE,KAAK,CAAC,GAAG,EAAE,IAAI,EAAE,uDAAuD,CAAC;IACzE,KAAK,CAAC,GAAG,EAAE,IAAI,EAAE,gCAAgC,CAAC;IAClD,KAAK,CAAC,GAAG,EAAE,IAAI,EAAE,6CAA6C,CAAC;IAC/D,2EAA2E;IAC3E,0CAA0C;IAC1C,KAAK,CAAC,GAAG,EAAE,KAAK,EAAE,4BAA4B,CAAC;IAC/C,KAAK,CAAC,IAAI,EAAE,QAAQ,EAAE,6CAA6C,CAAC;IACpE,KAAK,CAAC,GAAG,EAAE,KAAK,EAAE,oCAAoC,CAAC;IACvD,KAAK,CAAC,IAAI,EAAE,QAAQ,EAAE,gBAAgB,CAAC;IACvC,KAAK,CAAC,GAAG,EAAE,QAAQ,EAAE,iCAAiC,CAAC;IACvD,KAAK,CAAC,GAAG,EAAE,MAAM,EAAE,sBAAsB,CAAC;IAC1C,KAAK,CAAC,IAAI,EAAE,UAAU,EAAE,gBAAgB,CAAC;CAC1C,CAAC"}

package/dist/dictionary/free-readings.d.ts

@@ -0,0 +1,70 @@
+import { type Syllable } from "../syllable/syllable.js";
+/**
+ * A character whose reading as a word of its own is not the one it leads with.
+ */
+export interface FreeReading {
+    /** The character, in either script. */
+    readonly character: string;
+    /** The reading it takes standing alone, as tone-marked pinyin. */
+    readonly reading: string;
+    /** The words that carry the other reading, which is why the sources rank it
+     * first. */
+    readonly displaces: string;
+}
+/**
+ * Characters that read one way inside words and another standing alone.
+ *
+ * A character's default is ranked by `kHanyuPinlu` where it lists several
+ * readings and by the corpus mass of the dictionary's words where it does not,
+ * and both count the character *inside words*. That is the right evidence for
+ * most polyphones and the wrong evidence for these, which have a literary
+ * reading in their compounds and an everyday one as a free verb or classifier.
+ * 剥削, 剥夺 and 剥落 give 剥 `bō`, and a reader peeling an orange says `bāo`.
+ * `kHanyuPinlu` counts 切 `qiè` 998 times against `qiē` 70, nearly all of it
+ * 一切, and 切土豆 is `qiē`.
+ *
+ * The compounds are dictionary words and keep their own readings. The default
+ * is what a character segmented on its own is read with, which is where these
+ * stand as free words, and what an unknown compound is derived from.
+ *
+ * **What is listed here.** A character whose 现代汉语词典 entry gives the listed
+ * reading to its free use and the other to its compounds or a literary sense,
+ * found by reading example sentences rather than by a measure. CPP labels
+ * news text, where these characters mostly sit inside words, so it is the check
+ * that nothing broke rather than the evidence for the list. What is *not*
+ * listed is a character whose free uses split, such as 卡 (a card, `kǎ`, and
+ * stuck, `qiǎ`) or 背 (the back and to recite, `bèi`, and to carry, `bēi`): a
+ * default cannot choose between two free words, and a context rule can. Nor
+ * is a character that names and transliterations use on its own: 勒 is `lēi`
+ * in 勒得难受 and `lè` in every 希特勒 and 达勒姆 the dictionary lacks, and 盛 is
+ * `chéng` in 盛汤 and the surname `Shèng` in 刘盛. Listing either cost CPP 25
+ * and 7 cases.
+ *
+ * **The reading must be one the sources already give the character.** It is
+ * moved to the front of the ranking rather than written over it, so a source
+ * refresh that dropped it would fail the build rather than leave a reading
+ * nothing attests. That is why 剝 and 掙 are missing beside 剥 and 挣: Unihan
+ * gives them `bō` and `zhēng` alone.
+ */
+export declare const FREE_READINGS: readonly FreeReading[];
+/**
+ * Read a free reading into a syllable, throwing if it is not one.
+ *
+ * Throwing for the reason {@link import("./overrides.js").readOverrideReading}
+ * does: an entry that failed to parse would leave the ranking it was written
+ * to correct, and nothing would notice.
+ */
+export declare function freeSyllable(free: FreeReading): Syllable;
+/**
+ * The table, keyed by character and parsed into syllables.
+ */
+export declare const FREE_SYLLABLES: ReadonlyMap<string, Syllable>;
+/**
+ * Lead a character's ranked readings with its free reading, where it has one.
+ *
+ * The rest keep the order the ranking gave them. A listed reading the ranking
+ * does not hold throws, which is the build failing on a source that stopped
+ * attesting it.
+ */
+export declare function leadWithFreeReading(character: string, ranked: readonly Syllable[]): readonly Syllable[];
+//# sourceMappingURL=free-readings.d.ts.map

package/dist/dictionary/free-readings.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"free-readings.d.ts","sourceRoot":"","sources":["../../src/dictionary/free-readings.ts"],"names":[],"mappings":"AAAA,OAAO,EAAgB,KAAK,QAAQ,EAAE,MAAM,yBAAyB,CAAC;AAEtE;;GAEG;AACH,MAAM,WAAW,WAAW;IAC1B,uCAAuC;IACvC,QAAQ,CAAC,SAAS,EAAE,MAAM,CAAC;IAC3B,kEAAkE;IAClE,QAAQ,CAAC,OAAO,EAAE,MAAM,CAAC;IACzB;gBACY;IACZ,QAAQ,CAAC,SAAS,EAAE,MAAM,CAAC;CAC5B;AAED;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAkCG;AACH,eAAO,MAAM,aAAa,EAAE,SAAS,WAAW,EA0B/C,CAAC;AAEF;;;;;;GAMG;AACH,wBAAgB,YAAY,CAAC,IAAI,EAAE,WAAW,GAAG,QAAQ,CAQxD;AAED;;GAEG;AACH,eAAO,MAAM,cAAc,EAAE,WAAW,CAAC,MAAM,EAAE,QAAQ,CAExD,CAAC;AAaF;;;;;;GAMG;AACH,wBAAgB,mBAAmB,CACjC,SAAS,EAAE,MAAM,EACjB,MAAM,EAAE,SAAS,QAAQ,EAAE,GAC1B,SAAS,QAAQ,EAAE,CAYrB"}

package/dist/dictionary/free-readings.js

@@ -0,0 +1,108 @@
+import { readSyllable } from "../syllable/syllable.js";
+/**
+ * Characters that read one way inside words and another standing alone.
+ *
+ * A character's default is ranked by `kHanyuPinlu` where it lists several
+ * readings and by the corpus mass of the dictionary's words where it does not,
+ * and both count the character *inside words*. That is the right evidence for
+ * most polyphones and the wrong evidence for these, which have a literary
+ * reading in their compounds and an everyday one as a free verb or classifier.
+ * 剥削, 剥夺 and 剥落 give 剥 `bō`, and a reader peeling an orange says `bāo`.
+ * `kHanyuPinlu` counts 切 `qiè` 998 times against `qiē` 70, nearly all of it
+ * 一切, and 切土豆 is `qiē`.
+ *
+ * The compounds are dictionary words and keep their own readings. The default
+ * is what a character segmented on its own is read with, which is where these
+ * stand as free words, and what an unknown compound is derived from.
+ *
+ * **What is listed here.** A character whose 现代汉语词典 entry gives the listed
+ * reading to its free use and the other to its compounds or a literary sense,
+ * found by reading example sentences rather than by a measure. CPP labels
+ * news text, where these characters mostly sit inside words, so it is the check
+ * that nothing broke rather than the evidence for the list. What is *not*
+ * listed is a character whose free uses split, such as 卡 (a card, `kǎ`, and
+ * stuck, `qiǎ`) or 背 (the back and to recite, `bèi`, and to carry, `bēi`): a
+ * default cannot choose between two free words, and a context rule can. Nor
+ * is a character that names and transliterations use on its own: 勒 is `lēi`
+ * in 勒得难受 and `lè` in every 希特勒 and 达勒姆 the dictionary lacks, and 盛 is
+ * `chéng` in 盛汤 and the surname `Shèng` in 刘盛. Listing either cost CPP 25
+ * and 7 cases.
+ *
+ * **The reading must be one the sources already give the character.** It is
+ * moved to the front of the ranking rather than written over it, so a source
+ * refresh that dropped it would fail the build rather than leave a reading
+ * nothing attests. That is why 剝 and 掙 are missing beside 剥 and 挣: Unihan
+ * gives them `bō` and `zhēng` alone.
+ */
+export const FREE_READINGS = [
+    { character: "幢", reading: "zhuàng", displaces: "幢幢 and 经幢, a banner" },
+    { character: "抡", reading: "lūn", displaces: "抡材 and 抡元, to select" },
+    { character: "掄", reading: "lūn", displaces: "掄材 and 掄元, to select" },
+    { character: "剥", reading: "bāo", displaces: "剥削, 剥夺 and 剥落" },
+    { character: "削", reading: "xiāo", displaces: "削弱, 削减 and 剥削" },
+    { character: "铺", reading: "pū", displaces: "店铺 and 床铺, the noun" },
+    { character: "鋪", reading: "pū", displaces: "店鋪 and 床鋪, the noun" },
+    { character: "咳", reading: "ké", displaces: "the interjection hāi" },
+    { character: "哄", reading: "hǒng", displaces: "哄堂 and 哄抢, an uproar" },
+    { character: "刨", reading: "páo", displaces: "刨床 and 刨子, a plane" },
+    { character: "攒", reading: "zǎn", displaces: "攒聚 and 攒眉, to gather" },
+    { character: "攢", reading: "zǎn", displaces: "攢聚 and 攢眉, to gather" },
+    { character: "绷", reading: "bēng", displaces: "绷着脸, a stiff face" },
+    { character: "繃", reading: "bēng", displaces: "繃著臉, a stiff face" },
+    { character: "翘", reading: "qiào", displaces: "翘首 and 翘楚" },
+    { character: "翹", reading: "qiào", displaces: "翹首 and 翹楚" },
+    { character: "划", reading: "huá", displaces: "计划, 划分 and 规划" },
+    { character: "劃", reading: "huá", displaces: "計劃, 劃分 and 規劃" },
+    { character: "曲", reading: "qǔ", displaces: "曲线, 弯曲 and 曲折" },
+    { character: "切", reading: "qiē", displaces: "一切, 亲切 and 切实" },
+    { character: "嚼", reading: "jiáo", displaces: "咀嚼, the literary word" },
+    { character: "倔", reading: "juè", displaces: "倔强 and 倔犟" },
+    { character: "脏", reading: "zāng", displaces: "心脏 and 内脏, an organ" },
+    { character: "挣", reading: "zhèng", displaces: "挣扎, to struggle" },
+    { character: "漂", reading: "piāo", displaces: "漂亮 and 漂白" },
+];
+/**
+ * Read a free reading into a syllable, throwing if it is not one.
+ *
+ * Throwing for the reason {@link import("./overrides.js").readOverrideReading}
+ * does: an entry that failed to parse would leave the ranking it was written
+ * to correct, and nothing would notice.
+ */
+export function freeSyllable(free) {
+    const syllable = readSyllable(free.reading);
+    if (syllable?.tone === undefined) {
+        throw new Error(`free reading for ${free.character} is not a reading: ${free.reading}`);
+    }
+    return syllable;
+}
+/**
+ * The table, keyed by character and parsed into syllables.
+ */
+export const FREE_SYLLABLES = new Map(FREE_READINGS.map((free) => [free.character, freeSyllable(free)]));
+/**
+ * Whether two syllables are the same reading, 儿化 aside.
+ */
+function sameReading(left, right) {
+    return (left.initial === right.initial &&
+        left.final === right.final &&
+        left.tone === right.tone);
+}
+/**
+ * Lead a character's ranked readings with its free reading, where it has one.
+ *
+ * The rest keep the order the ranking gave them. A listed reading the ranking
+ * does not hold throws, which is the build failing on a source that stopped
+ * attesting it.
+ */
+export function leadWithFreeReading(character, ranked) {
+    const free = FREE_SYLLABLES.get(character);
+    if (free === undefined) {
+        return ranked;
+    }
+    const lead = ranked.find((syllable) => sameReading(syllable, free));
+    if (lead === undefined) {
+        throw new Error(`free reading for ${character} is not one the sources give it`);
+    }
+    return [lead, ...ranked.filter((syllable) => syllable !== lead)];
+}
+//# sourceMappingURL=free-readings.js.map

package/dist/dictionary/free-readings.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"free-readings.js","sourceRoot":"","sources":["../../src/dictionary/free-readings.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,YAAY,EAAiB,MAAM,yBAAyB,CAAC;AAetE;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAkCG;AACH,MAAM,CAAC,MAAM,aAAa,GAA2B;IACnD,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,QAAQ,EAAE,SAAS,EAAE,qBAAqB,EAAE;IACvE,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,sBAAsB,EAAE;IACrE,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,sBAAsB,EAAE;IACrE,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,eAAe,EAAE;IAC9D,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,MAAM,EAAE,SAAS,EAAE,eAAe,EAAE;IAC/D,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,IAAI,EAAE,SAAS,EAAE,qBAAqB,EAAE;IACnE,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,IAAI,EAAE,SAAS,EAAE,qBAAqB,EAAE;IACnE,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,IAAI,EAAE,SAAS,EAAE,sBAAsB,EAAE;IACpE,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,MAAM,EAAE,SAAS,EAAE,sBAAsB,EAAE;IACtE,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,oBAAoB,EAAE;IACnE,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,sBAAsB,EAAE;IACrE,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,sBAAsB,EAAE;IACrE,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,MAAM,EAAE,SAAS,EAAE,mBAAmB,EAAE;IACnE,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,MAAM,EAAE,SAAS,EAAE,mBAAmB,EAAE;IACnE,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,MAAM,EAAE,SAAS,EAAE,WAAW,EAAE;IAC3D,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,MAAM,EAAE,SAAS,EAAE,WAAW,EAAE;IAC3D,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,eAAe,EAAE;IAC9D,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,eAAe,EAAE;IAC9D,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,IAAI,EAAE,SAAS,EAAE,eAAe,EAAE;IAC7D,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,eAAe,EAAE;IAC9D,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,MAAM,EAAE,SAAS,EAAE,uBAAuB,EAAE;IACvE,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,WAAW,EAAE;IAC1D,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,MAAM,EAAE,SAAS,EAAE,qBAAqB,EAAE;IACrE,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,OAAO,EAAE,SAAS,EAAE,iBAAiB,EAAE;IAClE,EAAE,SAAS,EAAE,GAAG,EAAE,OAAO,EAAE,MAAM,EAAE,SAAS,EAAE,WAAW,EAAE;CAC5D,CAAC;AAEF;;;;;;GAMG;AACH,MAAM,UAAU,YAAY,CAAC,IAAiB;IAC5C,MAAM,QAAQ,GAAG,YAAY,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC;IAC5C,IAAI,QAAQ,EAAE,IAAI,KAAK,SAAS,EAAE,CAAC;QACjC,MAAM,IAAI,KAAK,CACb,oBAAoB,IAAI,CAAC,SAAS,sBAAsB,IAAI,CAAC,OAAO,EAAE,CACvE,CAAC;IACJ,CAAC;IACD,OAAO,QAAQ,CAAC;AAClB,CAAC;AAED;;GAEG;AACH,MAAM,CAAC,MAAM,cAAc,GAAkC,IAAI,GAAG,CAClE,aAAa,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,CAAC,IAAI,CAAC,SAAS,EAAE,YAAY,CAAC,IAAI,CAAC,CAAC,CAAC,CAClE,CAAC;AAEF;;GAEG;AACH,SAAS,WAAW,CAAC,IAAc,EAAE,KAAe;IAClD,OAAO,CACL,IAAI,CAAC,OAAO,KAAK,KAAK,CAAC,OAAO;QAC9B,IAAI,CAAC,KAAK,KAAK,KAAK,CAAC,KAAK;QAC1B,IAAI,CAAC,IAAI,KAAK,KAAK,CAAC,IAAI,CACzB,CAAC;AACJ,CAAC;AAED;;;;;;GAMG;AACH,MAAM,UAAU,mBAAmB,CACjC,SAAiB,EACjB,MAA2B;IAE3B,MAAM,IAAI,GAAG,cAAc,CAAC,GAAG,CAAC,SAAS,CAAC,CAAC;IAC3C,IAAI,IAAI,KAAK,SAAS,EAAE,CAAC;QACvB,OAAO,MAAM,CAAC;IAChB,CAAC;IACD,MAAM,IAAI,GAAG,MAAM,CAAC,IAAI,CAAC,CAAC,QAAQ,EAAE,EAAE,CAAC,WAAW,CAAC,QAAQ,EAAE,IAAI,CAAC,CAAC,CAAC;IACpE,IAAI,IAAI,KAAK,SAAS,EAAE,CAAC;QACvB,MAAM,IAAI,KAAK,CACb,oBAAoB,SAAS,iCAAiC,CAC/D,CAAC;IACJ,CAAC;IACD,OAAO,CAAC,IAAI,EAAE,GAAG,MAAM,CAAC,MAAM,CAAC,CAAC,QAAQ,EAAE,EAAE,CAAC,QAAQ,KAAK,IAAI,CAAC,CAAC,CAAC;AACnE,CAAC"}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@kensio/pinyinjs",
-  "version": "1.8.6",
+  "version": "1.9.0",
   "description": "Chinese hanzi to pinyin conversion",
   "repository": "https://github.com/KensioSoftware/pinyinjs",
   "homepage": "https://pinyinjs.dev",