refact: Class Pinyin for extend and override segment method

.umirc.js

@@ -27,7 +27,6 @@ export default defineConfig({
   hash: true,
   resolve: {
     includes: ['.'],
-    excludes: ['pinyin.ts', 'segment.ts'],
   },
   locales: [
     ['zh-CN', '中文'],

src/PinyinBase.ts

@@ -0,0 +1,275 @@
+import DICT_ZI from "./data/dict-zi"; // 单个汉字拼音数据。
+import DICT_PHRASES from "./data/phrases-dict"; // 词组拼音数据。
+import { segment } from "./segment-web";
+import { toFixed } from "./format";
+import SurnamePinyinData from "./data/surname";
+import CompoundSurnamePinyinData from "./data/compound_surname";
+import { hasKey, convertUserOptions, combo, compact } from "./util";
+import { ENUM_PINYIN_MODE, ENUM_PINYIN_STYLE } from "./constant";
+import type {
+  IPinyinAllOptions,
+  IPinyinOptions,
+  IPinyinSegment,
+} from "./declare";
+
+export default class PinyinBase {
+  // 兼容 v2.x 中的属性透出　
+  // pinyin styles:
+  STYLE_TONE = ENUM_PINYIN_STYLE.TONE;
+  STYLE_TONE2 = ENUM_PINYIN_STYLE.TONE2;
+  STYLE_TO3NE = ENUM_PINYIN_STYLE.TO3NE;
+  STYLE_NORMAL = ENUM_PINYIN_STYLE.NORMAL;
+  STYLE_INITIALS = ENUM_PINYIN_STYLE.INITIALS;
+  STYLE_FIRST_LETTER = ENUM_PINYIN_STYLE.FIRST_LETTER;
+
+  // 兼容 v2.x 中的属性透出　
+  // pinyin mode:
+  MODE_NORMAL = ENUM_PINYIN_MODE.NORMAL;
+  MODE_SURNAME = ENUM_PINYIN_MODE.SURNAME;
+  // MODE_PLACENAME = ENUM_PINYIN_MODE.PLACENAME;
+
+  /**
+   * 拼音转换入口。
+   */
+  pinyin(hans: string, options?: IPinyinOptions): string[][] {
+    if(typeof hans !== "string") {
+      return [];
+    }
+    const opt: IPinyinAllOptions = convertUserOptions(options);
+
+    let pys;
+    if (opt.mode === ENUM_PINYIN_MODE.SURNAME) {
+      pys = this.surname_pinyin(hans, opt);
+    } else {
+      // 因为分词结果有词性信息，结构不同，处理也不相同，所以需要分别处理。
+      if (opt.segment) {
+        // 分词加词性标注转换。
+        pys = this.segment_pinyin(hans, opt);
+      } else {
+        // 单字拆分转换。连续的非中文字符作为一个词（原样输出，不转换成拼音）。
+        pys = this.normal_pinyin(hans, opt);
+      }
+    }
+    if (options?.compact) {
+      pys = compact(pys);
+    }
+    return pys;
+  }
+
+  /**
+   * 不使用分词算法的拼音转换。
+   */
+  normal_pinyin(hans: string, options: IPinyinAllOptions): string[][] {
+    const pys: string[][] = [];
+    let nohans = "";
+
+    for(let i = 0, l = hans.length; i < l; i++) {
+      const words = hans[i];
+      const firstCharCode = words.charCodeAt(0);
+
+      if(DICT_ZI[firstCharCode]){
+        // 处理前面的“非中文”部分。
+        if (nohans.length > 0) {
+          pys.push([nohans]);
+          nohans = ""; // 重置“非中文”缓存。
+        }
+        pys.push(this.single_pinyin(words, options));
+      } else {
+        nohans += words;
+      }
+    }
+
+    // 清理最后的非中文字符串。
+    if(nohans.length > 0){
+      pys.push([nohans]);
+      nohans = ""; // reset non-chinese words.
+    }
+    return pys;
+  }
+
+  /**
+   * 单字拼音转换。
+   * @param {String} han, 单个汉字
+   * @return {Array} 返回拼音列表，多音字会有多个拼音项。
+   */
+  single_pinyin(han: string, options: IPinyinAllOptions): string[] {
+    if (typeof han !== "string") {
+      return [];
+    }
+    if (han.length !== 1) {
+      return this.single_pinyin(han.charAt(0), options);
+    }
+
+    const hanCode = han.charCodeAt(0);
+
+    if (!DICT_ZI[hanCode]) {
+      return [han];
+    }
+
+    const pys = DICT_ZI[hanCode].split(",");
+    if(!options.heteronym){
+      return [toFixed(pys[0], options.style)];
+    }
+
+    // 临时存储已存在的拼音，避免多音字拼音转换为非注音风格出现重复。
+    const py_cached: Record<string,string> = {};
+    const pinyins = [];
+    for(let i = 0, l = pys.length; i < l; i++){
+      const py = toFixed(pys[i], options.style);
+      if(hasKey(py_cached, py)){
+        continue;
+      }
+      py_cached[py] = py;
+
+      pinyins.push(py);
+    }
+    return pinyins;
+  }
+
+  segment(hans: string, segmentType?: IPinyinSegment): string[] {
+    return segment(hans, segmentType);
+  }
+
+  /**
+   * 将文本分词，并转换成拼音。
+   */
+  segment_pinyin(hans: string, options: IPinyinAllOptions): string[][] {
+    const phrases = this.segment(hans, options.segment);
+    let pys: string[][] = [];
+    let nohans = "";
+    for (let i = 0, l = phrases.length; i < l; i++) {
+      const words = phrases[i];
+      const firstCharCode = words.charCodeAt(0);
+
+      if(DICT_ZI[firstCharCode]){
+        // ends of non-chinese words.
+        if(nohans.length > 0){
+          pys.push([nohans]);
+          nohans = ""; // reset non-chinese words.
+        }
+
+        const newPys = words.length === 1
+          ? this.normal_pinyin(words, options)
+          : this.phrases_pinyin(words, options);
+
+        if (options.group) {
+          pys.push(this.groupPhrases(newPys));
+        } else {
+          pys = pys.concat(newPys);
+        }
+
+      } else {
+        nohans += words;
+      }
+    }
+
+    // 清理最后的非中文字符串。
+    if(nohans.length > 0){
+      pys.push([nohans]);
+      nohans = ""; // reset non-chinese words.
+    }
+    return pys;
+  }
+
+  /**
+   * 词语注音
+   * @param {String} phrases, 指定的词组。
+   * @param {Object} options, 选项。
+   * @return {Array}
+   */
+  phrases_pinyin(phrases: string, options: IPinyinAllOptions): string[][] {
+    const py: string[][] = [];
+    if (hasKey(DICT_PHRASES, phrases)) {
+      //! copy pinyin result.
+      DICT_PHRASES[phrases].forEach(function(item: string[], idx: number) {
+        py[idx] = [];
+        if (options.heteronym) {
+          item.forEach(function(py_item, py_index) {
+            py[idx][py_index] = toFixed(py_item, options.style);
+          });
+        } else {
+          py[idx][0] = toFixed(item[0], options.style);
+        }
+      });
+    } else {
+      for(let i = 0, l = phrases.length; i < l; i++) {
+        py.push(this.single_pinyin(phrases[i], options));
+      }
+    }
+    return py;
+  }
+
+  groupPhrases(phrases: string[][]): string[] {
+    if (phrases.length === 1) {
+      return phrases[0];
+    }
+
+    const grouped = combo(phrases);
+
+    return grouped;
+  }
+
+  // 姓名转成拼音
+  surname_pinyin(hans: string, options: IPinyinAllOptions): string[][] {
+    return this.compound_surname(hans, options);
+  }
+
+  // 复姓处理
+  compound_surname(hans: string, options: IPinyinAllOptions): string[][] {
+    const len = hans.length;
+    let prefixIndex = 0;
+    let result: string[][] = [];
+    for (let i = 0; i < len; i++) {
+      const twowords = hans.substring(i, i + 2);
+      if (hasKey(CompoundSurnamePinyinData, twowords)) {
+        if (prefixIndex <= i - 1) {
+          result = result.concat(
+            this.single_surname(
+              hans.substring(prefixIndex, i),
+              options
+            )
+          );
+        }
+        result = result.concat(CompoundSurnamePinyinData[twowords]);
+
+        i = i + 2;
+        prefixIndex = i;
+      }
+    }
+    // 处理复姓后面剩余的部分。
+    result = result.concat(
+      this.single_surname(
+        hans.substring(prefixIndex, len),
+        options
+      )
+    );
+    return result;
+  }
+
+  // 单姓处理
+  single_surname(hans: string, options: IPinyinAllOptions) {
+    let result: string[][] = [];
+    for (let i = 0, l = hans.length; i < l; i++) {
+      const word = hans.charAt(i);
+      if (hasKey(SurnamePinyinData, word)) {
+        result = result.concat(SurnamePinyinData[word]);
+      } else {
+        result.push(this.single_pinyin(word, options));
+      }
+    }
+    return result;
+  }
+
+  /**
+   * 比较两个汉字转成拼音后的排序顺序，可以用作默认的拼音排序算法。
+   *
+   * @param {String} hanA 汉字字符串 A。
+   * @return {String} hanB 汉字字符串 B。
+   * @return {Number} 返回 -1，0，或 1。
+   */
+  compare(hanA: string, hanB: string): number {
+    const pinyinA = this.pinyin(hanA);
+    const pinyinB = this.pinyin(hanB);
+    return String(pinyinA).localeCompare(String(pinyinB));
+  }
+}