fix(character-dictionary): add surname honorifics for Japanese localized aliases (#87)

src/main/character-dictionary-runtime/constants.ts

@@ -1,7 +1,7 @@
 export const ANILIST_GRAPHQL_URL = 'https://graphql.anilist.co';
 export const ANILIST_REQUEST_DELAY_MS = 2000;
 export const CHARACTER_IMAGE_DOWNLOAD_DELAY_MS = 250;
-export const CHARACTER_DICTIONARY_FORMAT_VERSION = 17;
+export const CHARACTER_DICTIONARY_FORMAT_VERSION = 18;
 export const CHARACTER_DICTIONARY_MERGED_TITLE = 'SubMiner Character Dictionary';
 
 export const HONORIFIC_SUFFIXES = [

src/main/character-dictionary-runtime/term-building.test.ts

@@ -0,0 +1,38 @@
+import assert from 'node:assert/strict';
+import test from 'node:test';
+import { buildNameTerms } from './term-building';
+import type { CharacterRecord } from './types';
+
+function characterRecord(overrides: Partial<CharacterRecord>): CharacterRecord {
+  return {
+    id: 136073,
+    role: 'primary',
+    firstNameHint: 'Chi-Yul',
+    fullName: 'Chi-Yul Song',
+    lastNameHint: 'Song',
+    nativeName: '송치율',
+    alternativeNames: [],
+    bloodType: '',
+    birthday: null,
+    description: '',
+    imageUrl: null,
+    age: '',
+    sex: '',
+    voiceActors: [],
+    ...overrides,
+  };
+}
+
+test('buildNameTerms adds surname honorifics from Japanese localized aliases', () => {
+  const terms = buildNameTerms(
+    characterRecord({
+      alternativeNames: ['Isao Mabuchi (馬渕勲)', 'Chi-Yeol (송치열)'],
+    }),
+  );
+
+  assert.ok(terms.includes('馬渕勲'));
+  assert.ok(terms.includes('馬渕勲さん'));
+  assert.ok(terms.includes('馬渕'));
+  assert.ok(terms.includes('馬渕さん'));
+  assert.ok(!terms.includes('송치'));
+});

src/main/character-dictionary-runtime/term-building.ts

@@ -3,6 +3,7 @@ import {
   addRomanizedKanaAliases,
   buildReading,
   buildReadingFromRomanized,
+  containsKanji,
   hasKanaOnly,
   isRomanizedName,
   splitJapaneseName,
@@ -39,6 +40,25 @@ function expandRawNameVariants(rawName: string): string[] {
   return [...variants];
 }
 
+function isJapaneseNameSplitCandidate(name: string): boolean {
+  const compact = name.replace(/[\s\u3000・･·•]/g, '');
+  return (
+    containsKanji(compact) && /^[\u3040-\u30ff\u3400-\u4dbf\u4e00-\u9fff々〆ヵヶー]+$/.test(compact)
+  );
+}
+
+function addJapaneseNameParts(character: CharacterRecord, name: string, terms: Set<string>): void {
+  if (!isJapaneseNameSplitCandidate(name)) return;
+
+  const nameParts = splitJapaneseName(name, character.firstNameHint, character.lastNameHint);
+  if (nameParts.family) {
+    terms.add(nameParts.family);
+  }
+  if (nameParts.given) {
+    terms.add(nameParts.given);
+  }
+}
+
 export function buildNameTerms(character: CharacterRecord): string[] {
   const base = new Set<string>();
   const romanizedBase = new Set<string>();
@@ -73,6 +93,10 @@ export function buildNameTerms(character: CharacterRecord): string[] {
           target.add(part);
         }
       }
+
+      if (target === base) {
+        addJapaneseNameParts(character, name, base);
+      }
     }
   }