feat: normalize Japanese artist/title names to romaji at storage time

- Add romajiForStorage() to romaji.js — smart word spacing + Title Case
- ytdlp.js: apply to folder names, filenames, and embedded tags
- novasonic-server.js: apply in edit-metadata endpoint and Spotify resolve
- spotify.js: update track artist/name from iTunes canonical English names

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
HoneyandClaude Sonnet 4.6 committed 2026-05-24 20:22:22 +02:00
1 parent c03463b66b
commit 446c583f7f
4 files changed
+56 -14

No files matched your search

+20 -1
View File
@@ -75,6 +75,25 @@ function toRomaji(text) {
return _hiraToRomaji(hira).toLowerCase();
}
// Human-readable romaji for storage: spaces between words, Title Case, ASCII preserved as-is.
function romajiForStorage(text) {
if (!text || !_tokenizer || !JP_RE.test(text)) return text || '';
const tokens = _tokenizer.tokenize(text);
const parts = tokens.map(t => {
const r = t.reading;
if (!r || r === '*') return t.surface_form; // ASCII — keep original
return _hiraToRomaji(_kataToHira(r)); // Japanese — convert (lowercase)
});
// Add space between adjacent alphabetic segments (word separation for names/titles)
const joined = parts.reduce((acc, part, i) => {
if (i === 0) return part;
const needsSpace = /[a-zA-Z0-9]$/.test(parts[i - 1]) && /^[a-zA-Z]/.test(part);
return acc + (needsSpace ? ' ' : '') + part;
}, '');
// Title-case: capitalize first lowercase letter of each word, preserve existing uppercase
return joined.replace(/\b([a-z])/g, c => c.toUpperCase()).replace(/\s+/g, ' ').trim();
}
// ─── INDEX BUILDER ────────────────────────────────────────────────────────────
const romajiCache = new Map();
@@ -123,4 +142,4 @@ function clearRomajiCache() { romajiCache.clear(); }
function isTokenizerReady() { return !!_tokenizer; }
module.exports = { buildRomajiIndex, clearRomajiCache, isTokenizerReady, fetchNdSongs: _fetchNdSongs };
module.exports = { buildRomajiIndex, clearRomajiCache, isTokenizerReady, fetchNdSongs: _fetchNdSongs, romajiForStorage };
+13 -6
View File
@@ -7,6 +7,7 @@ const {
sanitize, navidromeArtist,
downloadArtwork, findExistingTrack,
} = require('./metadata');
const { romajiForStorage } = require('./romaji');
function buildTitleKeywords(title) {
const VARIANT_WORDS = new Set(['sped','speedup','slowed','reverb','nightcore','karaoke','parody','remix','cover','acoustic','instrumental']);
@@ -49,16 +50,19 @@ function buildVariantFilter(title) {
async function downloadTrack({ jobId, artist, albumArtist, title, album, year, trackNum, totalTracks, discNum, discTotal, compilation, genre, artworkUrl, explicit, outputDir, durationMs, sourceUrl }) {
const mainArtist = albumArtist || artist;
const safeArtist = sanitize(mainArtist.split(/[,;&]/)[0].trim());
const mainArtistRomaji = romajiForStorage(mainArtist) || mainArtist;
const artistRomaji = romajiForStorage(artist) || artist;
const titleRomaji = romajiForStorage(title) || title;
const safeArtist = sanitize(mainArtistRomaji.split(/[,;&]/)[0].trim());
const safeAlbum = sanitize(album || 'Singles');
const dir = path.join(outputDir || CONFIG.musicPath, safeArtist, safeAlbum);
fs.mkdirSync(dir, { recursive: true });
const safeTitle = sanitize(title);
const safeTitle = sanitize(titleRomaji);
const trackPrefix = trackNum ? String(trackNum).padStart(2, '0') + ' - ' : '';
// Filename format: "{01 - }Artist1, Artist2 - Title.ext"
// Use track artist (not albumArtist) so all credited artists appear in the filename
const safeArtistForFile = sanitize(artist.replace(/;/g, ','));
const safeArtistForFile = sanitize(artistRomaji.replace(/;/g, ','));
const fileBase = `${trackPrefix}${safeArtistForFile} - ${safeTitle}`;
const existing = findExistingTrack(dir, trackPrefix, safeTitle, safeArtistForFile);
@@ -299,13 +303,16 @@ async function embedMetadata({ jobId, audioFile, dir, trackPrefix, safeTitle, ar
if (hasCover) ffArgs.push('-i', coverPath);
ffArgs.push('-map', '0:a');
if (hasCover) ffArgs.push('-map', '1:v');
const storedTitle = romajiForStorage(title) || title;
const storedArtist = romajiForStorage(navidromeArtist(artist)) || navidromeArtist(artist);
const storedAlbumArtist = romajiForStorage(mainArtist.replace(/\s+&\s+/g, ', ')) || mainArtist.replace(/\s+&\s+/g, ', ');
ffArgs.push(
'-c', 'copy',
'-map_metadata', '-1',
'-id3v2_version', '3',
'-metadata', `title=${title}`,
'-metadata', `artist=${navidromeArtist(artist)}`,
'-metadata', `album_artist=${mainArtist.replace(/\s+&\s+/g, ', ')}`,
'-metadata', `title=${storedTitle}`,
'-metadata', `artist=${storedArtist}`,
'-metadata', `album_artist=${storedAlbumArtist}`,
);
if (album) ffArgs.push('-metadata', `album=${album}`);
if (year) ffArgs.push('-metadata', `date=${String(year)}`);
+19 -7
View File
@@ -19,7 +19,7 @@ const CONFIG = require('./lib/config');
const { jobs, updateJob, log } = require('./lib/jobs');
const { sanitize, lookupItunesTrackMeta, fetchItunesTracks, searchItunesAlbumTracks } = require('./lib/metadata');
const { downloadTrack, downloadAlbumTracks } = require('./lib/downloader');
const { buildRomajiIndex, clearRomajiCache, isTokenizerReady, fetchNdSongs } = require('./lib/romaji');
const { buildRomajiIndex, clearRomajiCache, isTokenizerReady, fetchNdSongs, romajiForStorage } = require('./lib/romaji');
// ─── INIT ─────────────────────────────────────────────────────────────────────
const app = express();
@@ -225,6 +225,13 @@ app.post('/api/spotify/resolve', async (req, res) => {
setTimeout(() => reject(new Error('Zeitüberschreitung — Spotify API antwortet nicht (>120s)')), 120000)
);
const data = await Promise.race([_resolveSpotifyViaSpotdl(url), timeout]);
// Normalize Japanese artist/title names to romaji before sending to frontend
if (data.tracks) {
for (const t of data.tracks) {
if (t.artistName) t.artistName = romajiForStorage(t.artistName) || t.artistName;
if (t.trackName) t.trackName = romajiForStorage(t.trackName) || t.trackName;
}
}
res.json(data);
} catch(e) {
console.error('[spotify/resolve]', e.message);
@@ -609,6 +616,11 @@ app.post('/api/edit-metadata', async (req, res) => {
const normalArtist = _normalizeArtists(artist, ';');
const normalAlbumArtist = _normalizeArtists(albumArtist, ', ');
// Convert Japanese names to romaji for storage
const romajiArtist = normalArtist ? romajiForStorage(normalArtist) || normalArtist : normalArtist;
const romajiAlbumArtist = normalAlbumArtist ? romajiForStorage(normalAlbumArtist) || normalAlbumArtist : normalAlbumArtist;
const romajiTitle = title ? romajiForStorage(title) || title : title;
// Resolve file path from Navidrome
let filePath;
let ndSong;
@@ -673,9 +685,9 @@ app.post('/api/edit-metadata', async (req, res) => {
ffArgs.push('-map', '0:v?'); // '?' = optional, no error if file has no cover
}
ffArgs.push('-c', 'copy', '-map_metadata', '-1', '-id3v2_version', '3');
if (title) ffArgs.push('-metadata', `title=${title}`);
if (normalArtist) ffArgs.push('-metadata', `artist=${normalArtist}`);
const resolvedAlbumArtist = normalAlbumArtist || (normalArtist ? normalArtist.split(';')[0].trim() : null);
if (romajiTitle) ffArgs.push('-metadata', `title=${romajiTitle}`);
if (romajiArtist) ffArgs.push('-metadata', `artist=${romajiArtist}`);
const resolvedAlbumArtist = romajiAlbumArtist || (romajiArtist ? romajiArtist.split(';')[0].trim() : null);
if (resolvedAlbumArtist) ffArgs.push('-metadata', `album_artist=${resolvedAlbumArtist}`);
if (album) ffArgs.push('-metadata', `album=${album}`);
if (year) ffArgs.push('-metadata', `date=${year}`);
@@ -704,8 +716,8 @@ app.post('/api/edit-metadata', async (req, res) => {
// 1. Rename file if title changed — convention: {01 - }?{Artist} - {Title}.ext
if (title) {
const safeTitle = sanitize(title);
const fileArtist = (normalArtist || ndSong?.artist || '').replace(/;/g, ',');
const safeTitle = sanitize(romajiTitle || title);
const fileArtist = (romajiArtist || ndSong?.artist || '').replace(/;/g, ',');
const safeArtistPart = fileArtist ? sanitize(fileArtist) + ' - ' : '';
const prefix = trackNum ? String(trackNum).padStart(2, '0') + ' - ' : '';
const newName = `${prefix}${safeArtistPart}${safeTitle}${fileExt}`;
@@ -720,7 +732,7 @@ app.post('/api/edit-metadata', async (req, res) => {
const oldAlbumDir = path.dirname(currentFile);
const oldArtistDir = path.dirname(oldAlbumDir);
const newAlbumName = album ? sanitize(album) : path.basename(oldAlbumDir);
const newArtistName = normalAlbumArtist ? sanitize(normalAlbumArtist.split(/[,;&]/)[0].trim()) : path.basename(oldArtistDir);
const newArtistName = romajiAlbumArtist ? sanitize(romajiAlbumArtist.split(/[,;&]/)[0].trim()) : path.basename(oldArtistDir);
const newDir = path.join(CONFIG.musicPath, newArtistName, newAlbumName);
if (newDir !== oldAlbumDir) {
+4
View File
@@ -295,6 +295,10 @@ async function _loadSpotifyTrackArtwork(tracks, listEl) {
if (match.trackExplicitness) t.trackExplicitness = match.trackExplicitness;
if (match.previewUrl) t.previewUrl = match.previewUrl;
if (match.trackTimeMillis) t.durationMs = match.trackTimeMillis;
// Use iTunes canonical English/romaji names — more accurate than Spotify (which may use Japanese)
if (match.artistName) t.artistName = match.artistName;
const JP_RE = /[぀-ヿ一-鿿]/;
if (JP_RE.test(t.trackName) && match.trackName) t.trackName = match.trackName;
if (window._spotifyResolved?.tracks?.[idx]) Object.assign(window._spotifyResolved.tracks[idx], t);
const url = match.artworkUrl100 || match.artworkUrl60 || '';
if (url) _setArt(artEls[idx], url);