music: 新建 music app，替换 piano-sheet

听歌 + 练琴曲目管理： - 数据：piece (title/artist/category/lyrics/play_count/notes) + attachments (audio/video/pdf/image; image 带 role=chord/numbered/staff) - 后端 axum + sqlite，附件流式落 PVC，ServeFile 支持 Range（视频拖动） - 前端 guitar 风格 player：左 sidebar + tabs(歌词/吉他谱/简谱/五线谱/PDF/视频)，LRC 同步、快捷键、笔记自动保存 - ns cube-music + music.famzheng.me + bodylimit 5GiB - scripts/import_guitar.py 用于把 oci /data/guitar/ 旧曲库导入
2026-05-09 22:36:14 +01:00
parent 58f344db85
commit 1a8f297302
30 changed files with 2683 additions and 1314 deletions
@@ -0,0 +1,222 @@
+#!/usr/bin/env python3
+"""
+从 /data/guitar/ 把旧 guitar app 的曲库导入到 music API。
+
+跑法（在 oci 上，源数据原地）:
+    python3 import_guitar.py --src /data/guitar --api https://music.famzheng.me
+
+每个 mp3 = 一个 piece：
+- title / artist 从文件名 "{title} - {artist}.mp3" 拆
+- lyrics 来自同名 .lrc（UTF-8）
+- chord png 来自 chords/{sanitize(artist)}-{sanitize(title)}.png（如有）
+- play_count 来自 playcounts.json[mp3 文件名]
+- category 设为 "流行"（旧 guitar 里都是流行/吉他歌）
+
+幂等：以 (title, artist) 作主键去重，已有 piece 跳过附件重传。
+"""
+
+import argparse
+import json
+import re
+import sys
+import urllib.request
+import urllib.parse
+import urllib.error
+import os
+from pathlib import Path
+
+CATEGORY = "流行"
+
+
+def sanitize(text: str) -> str:
+    """匹配旧 chord_server.py 的 sanitize：替换非法文件名字符为 -"""
+    return re.sub(r'[<>:"/\\|?*]', '-', text).strip()
+
+
+def parse_filename(stem: str) -> tuple[str, str]:
+    """'{title} - {artist}' → (title, artist)"""
+    parts = stem.split(' - ')
+    title = parts[0].strip()
+    artist = ' - '.join(parts[1:]).strip() if len(parts) > 1 else ''
+    return title, artist
+
+
+def http_request(method: str, url: str, *, json_body=None, file_field=None,
+                 timeout: int = 600) -> dict:
+    """简单 HTTP 客户端，return JSON dict。失败抛异常。"""
+    headers = {}
+    data = None
+    if json_body is not None:
+        data = json.dumps(json_body).encode('utf-8')
+        headers['Content-Type'] = 'application/json'
+    elif file_field is not None:
+        boundary = '----music-import-' + os.urandom(8).hex()
+        body = []
+        for field_name, filename, content, mime in file_field:
+            body.append(f'--{boundary}\r\n'.encode())
+            disp = (f'Content-Disposition: form-data; name="{field_name}"; '
+                    f'filename="{filename}"\r\n').encode()
+            body.append(disp)
+            body.append(f'Content-Type: {mime}\r\n\r\n'.encode())
+            body.append(content)
+            body.append(b'\r\n')
+        body.append(f'--{boundary}--\r\n'.encode())
+        data = b''.join(body)
+        headers['Content-Type'] = f'multipart/form-data; boundary={boundary}'
+    req = urllib.request.Request(url, data=data, method=method, headers=headers)
+    with urllib.request.urlopen(req, timeout=timeout) as resp:
+        return json.loads(resp.read().decode('utf-8'))
+
+
+def list_pieces(api: str) -> list:
+    return http_request('GET', f'{api}/api/pieces')
+
+
+def create_piece(api: str, title: str, artist: str, lyrics: str | None) -> int:
+    body = {
+        'title': title,
+        'artist': artist or None,
+        'category': CATEGORY,
+        'lyrics': lyrics or None,
+    }
+    r = http_request('POST', f'{api}/api/pieces', json_body=body)
+    return int(r['id'])
+
+
+def patch_play_count(api: str, piece_id: int, count: int):
+    http_request('PATCH', f'{api}/api/pieces/{piece_id}',
+                 json_body={'play_count': count})
+
+
+def upload_attachment(api: str, piece_id: int, path: Path,
+                      mime: str, role: str | None = None) -> dict:
+    url = f'{api}/api/pieces/{piece_id}/attachments'
+    if role:
+        url += '?role=' + urllib.parse.quote(role)
+    bytes_content = path.read_bytes()
+    return http_request('POST', url, file_field=[
+        ('files', path.name, bytes_content, mime),
+    ])
+
+
+def detect_audio_mime(path: Path) -> str:
+    ext = path.suffix.lower()
+    return {
+        '.mp3': 'audio/mpeg',
+        '.m4a': 'audio/mp4',
+        '.flac': 'audio/flac',
+        '.ogg': 'audio/ogg',
+        '.wav': 'audio/wav',
+    }.get(ext, 'application/octet-stream')
+
+
+def main():
+    ap = argparse.ArgumentParser()
+    ap.add_argument('--src', default='/data/guitar', help='guitar 数据根目录')
+    ap.add_argument('--api', required=True, help='music API base，e.g. https://music.famzheng.me')
+    ap.add_argument('--dry-run', action='store_true', help='只打印不实际调用')
+    ap.add_argument('--limit', type=int, default=0, help='测试用，最多导入 N 首（0=全部）')
+    args = ap.parse_args()
+
+    src = Path(args.src)
+    if not src.exists():
+        print(f'[!] {src} 不存在', file=sys.stderr)
+        sys.exit(1)
+
+    chords_dir = src / 'chords'
+    playcounts_file = src / 'playcounts.json'
+    playcounts = {}
+    if playcounts_file.exists():
+        try:
+            playcounts = json.loads(playcounts_file.read_text())
+        except Exception as e:
+            print(f'[!] 读 playcounts.json 失败: {e}', file=sys.stderr)
+
+    # 已有 pieces 去重
+    existing = {}
+    if not args.dry_run:
+        try:
+            for p in list_pieces(args.api):
+                key = (p['title'], p.get('artist') or '')
+                existing[key] = p['id']
+            print(f'[i] 远端已有 {len(existing)} 首曲目，重复的会跳过创建')
+        except urllib.error.URLError as e:
+            print(f'[!] 连不上 {args.api}: {e}', file=sys.stderr)
+            sys.exit(1)
+
+    mp3s = sorted(src.glob('*.mp3'))
+    if args.limit:
+        mp3s = mp3s[:args.limit]
+
+    total = len(mp3s)
+    ok = skipped = failed = 0
+
+    for i, mp3 in enumerate(mp3s, 1):
+        stem = mp3.stem
+        title, artist = parse_filename(stem)
+        if not title:
+            print(f'[!] [{i}/{total}] 无法解析文件名: {mp3.name}', file=sys.stderr)
+            failed += 1
+            continue
+
+        key = (title, artist)
+        prefix = f'[{i}/{total}] {title} - {artist}'
+
+        if key in existing:
+            print(f'  ↻ {prefix} (已存在 id={existing[key]}, 跳过)')
+            skipped += 1
+            continue
+
+        # lrc
+        lrc_path = mp3.with_suffix('.lrc')
+        lyrics = None
+        if lrc_path.exists():
+            try:
+                lyrics = lrc_path.read_text(encoding='utf-8')
+            except UnicodeDecodeError:
+                lyrics = lrc_path.read_text(encoding='gbk', errors='replace')
+
+        # chord png
+        chord_path = chords_dir / f'{sanitize(artist)}-{sanitize(title)}.png'
+        if not chord_path.exists():
+            chord_path = None
+
+        play_count = int(playcounts.get(mp3.name, 0))
+
+        if args.dry_run:
+            print(f'  + {prefix}'
+                  f'{" [词]" if lyrics else ""}'
+                  f'{" [谱]" if chord_path else ""}'
+                  f'{f" [{play_count}次]" if play_count else ""}')
+            ok += 1
+            continue
+
+        try:
+            piece_id = create_piece(args.api, title, artist, lyrics)
+            print(f'  + {prefix} → id={piece_id}', end='', flush=True)
+
+            upload_attachment(args.api, piece_id, mp3,
+                              detect_audio_mime(mp3))
+            print(' [audio]', end='', flush=True)
+
+            if chord_path:
+                upload_attachment(args.api, piece_id, chord_path,
+                                  'image/png', role='chord')
+                print(' [chord]', end='', flush=True)
+
+            if play_count:
+                patch_play_count(args.api, piece_id, play_count)
+                print(f' [{play_count}次]', end='', flush=True)
+
+            print()
+            ok += 1
+        except Exception as e:
+            print(f'\n[!] {prefix} 失败: {e}', file=sys.stderr)
+            failed += 1
+
+    print()
+    print(f'完成: ok={ok}  skipped={skipped}  failed={failed}  total={total}')
+
+
+if __name__ == '__main__':
+    main()