feat(extension+backend): 插件直接在浏览器里抓 B 站字幕，跳过后端 download_subtitles

之前 B 站字幕优先逻辑放在后端的 BilibiliSubtitleFetcher，需要后端通过 CookieConfigManager
管理 SESSDATA cookie 才能拿 AI 字幕。这次改为：插件在用户浏览器里直接抓字幕，
天然带着用户当前登录态的 cookie；后端只负责把传过来的字幕当作转写缓存。

extension：
- 新增 logic/bilibili-subtitle.ts，调 /x/web-interface/view → /x/player/wbi/v2 → 字幕 URL JSON
  · service worker fetch 走 credentials:'include'，借 manifest host_permissions:'*://*/*'
    自动带 .bilibili.com 域 cookie，并绕过 CORS
  · 优先级：人工中文 > AI 中文 > 任意非空
- popup start() 与 background startTask() 在 platform === 'bilibili' 时先调一次抓取，
  结果作为 prefetched_transcript 字段塞到 /api/generate_note payload
- types.ts GenerateRequest 增加 prefetched_transcript 字段

backend：
- VideoRequest 增加可选 prefetched_transcript: dict
- generate_note endpoint 收到时调 _persist_prefetched_transcript() 写到
  NOTE_OUTPUT_DIR/<task_id>_transcript.json；NoteGenerator 的 cache-hit 逻辑天然命中，
  跳过 downloader.download_subtitles 和音频转写，直接走 GPT 总结
- 字幕清洗：去掉空 segment、必要时合成 full_text、language 默认 'zh'

效果：B 站登录用户的视频，从用户点击到 GPT 拿到全文，省掉一次后端 → B 站 API 的来回，
也彻底告别了 backend 那侧的 cookie 配置心智负担。

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

BillNote_extension/src/background/main.ts

@@ -2,6 +2,7 @@ import { onMessage } from 'webext-bridge/background'
 import type { Settings, TaskRecord } from '~/logic/types'
 import { DEFAULT_SETTINGS, MAX_TASKS, SETTINGS_KEY, TASKS_KEY } from '~/logic/constants'
 import { detectPlatform } from '~/logic/platform'
+import { fetchBilibiliSubtitle } from '~/logic/bilibili-subtitle'
 
 // only on dev mode
 if (import.meta.hot) {
@@ -65,6 +66,10 @@ async function startTask(url: string): Promise<{ ok: boolean, taskId?: string, e
     return { ok: false, error: '请先在设置页选择供应商与模型' }
 
   const backend = settings.backendUrl.replace(/\/$/, '')
+
+  // B 站：先在浏览器里抓字幕（带本地登录态 cookie），随提交带过去
+  const prefetched = platform === 'bilibili' ? await fetchBilibiliSubtitle(url) : null
+
   try {
     const res = await fetch(`${backend}/api/generate_note`, {
       method: 'POST',
@@ -82,6 +87,7 @@ async function startTask(url: string): Promise<{ ok: boolean, taskId?: string, e
           ...(settings.screenshot ? ['screenshot'] : []),
           ...(settings.link ? ['link'] : []),
         ],
+        prefetched_transcript: prefetched ?? undefined,
       }),
     })
     if (!res.ok)

BillNote_extension/src/logic/bilibili-subtitle.ts

@@ -0,0 +1,125 @@
+// 在浏览器里直接调 B 站 player API 抓字幕。
+// 因为 manifest host_permissions: '*://*/*' 覆盖 api.bilibili.com，service worker 里的
+// fetch 会自动带 .bilibili.com 域下的用户 cookie，并且绕过 CORS——AI 字幕需要登录态，
+// 这等于用用户当前浏览器的登录身份代替了 backend 那边的 SESSDATA 配置。
+//
+// 与 backend/app/downloaders/bilibili_subtitle.py 的 BilibiliSubtitleFetcher 行为对齐。
+
+const UA
+  = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36'
+
+export interface PrefetchedTranscript {
+  language: string
+  full_text: string
+  segments: Array<{ start: number, end: number, text: string }>
+  source: 'bilibili_extension'
+}
+
+interface SubtitleEntry {
+  lan?: string
+  ai_type?: number
+  subtitle_url?: string
+}
+
+function extractBvid(url: string): string | null {
+  const m = url.match(/BV([0-9A-Za-z]+)/)
+  return m ? `BV${m[1]}` : null
+}
+
+async function jsonGet<T>(url: string): Promise<T | null> {
+  try {
+    const res = await fetch(url, {
+      credentials: 'include',
+      headers: { 'User-Agent': UA, 'Referer': 'https://www.bilibili.com' },
+    })
+    if (!res.ok)
+      return null
+    return await res.json() as T
+  }
+  catch (e) {
+    console.warn('[bilinote] B 站 API 请求失败:', url, e)
+    return null
+  }
+}
+
+async function getCid(bvid: string): Promise<number | null> {
+  const data = await jsonGet<{ code: number, data?: { cid?: number } }>(
+    `https://api.bilibili.com/x/web-interface/view?bvid=${bvid}`,
+  )
+  if (!data || data.code !== 0)
+    return null
+  return data.data?.cid ?? null
+}
+
+async function listSubtitles(bvid: string, cid: number): Promise<SubtitleEntry[]> {
+  const data = await jsonGet<{
+    code: number
+    data?: { subtitle?: { subtitles?: SubtitleEntry[] } }
+  }>(`https://api.bilibili.com/x/player/wbi/v2?bvid=${bvid}&cid=${cid}`)
+  if (!data || data.code !== 0)
+    return []
+  return data.data?.subtitle?.subtitles ?? []
+}
+
+function pickSubtitle(subtitles: SubtitleEntry[]): SubtitleEntry | null {
+  if (!subtitles.length)
+    return null
+  const isZh = (s: SubtitleEntry) => {
+    const lan = (s.lan || '').toLowerCase()
+    return lan.startsWith('zh') || lan === 'ai-zh'
+  }
+  // 优先级：人工中文 > AI 中文 > 任意非空
+  return (
+    subtitles.find(s => isZh(s) && !s.ai_type)
+    || subtitles.find(s => isZh(s))
+    || subtitles[0]
+  )
+}
+
+function normalizeUrl(url: string): string {
+  return url.startsWith('//') ? `https:${url}` : url
+}
+
+interface SubtitleBody {
+  body?: Array<{ from?: number, to?: number, content?: string }>
+}
+
+export async function fetchBilibiliSubtitle(videoUrl: string): Promise<PrefetchedTranscript | null> {
+  const bvid = extractBvid(videoUrl)
+  if (!bvid)
+    return null
+
+  const cid = await getCid(bvid)
+  if (!cid)
+    return null
+
+  const subtitles = await listSubtitles(bvid, cid)
+  const track = pickSubtitle(subtitles)
+  if (!track?.subtitle_url) {
+    console.info(`[bilinote] B 站 ${bvid} 没找到可用字幕轨（可能未登录或视频无字幕）`)
+    return null
+  }
+
+  const sub = await jsonGet<SubtitleBody>(normalizeUrl(track.subtitle_url))
+  const body = sub?.body || []
+  const segments: PrefetchedTranscript['segments'] = []
+  for (const item of body) {
+    const text = (item.content || '').trim()
+    if (!text)
+      continue
+    segments.push({
+      start: Number(item.from || 0),
+      end: Number(item.to || 0),
+      text,
+    })
+  }
+  if (!segments.length)
+    return null
+
+  return {
+    language: track.lan || 'zh',
+    full_text: segments.map(s => s.text).join(' '),
+    segments,
+    source: 'bilibili_extension',
+  }
+}

BillNote_extension/src/logic/types.ts

@@ -40,6 +40,13 @@ export interface GenerateRequest {
   format?: string[]
   style?: string
   extras?: string
+  // 客户端在浏览器里直接抓到的字幕，跳过后端的 download_subtitles + 音频转写
+  prefetched_transcript?: {
+    language: string
+    full_text: string
+    segments: Array<{ start: number, end: number, text: string }>
+    source?: string
+  }
 }
 
 export interface NoteResult {

BillNote_extension/src/popup/Popup.vue

@@ -3,6 +3,7 @@ import { computed, onMounted, onUnmounted, ref } from 'vue'
 import { detectPlatform } from '~/logic/platform'
 import { settings, settingsReady, tasks, tasksReady, upsertTask } from '~/logic/storage'
 import { generateNote, getTaskStatus, resolveImageUrl } from '~/logic/api'
+import { fetchBilibiliSubtitle } from '~/logic/bilibili-subtitle'
 import type { TaskRecord } from '~/logic/types'
 
 const tabUrl = ref<string>('')
@@ -64,6 +65,8 @@ async function start() {
   }
   submitting.value = true
   try {
+    // B 站：在用户浏览器里直接抓字幕（带本地登录态 cookie），跳过后端的 download_subtitles 与音频转写
+    const prefetched = platform.value === 'bilibili' ? await fetchBilibiliSubtitle(tabUrl.value) : null
     const { task_id } = await generateNote({
       video_url: tabUrl.value,
       platform: platform.value!,
@@ -77,6 +80,7 @@ async function start() {
         ...(settings.value.screenshot ? ['screenshot'] : []),
         ...(settings.value.link ? ['link'] : []),
       ],
+      prefetched_transcript: prefetched ?? undefined,
     })
     activeTaskId.value = task_id
     upsertTask({

backend/app/routers/note.py

@@ -50,6 +50,10 @@ class VideoRequest(BaseModel):
     video_understanding: Optional[bool] = False
     video_interval: Optional[int] = 0
     grid_size: Optional[list] = []
+    # 客户端（如浏览器插件）已经在用户浏览器里抓到字幕，直接传给后端复用，
+    # 跳过 download_subtitles 和音频转写。形如：
+    #   {"language": "zh", "full_text": "...", "segments": [{"start","end","text"}, ...]}
+    prefetched_transcript: Optional[dict] = None
 
     @field_validator("video_url")
     def validate_supported_url(cls, v):
@@ -74,6 +78,40 @@ def save_note_to_file(task_id: str, note):
         json.dump(asdict(note), f, ensure_ascii=False, indent=2)
 
 
+def _persist_prefetched_transcript(task_id: str, transcript: dict) -> None:
+    """把客户端预取的字幕写到 NoteGenerator 期望的转写缓存文件里。
+
+    NoteGenerator.generate 会优先读 <task_id>_transcript.json，命中即跳过 download_subtitles
+    与音频转写流程。要求字段：language(可空)/full_text/segments[{start,end,text}]
+    """
+    segments = transcript.get("segments") or []
+    cleaned_segments = []
+    for s in segments:
+        text = (s.get("text") or "").strip()
+        if not text:
+            continue
+        cleaned_segments.append({
+            "start": float(s.get("start", 0)),
+            "end": float(s.get("end", 0)),
+            "text": text,
+        })
+    if not cleaned_segments:
+        raise ValueError("prefetched_transcript 没有可用的 segments")
+
+    full_text = transcript.get("full_text") or " ".join(s["text"] for s in cleaned_segments)
+    payload = {
+        "language": transcript.get("language") or "zh",
+        "full_text": full_text,
+        "segments": cleaned_segments,
+    }
+
+    os.makedirs(NOTE_OUTPUT_DIR, exist_ok=True)
+    target = os.path.join(NOTE_OUTPUT_DIR, f"{task_id}_transcript.json")
+    with open(target, "w", encoding="utf-8") as f:
+        json.dump(payload, f, ensure_ascii=False, indent=2)
+    logger.info(f"已写入客户端预取字幕缓存: {target} ({len(cleaned_segments)} 段)")
+
+
 def run_note_task(task_id: str, video_url: str, platform: str, quality: DownloadQuality,
                   link: bool = False, screenshot: bool = False, model_name: str = None, provider_id: str = None,
                   _format: list = None, style: str = None, extras: str = None, video_understanding: bool = False,
@@ -163,6 +201,13 @@ def generate_note(data: VideoRequest, background_tasks: BackgroundTasks):
         # 统一先写入 PENDING，表示已进入队列等待串行执行
         NoteGenerator()._update_status(task_id, TaskStatus.PENDING)
 
+        # 客户端已经抓好字幕的话，写到转写缓存文件，NoteGenerator 的 cache-hit 逻辑会直接用上
+        if data.prefetched_transcript:
+            try:
+                _persist_prefetched_transcript(task_id, data.prefetched_transcript)
+            except Exception as e:
+                logger.warning(f"写入预取字幕失败 (task_id={task_id}): {e}")
+
         background_tasks.add_task(run_note_task, task_id, data.video_url, data.platform, data.quality, data.link,
                                   data.screenshot, data.model_name, data.provider_id, data.format, data.style,
                                   data.extras, data.video_understanding, data.video_interval, data.grid_size)