fix(scrape): 整理后刮削照 MP 对作品目录刮,集只刮本次整理的文件

原地入库逐个文件排刮削,单文件刮削只写该集,剧集根目录(tvshow.nfo、海报、季海报)与季目录 (season.nfo)没有元数据。照 MP ScrapingChain.scrape_metadata_event:按作品目录分组,目录元数据照常初始化,集只刮本次文件;复制 / 移动同样改为只刮本次文件(此前刮整个作品目录)。 - 迁移 0062:scrape_record.videos(null = 目录下全部) - queueScrape 支持 videos;同目录排队中还没开始的刮削合成一批(有一方刮全部则刮全部) - 作品详情「重新刮削」与文件管理刮削不变(目录下全部) Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-02 18:20:52 +08:00
1 parent 501b4eb403
commit f5bc6e9c40
@@ -77,6 +77,8 @@ export interface QueueScrapeInput extends ScrapeLocation {
  actor?: { type: string; id: string | null } | undefined;
  /** 引擎幂等键(入库后自动刮削按整理记录) */
  idempotencyKey?: string | undefined;
+  /** 只刮这些视频(库根相对路径;整理后刮削用,作品 / 季目录元数据照常初始化);缺省 = 目录下全部 */
+  videos?: readonly string[] | undefined;
 }

 type Row = {
@@ -93,6 +95,7 @@ type Row = {
  reason: string | null;
  files: unknown;
  organization_record_id: string | null;
+  videos: string[] | null;
  engine_run_id: string | null;
  requested_at: Date | string;
  started_at: Date | string | null;
@@ -116,6 +119,7 @@ export function scrapeRecordView(r: Row): ScrapeRecord {
    reason: r.reason,
    files: (r.files ?? []) as ScrapeFileResult[],
    organizationRecordId: r.organization_record_id,
+    videos: r.videos ?? null,
    taskId: r.engine_run_id,
    requestedAt: iso(r.requested_at)!,
    startedAt: iso(r.started_at),
@@ -157,7 +161,7 @@ export async function queueScrape(
  const mode = input.mode ?? (await scrapeSettings(d)).scrapeMode;
  const cur = await db
    .selectFrom("scrape_record")
-    .select(["id", "engine_run_id", "media_entity_id", "match"])
+    .select(["id", "engine_run_id", "media_entity_id", "match", "videos"])
    .where("root_id", "=", input.rootId)
    .where("path", "=", input.path)
    .where("mode", "=", mode)
@@ -165,10 +169,17 @@ export async function queueScrape(
    .where("requested_at", ">", new Date(Date.now() - DEDUPE_MS))
    .orderBy("requested_at", "desc")
    .executeTakeFirst();
+  const videos = input.videos?.length ? [...new Set(input.videos)].sort() : null;
  const sameMatch =
    !!cur &&
    (cur.media_entity_id ?? null) === input.match.mediaEntityId &&
    ((cur.match as ScrapeMatch | null)?.externalId ?? null) === input.match.externalId;
+  if (cur && sameMatch) {
+    // 同一目录还没开始的刮削合成一批(MP 按目标目录合批):有一方刮全部则刮全部,否则合并文件
+    const merged = cur.videos && videos ? [...new Set([...cur.videos, ...videos])].sort() : null;
+    if (JSON.stringify(merged) !== JSON.stringify(cur.videos ? [...cur.videos].sort() : null))
+      await db.updateTable("scrape_record").set({ videos: merged }).where("id", "=", cur.id).execute();
+  }
  if (cur && sameMatch && cur.engine_run_id) return { recordId: cur.id, taskId: cur.engine_run_id, deduplicated: true };
  const reused = !!cur && sameMatch;
  const rec = reused
@@ -188,6 +199,7 @@ export async function queueScrape(
          organization_record_id: input.organizationRecordId ?? null,
          actor_type: input.actor?.type ?? (input.trigger === "auto" ? "system" : null),
          actor_id: input.actor?.id ?? null,
+          videos,
        })
        .returning("id")
        .executeTakeFirstOrThrow();
@@ -362,7 +374,7 @@ export async function scrapeEntity(
  return out;
 }

-/** 整理入库成功后:设置打开或本次 Local 补全要求写本地元数据时,复用作品目录刮削任务。 */
+/** 整理入库成功后(设置打开或本次要求刮削):照 MP 对作品目录刮削,集只刮本次整理的文件。 */
 export async function scrapeAfterImport(
  d: AoDeps,
  entityId: string | null,
@@ -376,40 +388,46 @@ export async function scrapeAfterImport(
    if (!s.scrapeAfterImport && !opts.force && opts.enabled !== true) return;
    const organization = await d.ctx.db
      .selectFrom("organization_record")
-      .select(["action", "target_library_root_id"])
+      .select(["target_library_root_id"])
      .where("id", "=", recordId)
      .executeTakeFirst();
-    if (organization?.action === "in_place" && organization.target_library_root_id) {
-      const match = await entityScrapeMatch(d, entityId, "library");
-      const root = await d.ctx.db
-        .selectFrom("library_root")
-        .selectAll()
-        .where("id", "=", organization.target_library_root_id)
-        .executeTakeFirst();
-      if (!match || !root || root.management_mode !== "managed") return;
-      const files = await d.ctx.db
-        .selectFrom("organization_file_result")
-        .select("target_path")
-        .where("organization_record_id", "=", recordId)
-        .where("role", "=", "main")
-        .where("status", "=", "succeeded")
-        .orderBy("ordinal")
-        .execute();
-      for (const file of files)
-        if (file.target_path)
-          await queueScrape(d, {
-            rootId: `lib:${root.id}`,
-            rootPath: root.local_path,
-            libraryRootId: root.id,
-            path: file.target_path,
-            match,
-            trigger: "auto",
-            organizationRecordId: recordId,
-            idempotencyKey: `${SCRAPE_TASK}:record:${recordId}:${file.target_path}`,
-          });
-      return;
+    const match = await entityScrapeMatch(d, entityId, "library");
+    const root = organization?.target_library_root_id
+      ? await d.ctx.db
+          .selectFrom("library_root")
+          .selectAll()
+          .where("id", "=", organization.target_library_root_id)
+          .executeTakeFirst()
+      : undefined;
+    if (!match || !root || root.management_mode !== "managed") return;
+    const files = await d.ctx.db
+      .selectFrom("organization_file_result")
+      .select("target_path")
+      .where("organization_record_id", "=", recordId)
+      .where("role", "=", "main")
+      .where("status", "=", "succeeded")
+      .orderBy("ordinal")
+      .execute();
+    // 照 MP 整理后刮削:按作品目录分组,对作品目录刮(作品 / 季的 NFO 与图片照常初始化),集只刮本次整理的文件;
+    // 视频直接在库根(推不出作品目录)时对文件本身刮。
+    const byDir = new Map<string, string[]>();
+    for (const f of files) {
+      if (!f.target_path) continue;
+      const dir = inferWorkDir(match.kind, f.target_path) ?? f.target_path;
+      byDir.set(dir, [...(byDir.get(dir) ?? []), f.target_path]);
    }
-    await scrapeEntity(d, entityId, { trigger: "auto", organizationRecordId: recordId });
+    for (const [dir, videos] of byDir)
+      await queueScrape(d, {
+        rootId: `lib:${root.id}`,
+        rootPath: root.local_path,
+        libraryRootId: root.id,
+        path: dir,
+        match,
+        trigger: "auto",
+        organizationRecordId: recordId,
+        idempotencyKey: `${SCRAPE_TASK}:record:${recordId}:${dir}`,
+        ...(dir === videos[0] ? {} : { videos }),
+      });
  } catch (err) {
    d.ctx.logger.warn({ entityId, recordId, err: { message: errMessage(err) } }, "入库后自动刮削没排上");
  }
@@ -639,6 +657,8 @@ export async function runScrapeRecord(

    const kind = match.kind;
    const lay = await layoutOf(rootReal, rec.path, isDir, kind);
+    // 整理后刮削:目录布局照常(作品 / 季元数据),集只刮本次整理的文件
+    if (rec.videos) lay.videos = rec.videos.filter((v) => !rec.path || v === rec.path || v.startsWith(`${rec.path}/`));
    if (!lay.videos.length) return await fail("scrape_no_videos");

    // ── 字段:本地作品(有元数据)优先 ──