fix(sites): normalize HHClub RSS items, complete detail page parsing (HH-2)

- RSS: split NexusPHP "[category]name[subtitle][size][uploader]" titles in the
  site adapter layer (category id/name/media type, clean title, subtitle,
  size); snapshot fields win on merge; lazily renormalize old RSS rows and
  re-emit release.observed; unknown list fields show an em dash
- Detail: HHClub section layout (douban info block as description,
  screenshots section, IMDb/TMDB from MediaInfo lines, upload time field,
  on-demand viewfilelist.php once); site CDN screenshots get a Referer
- Site request count now includes site-bound RSS fetches
- Origin view shows "自动化" when idle and "Agent" only after recent
  automation actions (lastAutomationAt); agent step titles no longer get
  squeezed to a sliver by long notes

Co-Authored-By: Claude Opus 5.5 <[email protected]>

packages/adapters/src/sites/spider/detail-page.ts

@@ -25,10 +25,13 @@ type El = ReturnType<CheerioAPI>;
 
 const LABELS = {
   description: ["简介", "Description"],
+  /** 没有简介容器时的次选:豆瓣 / IMDb 信息块(憨憨把豆瓣资料拆成独立一节) */
+  descriptionFallback: ["豆瓣", "豆瓣信息", "影片信息", "IMDb信息", "IMDb"],
+  screenshots: ["截图", "Screenshots", "Screenshot"],
   mediaInfo: ["MediaInfo", "媒体信息", "BDInfo"],
   tags: ["标签", "Tags"],
   uploader: ["上传者", "发布者", "发布人", "上传人", "Uploader"],
-  files: ["文件列表", "文件清单", "File list", "Files"],
+  files: ["文件列表", "文件清单", "种子文件", "File list", "Files"],
 };
 
 /** 简介里可能装 MediaInfo 的块 */
@@ -62,6 +65,19 @@ function fileFromText(text: string): SiteDetailFile | null {
   return HAS_EXT.test(t) ? { name: t, sizeBytes: null } : null;
 }
 
+/** 文本里的 IMDb ID:imdb.com/title 链接或「IMDB : tt…」一行;不接受孤立的 tt 串(避免误中哈希) */
+export function imdbFromText(text: string): string | null {
+  const m =
+    /imdb\.com\/title\/(tt\d{7,8})\b/i.exec(text) ??
+    /\bIMDb\s*(?:ID|链接|編號|编号)?\s*[:：]?\s*(tt\d{7,8})\b/i.exec(text);
+  return m?.[1] ?? null;
+}
+
+/** NexusPHP viewfilelist.php 返回的文件表(按需 AJAX 取到的 HTML 片段) */
+export function filesFromFileListHtml($: CheerioAPI): SiteDetailFile[] {
+  return filesFrom($, $("tr"));
+}
+
 function filesFrom($: CheerioAPI, rows: El): SiteDetailFile[] {
   const out: SiteDetailFile[] = [];
   rows.each((_, n) => {
@@ -70,7 +86,7 @@ function filesFrom($: CheerioAPI, rows: El): SiteDetailFile[] {
     const cells = row.children("td");
     if (cells.length >= 2) {
       const name = cells.first().text().replace(/\s+/g, " ").trim();
-      const size = cells.eq(1).text().trim();
+      const size = cells.eq(1).text().replace(/\s+/g, " ").trim();
       if (name && HAS_EXT.test(name) && /\d/.test(size)) out.push({ name, sizeBytes: parseSizeMp(size) || null });
       return;
     }
@@ -106,6 +122,7 @@ export function extractDetailPage($: CheerioAPI, conf: DetailConf, base: string)
   let desc: El | null = descSel ? $(descSel).first() : null;
   if (!desc?.length) desc = byLabel(LABELS.description);
   page.probe.hasDescription = !!desc?.length;
+  const infoBlock = desc?.length ? null : byLabel(LABELS.descriptionFallback);
 
   // MediaInfo:站点选择器 → 字段标签 → 简介里的引用 / 代码块(取最内层文本,整块从简介里去掉)
   let mediaInfoEl: El | null = null;
@@ -137,12 +154,19 @@ export function extractDetailPage($: CheerioAPI, conf: DetailConf, base: string)
 
   // 截图
   const imgs: string[] = [];
-  const imgEls = p.screenshots ? $(p.screenshots) : desc?.length ? desc.find("img") : null;
-  imgEls?.each((_, n) => {
-    const e = $(n);
-    const u = screenshotUrl(e.attr("data-src") ?? e.attr("data-original") ?? e.attr("src"), base);
-    if (u && !imgs.includes(u)) imgs.push(u);
-  });
+  const pushImgs = (els: El | null) =>
+    els?.each((_, n) => {
+      const e = $(n);
+      const u = screenshotUrl(e.attr("data-src") ?? e.attr("data-original") ?? e.attr("src"), base);
+      if (u && !imgs.includes(u)) imgs.push(u);
+    });
+  if (p.screenshots) pushImgs($(p.screenshots));
+  else {
+    // 独立的「截图」一节(憨憨)优先;没有再取简介里的图
+    const shotBlock = byLabel(LABELS.screenshots);
+    if (shotBlock) pushImgs(shotBlock.is("img") ? shotBlock : shotBlock.find("img"));
+    if (!imgs.length && desc?.length) pushImgs(desc.find("img"));
+  }
   page.probe.imageCount = imgs.length;
   page.screenshots = imgs.slice(0, MAX_SCREENSHOTS);
 
@@ -153,18 +177,36 @@ export function extractDetailPage($: CheerioAPI, conf: DetailConf, base: string)
     c.find("img").remove();
     const t = blockText($, c);
     if (t) page.descriptionText = t.slice(0, 20_000);
+  } else if (infoBlock) {
+    const c = infoBlock.clone();
+    c.find("img").remove();
+    // 块里的「标签」span 与值之间补上冒号 / 换行,读起来是「片名:xxx」一行一项
+    c.find(p.labelSelectors?.join(", ") || "span.font-bold").each((_, n) => {
+      const e = $(n);
+      const t = e.text().trim();
+      if (t && t.length <= 16 && !/[:：]$/.test(t)) e.text(`${t}：`);
+      e.before("\n");
+    });
+    const t = blockText($, c)
+      .replace(/[ \t]*\n[ \t]*/g, "\n")
+      .replace(/：\s*\n(?=\S)/g, "：");
+    if (t) page.descriptionText = t.slice(0, 20_000);
   }
 
-  // 外部 ID:页面上的链接优先,其次简介正文
-  const descHtml = desc?.html() ?? "";
+  // 外部 ID:页面上的链接优先,其次简介 / 信息块正文,再次 MediaInfo 里的 IMDB / TMDB 行(憨憨的 MediaInfo 带这两行)
+  const descHtml = `${desc?.html() ?? ""}\n${infoBlock?.html() ?? ""}`;
+  const idText = `${descHtml}\n${page.mediaInfo ?? ""}`;
   const f = conf.fields ?? {};
   page.imdbId =
-    extractImdbId(mcSelect($, null, f.imdb ?? "a[href*='imdb.com']::attr(href)")) ?? extractImdbId(descHtml) ?? null;
+    extractImdbId(mcSelect($, null, f.imdb ?? "a[href*='imdb.com']::attr(href)")) ??
+    imdbFromText(idText) ??
+    extractImdbId(descHtml) ??
+    null;
   page.doubanId =
     extractDoubanId(mcSelect($, null, f.douban ?? "a[href*='douban.com']::attr(href)")) ??
-    extractDoubanId(descHtml) ??
+    extractDoubanId(idText) ??
     null;
-  page.tmdb = extractTmdb($("a[href*='themoviedb.org']").first().attr("href")) ?? extractTmdb(descHtml);
+  page.tmdb = extractTmdb($("a[href*='themoviedb.org']").first().attr("href")) ?? extractTmdb(idText);
 
   // 文件清单:站点选择器 → MovieClaw file_list(去伪元素)→ 字段标签 → #filelist 表格
   const fileSel = p.files ?? f.fileList?.replace(/::text$/, "");