feat: DSL 磁力聚合 — 搜索/解析/预览/投递 + cilisousuo 源 + 中英双语 + Docker 部署

2026-10-16 18:53:45 +08:00
commit d6999104bf
64 changed files with 16524 additions and 0 deletions
@@ -0,0 +1,145 @@
+import { buildMagnet } from '../torrent'
+import { DEFAULT_UA, httpText } from '../http'
+import type { ProviderContext, SearchProvider, SearchResult } from './types'
+
+const HOMEPAGE = 'https://cilisousuo.cc'
+
+const SIZE_UNITS: Record<string, number> = {
+  b: 1,
+  kb: 1024,
+  kib: 1024,
+  mb: 1024 ** 2,
+  mib: 1024 ** 2,
+  gb: 1024 ** 3,
+  gib: 1024 ** 3,
+  tb: 1024 ** 4,
+  tib: 1024 ** 4
+}
+
+/** "330.3MB" / "1.1TB" → 字节数 */
+function parseSize(input: unknown): number {
+  const text = String(input || '').trim()
+  const m = text.match(/^([\d.]+)\s*([a-zA-Z]+)?$/)
+  if (!m) return 0
+  const value = Number(m[1]) || 0
+  const unit = (m[2] || 'b').toLowerCase()
+  return Math.round(value * (SIZE_UNITS[unit] ?? 1))
+}
+
+function cilisousuoBase() {
+  return (process.env.CILISOUSUO_BASE || HOMEPAGE).replace(/\/+$/, '')
+}
+
+/** 去掉 HTML 标签并解码常见实体 */
+function stripHtml(input: string): string {
+  return input
+    .replace(/<[^>]+>/g, '')
+    .replace(/&nbsp;/g, ' ')
+    .replace(/&amp;/g, '&')
+    .replace(/&lt;/g, '<')
+    .replace(/&gt;/g, '>')
+    .replace(/&#0?39;|&apos;/g, "'")
+    .replace(/&quot;/g, '"')
+    .replace(/\s+/g, ' ')
+    .trim()
+}
+
+interface ListItem {
+  title: string
+  filename: string
+  size: number
+  code: string
+}
+
+/** 从列表页解析出条目（列表页只含名称/大小与 /magnet/{code}，infoHash 在详情页） */
+function parseList(html: string): ListItem[] {
+  const items: ListItem[] = []
+  const liRe = /<li class="item">([\s\S]*?)<\/li>/g
+  let m: RegExpExecArray | null
+  while ((m = liRe.exec(html))) {
+    const block = m[1]
+    const titleM = block.match(/<div class="result-title">([\s\S]*?)<\/div>/)
+    const fileM = block.match(/<div class="filename">([\s\S]*?)<\/div>/)
+    const sizeM = block.match(/<div class="size">([\s\S]*?)<\/div>/)
+    const codeM = block.match(/href="\/magnet\/([^"]+)"/)
+    if (!codeM) continue
+    const title = titleM ? stripHtml(titleM[1]) : ''
+    const filename = fileM ? stripHtml(fileM[1]) : ''
+    items.push({
+      title,
+      filename,
+      size: sizeM ? parseSize(sizeM[1]) : 0,
+      code: codeM[1]
+    })
+  }
+  return items
+}
+
+/** 并发受限地映射 */
+async function mapWithConcurrency<T, R>(items: T[], concurrency: number, fn: (item: T) => Promise<R>): Promise<R[]> {
+  const out: R[] = new Array(items.length)
+  let cursor = 0
+  async function worker() {
+    while (cursor < items.length) {
+      const i = cursor++
+      out[i] = await fn(items[i])
+    }
+  }
+  await Promise.all(Array.from({ length: Math.min(concurrency, items.length) }, () => worker()))
+  return out
+}
+
+async function fetchDetail(it: ListItem, ctx: ProviderContext): Promise<SearchResult | null> {
+  const url = `${cilisousuoBase()}/magnet/${it.code}`
+  const html = await httpText(url, {
+    proxy: ctx.proxy,
+    timeoutMs: Math.min(ctx.timeoutMs, 8000),
+    headers: { accept: 'text/html,application/xhtml+xml', 'user-agent': DEFAULT_UA }
+  })
+
+  const hashM = html.match(/xt=urn:btih:([a-f0-9]{40})/i)
+  if (!hashM) return null
+  const infoHash = hashM[1].toLowerCase()
+  const fileCountM = html.match(/文件数量[^\d]*<dd>\s*(\d+)/)
+  const fileCount = fileCountM ? Number(fileCountM[1]) : null
+  const name = it.title || it.filename || infoHash
+
+  return {
+    infoHash,
+    name,
+    size: it.size,
+    seeders: null,
+    leechers: null,
+    fileCount,
+    category: null,
+    source: 'cilisousuo',
+    sourceLabel: '磁力搜索',
+    sourceId: it.code,
+    sourceUrl: `${cilisousuoBase()}/magnet/${it.code}`,
+    addedAt: null,
+    magnet: buildMagnet(infoHash, name)
+  } satisfies SearchResult
+}
+
+export const cilisousuoProvider: SearchProvider = {
+  id: 'cilisousuo',
+  label: '磁力搜索',
+  homepage: HOMEPAGE,
+
+  async search(query: string, ctx: ProviderContext): Promise<SearchResult[]> {
+    const url = `${cilisousuoBase()}/search?q=${encodeURIComponent(query)}`
+    const html = await httpText(url, {
+      proxy: ctx.proxy,
+      timeoutMs: ctx.timeoutMs,
+      headers: { accept: 'text/html,application/xhtml+xml', 'user-agent': DEFAULT_UA }
+    })
+
+    const items = parseList(html)
+    if (!items.length) return []
+
+    // 列表页不含 infoHash，逐条抓详情页补全磁力链；最多取前 24 条并发抓取
+    const capped = items.slice(0, 24)
+    const results = await mapWithConcurrency(capped, 5, (it) => fetchDetail(it, ctx).catch(() => null))
+    return results.filter((x): x is SearchResult => x !== null)
+  }
+}