feat(adapters): port MoviePilot open site framework (7 spiders, 20 user parsers) + MovieClaw v0.27.0 site definitions + mp-sync tools (W1-3c)

- sites/spiders: Haidan/HDDolby/RousiPro/SunnyPT/TNode/TorrentLeech/Yema (+mTorrent wrapper), specialized SiteAdapter, indirect download resolved at download time - sites/user-parsers: SiteParserBase + all 20 MP SiteSchema parsers (lxml XPath via xpath+xmldom), refreshSiteUserData with schema fallback - sites/definitions: zod SiteDefinition schemas, 23 MovieClaw v0.27.0 (MIT) definitions via scripts/mp-sync/movieclaw-import.mjs, registry getSiteDefinition/listSiteDefinitions, NOTICE - scripts/mp-sync: ledger.json (upstream sha256), check.mjs, build-ledger.mjs; docs/porting/mp-sync-ledger.md - license-and-provenance: add copied-mit kind, re-anchor MovieClaw rows to v0.27.0 Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-27 18:39:27 +08:00
1 parent cf85481d95
commit d9f2af2f3c
@@ -0,0 +1,204 @@
+{
+  "$comment": "W1-3b 手写观众定义(按 MP indexer schema;仅对 testkit mock 验证)。深合并到 MovieClaw v0.27.0 转换结果之上,本文件的值优先,MovieClaw 值只补空。 umt.detail.fields=null:保留 W1-3b rowhead 详情模型(已对 mock 验证),不采用 MovieClaw 默认逐字段选择器。",
+  "id": "audiences",
+  "name": "观众",
+  "domain": "https://audiences.me/",
+  "encoding": "UTF-8",
+  "public": false,
+  "proxy": false,
+  "language": "zh",
+  "result_num": 100,
+  "timeout": 30,
+  "search": {
+    "paths": [
+      {
+        "path": "torrents.php",
+        "type": "all"
+      }
+    ],
+    "params": {
+      "search": "{keyword}",
+      "search_area": 4
+    }
+  },
+  "batch": {
+    "delimiter": " ",
+    "space_replace": "_"
+  },
+  "browse": {
+    "path": "torrents.php?page={page}",
+    "start": 0
+  },
+  "category": {
+    "movie": [
+      {
+        "id": "401",
+        "cat": "Movies",
+        "desc": "电影"
+      }
+    ],
+    "tv": [
+      {
+        "id": "402",
+        "cat": "TV Series",
+        "desc": "剧集"
+      },
+      {
+        "id": "403",
+        "cat": "TV Shows",
+        "desc": "综艺"
+      }
+    ],
+    "music": []
+  },
+  "torrents": {
+    "list": {
+      "selector": "table.torrents > tbody > tr:has(table.torrentname), table.torrents > tr:has(table.torrentname)"
+    },
+    "fields": {
+      "title_default": {
+        "selector": "a[href*=\"details.php?id=\"]"
+      },
+      "title_optional": {
+        "selector": "a[title][href*=\"details.php?id=\"]",
+        "attribute": "title"
+      },
+      "title": {
+        "text": "{% if fields['title_optional'] %}{{ fields['title_optional'] }}{% else %}{{ fields['title_default'] }}{% endif %}"
+      },
+      "details": {
+        "selector": "a[href*=\"details.php?id=\"]",
+        "attribute": "href"
+      },
+      "download": {
+        "selector": "a[href*=\"download.php?id=\"]",
+        "attribute": "href"
+      },
+      "imdbid": {
+        "selector": "a[href*=\"imdb.com/title/tt\"]",
+        "attribute": "href",
+        "filters": [
+          {
+            "name": "re_search",
+            "args": ["tt\\d+", 0]
+          }
+        ]
+      },
+      "date_elapsed": {
+        "selector": "td:nth-child(4) > span"
+      },
+      "date_added": {
+        "selector": "td:nth-child(4) > span",
+        "attribute": "title"
+      },
+      "date": {
+        "text": "{% if fields['date_elapsed'] or fields['date_added'] %}{{ fields['date_added'] if fields['date_added'] else fields['date_elapsed'] }}{% else %}now{% endif %}",
+        "filters": [
+          {
+            "name": "dateparse",
+            "args": "%Y-%m-%d %H:%M:%S"
+          }
+        ]
+      },
+      "size": {
+        "selector": "td:nth-child(5)"
+      },
+      "seeders": {
+        "selector": "td:nth-child(6)"
+      },
+      "leechers": {
+        "selector": "td:nth-child(7)"
+      },
+      "grabs": {
+        "selector": "td:nth-child(8)"
+      },
+      "downloadvolumefactor": {
+        "case": {
+          "img.pro_free": 0,
+          "img.pro_free2up": 0,
+          "img.pro_50pctdown": 0.5,
+          "img.pro_50pctdown2up": 0.5,
+          "img.pro_30pctdown": 0.3,
+          "*": 1
+        }
+      },
+      "uploadvolumefactor": {
+        "case": {
+          "img.pro_50pctdown2up": 2,
+          "img.pro_free2up": 2,
+          "img.pro_2up": 2,
+          "*": 1
+        }
+      },
+      "freedate": {
+        "selector": "img[class^=\"pro_\"]",
+        "attribute": "title",
+        "filters": [
+          {
+            "name": "re_search",
+            "args": ["\\d{4}-\\d{2}-\\d{2} \\d{2}:\\d{2}:\\d{2}", 0]
+          }
+        ]
+      },
+      "description": {
+        "selector": "span.torrent-subtitle-line"
+      },
+      "labels": {
+        "selector": "span.tags"
+      },
+      "category": {
+        "selector": "a[href*=\"?cat=\"]",
+        "attribute": "href",
+        "filters": [
+          {
+            "name": "querystring",
+            "args": "cat"
+          }
+        ]
+      }
+    }
+  },
+  "umt": {
+    "version": "[email protected]",
+    "kind": "nexusphp",
+    "displayName": "观众",
+    "timezone": "+08:00",
+    "rateLimit": {
+      "minIntervalMs": 3000,
+      "burst": 1
+    },
+    "capabilities": {
+      "list": true,
+      "search": true,
+      "detail": true,
+      "download": true,
+      "rss": true
+    },
+    "categoryNames": {
+      "401": "电影",
+      "402": "剧集",
+      "403": "综艺",
+      "406": "纪录片"
+    },
+    "loginSuccess": "a[href*=\"logout\"]",
+    "emptyMarkers": ["没有种子", "没有结果", "Nothing found"],
+    "pagination": "a[href*=\"page=\"]",
+    "detailPath": "details.php",
+    "downloadPath": "download.php",
+    "detail": {
+      "path": "details.php",
+      "title": "h1#top",
+      "rowLabel": "td.rowhead",
+      "rowValue": "td.rowfollow",
+      "labels": {
+        "subtitle": ["副标题", "Small Description"],
+        "basicInfo": ["基本信息", "Basic Info"],
+        "peers": ["同伴", "Peers"]
+      },
+      "description": "#kdescr",
+      "headerScope": "h1#top",
+      "notFoundMarkers": ["没有该ID的种子", "No torrent with ID"],
+      "fields": null
+    }
+  }
+}
@@ -0,0 +1,16 @@
+{
+  "$comment": "CHDBits 分类名取自 MovieClaw chdbits.yaml 注释(TV Series / TV Shows / Sports / Music Videos / HQ Audio)。",
+  "umt": {
+    "categoryNames": {
+      "401": "电影",
+      "402": "剧集",
+      "403": "综艺",
+      "404": "纪录片",
+      "405": "动漫",
+      "406": "MV",
+      "407": "体育",
+      "408": "HQ Audio",
+      "409": "其他"
+    }
+  }
+}
@@ -0,0 +1,17 @@
+{
+  "$comment": "HHClub 分类名取自 MovieClaw hhanclub.yaml 注释(导航栏分类表);REQ §4.5 要求综艺/体育/短剧保留原始分类。",
+  "umt": {
+    "categoryNames": {
+      "401": "电影",
+      "402": "电视剧",
+      "403": "综艺",
+      "404": "纪录片",
+      "405": "动漫",
+      "407": "体育",
+      "409": "其他",
+      "410": "音乐",
+      "411": "MV",
+      "412": "短剧"
+    }
+  }
+}
@@ -0,0 +1,26 @@
+{
+  "$comment": "W1-3b 馒头定义(mteam/definition.ts 原值,按 MP mtorrent.py)。本文件优先,MovieClaw 值只补空;categoryNames 为本项目编制的粗分类名。",
+  "name": "馒头",
+  "timeout": 15,
+  "umt": {
+    "version": "[email protected]",
+    "displayName": "馒头",
+    "categoryNames": {
+      "401": "电影",
+      "419": "电影",
+      "420": "电影",
+      "421": "电影",
+      "439": "电影",
+      "402": "剧集",
+      "403": "综艺",
+      "435": "剧集",
+      "438": "剧集",
+      "404": "纪录",
+      "405": "动画",
+      "407": "体育",
+      "406": "音乐",
+      "408": "音乐",
+      "434": "音乐"
+    }
+  }
+}