850 lines
34 KiB
JavaScript
850 lines
34 KiB
JavaScript
// NexusPHP 框架共用脚本：所有 NexusPHP 站点用同一份，站点差异由各自的 definition 选择器提供。
|
||
// mediabot2 经 Jint 沙箱执行，注入：content, url, http, html, json, utils
|
||
// 入口函数（与注册中心 7 个标准能力模板一一对应，能力名 = 函数名）：
|
||
// search(ctx) —— 搜索种子（能力: search）
|
||
// detail(ctx) —— 种子详情（能力: detail）
|
||
// index(ctx) —— 用户统计（能力: index，解析首页信息块）
|
||
// userdetail(ctx) —— 用户详情（能力: userdetail，解析 userdetails.php）
|
||
// signIn(ctx) —— 签到（能力: signIn）
|
||
// download(ctx) —— 下载 .torrent 种子文件字节（能力: download）
|
||
// sync(ctx) —— 站点信息同步（mediabot2 Host 调用：首页统计 + 账号资料）
|
||
// parse(content, url) —— 解析用户统计页（校验沙箱 /validate 使用，http 走 mock）
|
||
// rss(ctx) —— RSS/Atom 刷流（供 RssDiscoveryJob，非标准能力）
|
||
|
||
// ==================== 运行环境兼容层 ====================
|
||
// mediabot2 宿主注入: http / html / utils / json；
|
||
// 注册中心 Jint SDK 注入: parseHtml / httpClient / console / __wrapNode。
|
||
// 以下兼容代码仅在注册中心沙箱中生效（mediabot2 环境已有同名对象，条件不成立时零影响）。
|
||
if (typeof http === 'undefined') {
|
||
var http = {
|
||
get: function(url) { return httpClient.getString(url); },
|
||
// SDK 未提供 POST，注册中心沙箱中回退为 GET（mock 环境仅用于样本解析验证）
|
||
post: function(url) { return httpClient.getString(url); },
|
||
getBytes: function(url) { return httpClient.getBytes(url); }
|
||
};
|
||
}
|
||
if (typeof html === 'undefined') {
|
||
var html = {
|
||
parse: function(h) {
|
||
var doc = parseHtml(h);
|
||
var oQS = doc.querySelector, oQSA = doc.querySelectorAll;
|
||
var oGEI = doc.getElementById, oGET = doc.getElementsByTagName;
|
||
doc.querySelector = function(sel) { return __adapt(oQS.call(doc, sel)); };
|
||
doc.querySelectorAll = function(sel) {
|
||
var r = oQSA.call(doc, sel) || [], out = [];
|
||
for (var i = 0; i < r.length; i++) out.push(__adapt(r[i]));
|
||
return out;
|
||
};
|
||
doc.getElementById = function(id) { return __adapt(oGEI.call(doc, id)); };
|
||
doc.getElementsByTagName = function(tag) {
|
||
var r = oGET.call(doc, tag) || [], out = [];
|
||
for (var j = 0; j < r.length; j++) out.push(__adapt(r[j]));
|
||
return out;
|
||
};
|
||
doc.queryList = doc.querySelectorAll;
|
||
return doc;
|
||
}
|
||
};
|
||
}
|
||
|
||
// 为 SDK wrapNode 补齐 mediabot2 DOM 约定: attr / queryList / nextElementSibling
|
||
function __adapt(n) {
|
||
if (!n || typeof n !== 'object' || n.__adapted) return n;
|
||
try {
|
||
n.__adapted = true;
|
||
if (typeof n.attr !== 'function') n.attr = function(name) { return n.getAttribute(name); };
|
||
if (typeof n.queryList !== 'function') n.queryList = function(sel) { return n.querySelectorAll(sel) || []; };
|
||
if (typeof n.nextElementSibling === 'undefined' && n._node) {
|
||
try { n.nextElementSibling = __adapt(__wrapNode(n._node.NextElementSibling)); }
|
||
catch (e) { n.nextElementSibling = null; }
|
||
}
|
||
// 递归: 子查询返回的新包装对象也要经过 __adapt
|
||
if (typeof n.querySelector === 'function' && !n.__qsw) {
|
||
n.__qsw = true;
|
||
var oqs = n.querySelector, oqsa = n.querySelectorAll;
|
||
n.querySelector = function(sel) { return __adapt(oqs.call(n, sel)); };
|
||
n.querySelectorAll = function(sel) {
|
||
var r = oqsa.call(n, sel) || [], out = [];
|
||
for (var i = 0; i < r.length; i++) out.push(__adapt(r[i]));
|
||
return out;
|
||
};
|
||
}
|
||
} catch (e) {}
|
||
return n;
|
||
}
|
||
|
||
// 是否运行在注册中心沙箱（SDK 才有 __wrapNode）
|
||
var __inRegistrySandbox = (typeof __wrapNode === 'function');
|
||
|
||
// 注册中心 ParseInputArgs 把 testInput 原始 JSON 串按位置传参; mediabot2 直接传对象。
|
||
// 入口函数统一经 __asCtx 归一化，两种调用方式都兼容。
|
||
function __asCtx(arg) {
|
||
if (typeof arg === 'string' && arg.length > 0) {
|
||
try {
|
||
var p = JSON.parse(arg);
|
||
if (p && typeof p === 'object') return p;
|
||
} catch (e) {}
|
||
return {};
|
||
}
|
||
return (arg && typeof arg === 'object') ? arg : {};
|
||
}
|
||
|
||
// 取 baseUrl；注册中心沙箱中为空时用占位地址（mock http 会回退返回样本 HTML，不影响解析链路）
|
||
function __resolveBase(ctx) {
|
||
var b = String(ctx && ctx.baseUrl ? ctx.baseUrl : '').replace(/\/+$/, '');
|
||
if (!b && __inRegistrySandbox) b = 'http://mock.local';
|
||
return b;
|
||
}
|
||
|
||
function unitToBytes(unit) {
|
||
if (!unit) return 1;
|
||
switch (String(unit).toUpperCase()) {
|
||
case 'PB': return 1024 * 1024 * 1024 * 1024 * 1024;
|
||
case 'TB': return 1024 * 1024 * 1024 * 1024;
|
||
case 'GB': return 1024 * 1024 * 1024;
|
||
case 'MB': return 1024 * 1024;
|
||
case 'KB': return 1024;
|
||
default: return 1;
|
||
}
|
||
}
|
||
|
||
// 标签后直接匹配数值（兼容 "上传量: 1.23 TB" / "上传量 1.23 TB" / 表格布局）
|
||
function parseSize(text, labelPattern) {
|
||
var re = new RegExp(labelPattern + '\\s*[:：]?\\s*([\\d.,]+)\\s*(TB|GB|MB|KB|PB|B)?', 'i');
|
||
var m = text.match(re);
|
||
if (!m) return null;
|
||
var num = parseFloat(m[1].replace(/,/g, ''));
|
||
if (isNaN(num)) return null;
|
||
return Math.round(num * unitToBytes(m[2]));
|
||
}
|
||
|
||
function parseDecimal(text, labelPattern) {
|
||
var re = new RegExp(labelPattern + '\\s*[:：]?\\s*([\\d.,]+)', 'i');
|
||
var m = text.match(re);
|
||
if (!m) return null;
|
||
var v = parseFloat(m[1].replace(/,/g, ''));
|
||
return isNaN(v) ? null : v;
|
||
}
|
||
|
||
function parseInteger(text, labelPattern) {
|
||
var re = new RegExp(labelPattern + '\\s*[:：]?\\s*([\\d,]+)', 'i');
|
||
var m = text.match(re);
|
||
if (!m) return null;
|
||
var v = parseInt(m[1].replace(/,/g, ''), 10);
|
||
return isNaN(v) ? null : v;
|
||
}
|
||
|
||
// 从信息块解析 uid / 用户名（userdetails 链接）
|
||
function parseIdentity(block) {
|
||
var uid = null;
|
||
var userName = null;
|
||
var link = block ? block.querySelector('a[href*="userdetails"]') : null;
|
||
if (link) {
|
||
var href = link.attr('href') || '';
|
||
var um = href.match(/[?&]id=(\d+)/i);
|
||
if (um) uid = um[1];
|
||
var t = link.text();
|
||
if (t) userName = t;
|
||
}
|
||
return { uid: uid, userName: userName };
|
||
}
|
||
|
||
// 解析首页统计：上传/下载/魔力/分享率/做种/下载活动/H&R
|
||
function parseIndexStats(indexHtml) {
|
||
var doc = html.parse(indexHtml);
|
||
var block = doc.querySelector('#info_block');
|
||
var text = block ? block.text() : indexHtml;
|
||
|
||
var up = parseSize(text, '(?:上传量?|上傳量?)');
|
||
var down = parseSize(text, '(?:下载量?|下載量?)');
|
||
if (up === null && down === null) return null;
|
||
|
||
var bonus = parseDecimal(text, '魔力值(?:\\[\\s*使用\\s*\\])?');
|
||
if (bonus === null) {
|
||
// PTTIME: 魔力值(77.45魔力/小时)[使用&说明]: 799327.2
|
||
var bm = text.match(/魔力值.*?[:：]\s*([\d.,]+)/i);
|
||
if (bm) bonus = parseFloat(bm[1].replace(/,/g, ''));
|
||
}
|
||
var ratio = parseDecimal(text, '(?:分享率|分享比率)');
|
||
|
||
var seeding = null, leeching = null;
|
||
var act = text.match(/当前活动\s*[:：]\s*(\d+)\s+(\d+)/i);
|
||
if (!act) act = text.match(/活动\s*[:：]\s*[^\d]*(\d+)[^\d]*(\d+)/i); // PTTIME: "活动: ⬆16 ⬇2"
|
||
if (act) {
|
||
seeding = parseInt(act[1], 10);
|
||
leeching = parseInt(act[2], 10);
|
||
} else {
|
||
seeding = parseInteger(text, '(?:做种|做種)');
|
||
}
|
||
|
||
var hr = null;
|
||
var hrM = text.match(/H&R\s*[:：]\s*\[\s*\d+\s*\/\s*(\d+)\s*\/\s*\d+\s*\]/i);
|
||
if (hrM) hr = parseInt(hrM[1], 10);
|
||
|
||
var identity = parseIdentity(block || doc);
|
||
return {
|
||
uploaded: up,
|
||
downloaded: down,
|
||
bonus: bonus,
|
||
ratio: ratio,
|
||
seeding: seeding,
|
||
leeching: leeching,
|
||
hr: hr,
|
||
uid: identity.uid,
|
||
userName: identity.userName
|
||
};
|
||
}
|
||
|
||
// 解析 userdetails.php 账号资料：邮箱/等级/注册日期/最近动向/做种积分/封存状态 + uid/用户名
|
||
// 注意：userdetails 的字段在 #info_block 之外的资料表格中，故用整页 body 文本解析
|
||
function parseUserDetails(userHtml) {
|
||
var doc = html.parse(userHtml);
|
||
var body = doc.querySelector('body') || doc;
|
||
var text = body.text() || userHtml;
|
||
|
||
var email = null;
|
||
// DOM 优先：从 <a href="mailto:"> 提取最准确
|
||
var mailLink = doc.querySelector('a[href*="mailto:"]');
|
||
if (mailLink) {
|
||
var href = mailLink.attr('href') || '';
|
||
var mm = href.match(/mailto:([\w.+-]+@[\w.-]+\.\w+)/i);
|
||
if (mm) email = mm[1];
|
||
else email = mailLink.text();
|
||
}
|
||
// 退路：正则从 body text 搜
|
||
if (!email) {
|
||
var em = text.match(/(?:邮箱|Email|E-mail)\s*[:：]?\s*([\w.+-]+@[\w.-]+\.\w+)/i);
|
||
if (em) email = em[1];
|
||
}
|
||
|
||
// 等级：DOM 解析，遍历每行 tr，找 label td 文本为 等级/级别/Class 的行，取其右边 td 的内容
|
||
var classLevel = null;
|
||
var rows = doc.queryList('tr');
|
||
for (var i = 0; i < rows.length && !classLevel; i++) {
|
||
var cells = rows[i].queryList('td');
|
||
if (cells.length >= 2) {
|
||
var label = cells[0].text();
|
||
if (label === '等级' || label === '级别' || label === 'Class' || label === '用户等级') {
|
||
var img = cells[1].querySelector('img');
|
||
var b = cells[1].querySelector('b');
|
||
if (img) {
|
||
classLevel = img.attr('alt');
|
||
} else if (b) {
|
||
classLevel = b.text(); // PTTIME: 等级在 <b class="fcr"> 内
|
||
} else {
|
||
classLevel = cells[1].text();
|
||
}
|
||
// 去掉前导括号前缀，如 "(幼儿园)User" → "User"
|
||
if (classLevel) {
|
||
classLevel = classLevel.replace(/^\s*\([^)]*\)\s*/, '');
|
||
}
|
||
}
|
||
}
|
||
}
|
||
|
||
var registerDate = null;
|
||
var regM = text.match(/(?:注册日期|加入日期|Registered)\s*[:：]?\s*(\d{4}-\d{2}-\d{2})/i);
|
||
if (regM) registerDate = regM[1];
|
||
|
||
// 最近动向：NexusPHP 常为 "xx 天/时 前" 或绝对时间，原样截取供上层判断
|
||
var lastActivityAt = null;
|
||
var laM = text.match(/(?:最近动向|最近活动|最后活动|最近上站|Last\s*(?:Seen|Active|Activity))\s*[:：]?\s*([^\n\r]{1,40})/i);
|
||
if (laM) lastActivityAt = laM[1].trim();
|
||
|
||
// 做种积分
|
||
var seedingScore = null;
|
||
var ssM = text.match(/(?:做种积分|做種積分|Seeding\s*(?:Bonus|Points?))\s*[:：]?\s*([\d.,]+)/i);
|
||
if (ssM) {
|
||
var sv = parseFloat(ssM[1].replace(/,/g, ''));
|
||
if (!isNaN(sv)) seedingScore = sv;
|
||
}
|
||
|
||
// 封存/封禁检测
|
||
var isSealed = /封存|帳號已?被封|账号已?被封/.test(text);
|
||
|
||
var identity = parseIdentity(doc.querySelector('#info_block') || doc);
|
||
return {
|
||
uid: identity.uid,
|
||
userName: identity.userName,
|
||
email: email,
|
||
classLevel: classLevel,
|
||
registerDate: registerDate,
|
||
lastActivityAt: lastActivityAt,
|
||
seedingScore: seedingScore,
|
||
isSealed: isSealed
|
||
};
|
||
}
|
||
|
||
// ================= 能力: index（用户统计）=================
|
||
// ctx: { baseUrl, ... }。解析首页信息块，返回当前登录用户统计。
|
||
function index(input) {
|
||
var ctx = __asCtx(input);
|
||
var base = __resolveBase(ctx);
|
||
if (!base) return { error: 'no_base_url' };
|
||
|
||
var indexHtml = null;
|
||
var paths = ['/index.php', '/'];
|
||
for (var i = 0; i < paths.length; i++) {
|
||
try {
|
||
indexHtml = http.get(base + paths[i]);
|
||
} catch (e) {
|
||
indexHtml = null;
|
||
}
|
||
if (indexHtml && indexHtml.length > 0) break;
|
||
}
|
||
if (!indexHtml) return { error: 'index_fetch_failed' };
|
||
|
||
var stats = parseIndexStats(indexHtml);
|
||
if (!stats) return { error: 'index_no_match', html: indexHtml };
|
||
return stats;
|
||
}
|
||
|
||
// ================= 能力: userdetail（用户详情）=================
|
||
// ctx: { baseUrl, userId (或 uid) }
|
||
function userdetail(input) {
|
||
var ctx = __asCtx(input);
|
||
var base = __resolveBase(ctx);
|
||
if (!base) return { error: 'no_base_url' };
|
||
var uid = (ctx && (ctx.userId || ctx.uid)) || '';
|
||
if (!uid) return { error: 'no_user_id' };
|
||
|
||
var userHtml;
|
||
try {
|
||
userHtml = http.get(base + '/userdetails.php?id=' + encodeURIComponent(uid));
|
||
} catch (e) {
|
||
return { error: String((e && e.message) || e) };
|
||
}
|
||
if (!userHtml) return { error: 'userdetails_fetch_failed' };
|
||
|
||
var profile = parseUserDetails(userHtml);
|
||
if (!profile) return { error: 'userdetails_no_match', html: userHtml };
|
||
return {
|
||
uid: profile.uid || uid,
|
||
userName: profile.userName,
|
||
userClass: profile.classLevel,
|
||
registeredAt: profile.registerDate,
|
||
lastActivityAt: profile.lastActivityAt,
|
||
seedingScore: profile.seedingScore,
|
||
profileEmail: profile.email,
|
||
isSealed: profile.isSealed
|
||
};
|
||
}
|
||
|
||
// ================= 站点信息同步（mediabot2 Host 调用）=================
|
||
// ctx: { siteCode, siteId, baseUrl, uid, userName, utcNow, frameworkVersion }
|
||
// Success: returns stats object. Failure: returns { error, html?, userHtml? } carrying raw HTML
|
||
// so Host can report it verbatim to Registry. NEVER return null; always include `error`.
|
||
function sync(input) {
|
||
var ctx = __asCtx(input);
|
||
var base = __resolveBase(ctx);
|
||
if (!base) return { error: 'no_base_url' };
|
||
|
||
// 1. fetch index page (index.php first, root fallback)
|
||
var indexHtml = null;
|
||
var paths = ['/index.php', '/'];
|
||
for (var i = 0; i < paths.length; i++) {
|
||
try {
|
||
indexHtml = http.get(base + paths[i]);
|
||
} catch (e) {
|
||
indexHtml = null;
|
||
}
|
||
if (indexHtml && indexHtml.length > 0) break;
|
||
}
|
||
if (!indexHtml) return { error: 'index_fetch_failed' };
|
||
|
||
var indexStats = parseIndexStats(indexHtml);
|
||
if (!indexStats) return { error: 'index_no_match', html: indexHtml };
|
||
|
||
// 2. if uid known, fetch userdetails.php?id={uid} to parse profile
|
||
var uid = indexStats.uid || (ctx && ctx.uid) || null;
|
||
var profile = null;
|
||
var userHtml = null;
|
||
if (uid) {
|
||
try {
|
||
userHtml = http.get(base + '/userdetails.php?id=' + encodeURIComponent(uid));
|
||
if (userHtml && userHtml.length > 0) {
|
||
profile = parseUserDetails(userHtml);
|
||
// content fetched but no profile fields parsed (CAPTCHA / login wall): treat as failure
|
||
if (profile && profile.email === null && profile.classLevel === null && profile.registerDate === null) {
|
||
profile = null;
|
||
}
|
||
}
|
||
} catch (e) {
|
||
profile = null;
|
||
}
|
||
}
|
||
// userdetails fetch/parse failed: stats available, attach userHtml for Host reporting
|
||
if (uid && !profile) {
|
||
return {
|
||
uploaded: indexStats.uploaded,
|
||
downloaded: indexStats.downloaded,
|
||
bonus: indexStats.bonus,
|
||
ratio: indexStats.ratio,
|
||
seeding: indexStats.seeding,
|
||
leeching: indexStats.leeching,
|
||
hr: indexStats.hr,
|
||
uid: uid,
|
||
userName: indexStats.userName || (ctx && ctx.userName) || null,
|
||
email: null,
|
||
classLevel: null,
|
||
registerDate: null,
|
||
error: 'userdetails_no_match',
|
||
html: indexHtml,
|
||
userHtml: userHtml
|
||
};
|
||
}
|
||
|
||
return {
|
||
uploaded: indexStats.uploaded,
|
||
downloaded: indexStats.downloaded,
|
||
bonus: indexStats.bonus,
|
||
ratio: indexStats.ratio,
|
||
seeding: indexStats.seeding,
|
||
leeching: indexStats.leeching,
|
||
hr: indexStats.hr,
|
||
uid: uid,
|
||
userName: (profile && profile.userName) || indexStats.userName || (ctx && ctx.userName) || null,
|
||
email: (profile && profile.email) || null,
|
||
classLevel: (profile && profile.classLevel) || null,
|
||
registerDate: (profile && profile.registerDate) || null
|
||
};
|
||
}
|
||
|
||
function parse(content, url) {
|
||
var doc = html.parse(content);
|
||
return {
|
||
uploaded: doc.queryText('#seedup a'),
|
||
downloaded: doc.queryText('#seeddown a'),
|
||
bonus: doc.queryText('#seedbonus')
|
||
};
|
||
}
|
||
|
||
// ================= 能力: signIn（签到）=================
|
||
// 签到（signInPath 默认 /attendance.php，可在站点 definition 中覆盖，如 BTSCHOOL: /index.php?action=addbonus）
|
||
// 成功时若响应页含 #info_block (含魔力值) 则一并返回 bonus 供服务端更新 Site.Statistics.Bonus;
|
||
// 站点页面不含 / 解析不到时缺省不返回该字段, 服务端不处理 (report §四 签到奖励自动更新).
|
||
function signIn(input) {
|
||
var ctx = __asCtx(input);
|
||
var signPath = ctx.signInPath || '/attendance.php';
|
||
var res = http.post(signPath);
|
||
if (res && res.indexOf('成功') >= 0) {
|
||
var out = { signed: true, message: '签到成功' };
|
||
try {
|
||
var stats = parseIndexStats(res);
|
||
if (stats && stats.bonus !== null && stats.bonus !== undefined && !isNaN(stats.bonus)) {
|
||
out.bonus = stats.bonus;
|
||
}
|
||
} catch (e) {
|
||
// bonus 解析失败不影响签到成功结果
|
||
}
|
||
return out;
|
||
}
|
||
if (res && (res.indexOf('已签到') >= 0 || res.indexOf('已经签到') >= 0 || res.indexOf('已經簽到') >= 0 || res.indexOf('重复签到') >= 0 || res.indexOf('重複簽到') >= 0)) {
|
||
return { signed: true, message: '今日已签到' };
|
||
}
|
||
return { signed: false, message: (res && res.length > 0) ? '未识别签到结果' : '签到请求失败' };
|
||
}
|
||
|
||
// 解析裸大小文本 "1.23 GB" / "986.5 MB" / "1.2 TB" → 字节数
|
||
function parseSizeText(text) {
|
||
if (!text) return 0;
|
||
var m = String(text).match(/([\d.,]+)\s*(TB|GB|MB|KB|PB|B)\b/i);
|
||
if (!m) return 0;
|
||
var num = parseFloat(m[1].replace(/,/g, ''));
|
||
if (isNaN(num)) return 0;
|
||
return Math.round(num * unitToBytes(m[2]));
|
||
}
|
||
|
||
// 提取非负整数（忽略逗号与装饰性字符）
|
||
function toInt(text) {
|
||
if (text === null || text === undefined) return 0;
|
||
var m = String(text).match(/[\d,]+/);
|
||
if (!m) return 0;
|
||
var v = parseInt(m[0].replace(/,/g, ''), 10);
|
||
return isNaN(v) ? 0 : v;
|
||
}
|
||
|
||
// 从标题里按正则提取首个媒体标签 (分辨率/来源/编码)，供订阅偏好过滤与质量评分
|
||
function parseMediaTag(title, re) {
|
||
if (!title) return null;
|
||
var m = String(title).match(re);
|
||
if (!m) return null;
|
||
var v = m[1] || m[0];
|
||
return v ? v.toUpperCase() : null;
|
||
}
|
||
|
||
// ================= 能力: search（搜索种子）=================
|
||
// ctx: { siteCode, baseUrl, keyword, page(=1), searchPath(可选, 默认 /torrents.php) }
|
||
// 走 http（已注入 Cookie 的 HttpClient），解析 table.torrents 种子行。
|
||
function search(input) {
|
||
var ctx = __asCtx(input);
|
||
var keyword = ctx && ctx.keyword ? String(ctx.keyword) : '';
|
||
if (!keyword) return [];
|
||
var base = __resolveBase(ctx);
|
||
if (!base) return [];
|
||
var page = ctx.page ? parseInt(ctx.page, 10) : 1;
|
||
if (isNaN(page) || page < 1) page = 1;
|
||
var searchPath = ctx.searchPath || '/torrents.php';
|
||
|
||
// 首页不拼 &page=N：NexusPHP 分页分 0 基/1 基两种，部分站点(如 carpt)在
|
||
// ?search=X&page=1 时实际请求的是"第 2 页"，而结果往往只有一页 → 返回空表(仅搜索框)。
|
||
// 统一用裸 ?search=X 取首页，对两种索引方案都正确返回第 1 页。
|
||
var query = '?search=' + encodeURIComponent(keyword);
|
||
var url = (page > 1)
|
||
? (base + searchPath + query + '&page=' + page)
|
||
: (base + searchPath + query);
|
||
var htmlText = http.get(url);
|
||
if (!htmlText) return [];
|
||
var doc = html.parse(htmlText);
|
||
if (!doc) return [];
|
||
|
||
// ---- 诊断: 校验返回结构, 便于排查搜索空结果 (源码原始 html 特征) ----
|
||
// 注意: 不能用 "Powered by NexusPHP" 或站名判断登录墙 —— 这些字串每个 NexusPHP 站页面都有.
|
||
// 登录墙的可靠特征是: 存在登录表单 (带 password input / action 指向 login 页).
|
||
var htmlTag = null;
|
||
var titleTag = (htmlText.match(/<title>([^<]*)<\/title>/i) || [])[1] || '';
|
||
var tableCount = (htmlText.match(/<table[^>]*class="torrents"[^>]*>/i) || []).length;
|
||
var hasLoginForm = /<input[^>]*type="Password"[^>]*>/i.test(htmlText)
|
||
|| /<form[^>]*action="[^"]*login[^"]*"[^>]*>/i.test(htmlText)
|
||
|| /name="(?:username|loginusername|userpasswd)"[^>]*>/i.test(htmlText)
|
||
|| /登录到|login in|请登陆|请登录/i.test(titleTag);
|
||
// 采集页面所有的 table class + 种子详情链接数, 用于判断选择器是否匹配.
|
||
var tableClasses = [];
|
||
var tcm = htmlText.match(/<table[^>]*class="([^"]*)"[^>]*>/gi);
|
||
if (tcm) for (var tc = 0; tc < tcm.length; tc++) {
|
||
var cls = (tcm[tc].match(/class="([^"]*)"/i) || [])[1] || '';
|
||
if (cls && tableClasses.indexOf(cls) < 0) tableClasses.push(cls);
|
||
}
|
||
var detailLinkCount = (htmlText.match(/href="[^"]*details\.php[^"]*"/gi) || []).length;
|
||
if (hasLoginForm) {
|
||
htmlTag = 'loginwall';
|
||
} else if (detailLinkCount > 0) {
|
||
htmlTag = 'has_details_' + detailLinkCount;
|
||
} else if (titleTag && /无.*种子|没有|未找到|No|结果/i.test(titleTag)) {
|
||
htmlTag = 'noresult:' + titleTag;
|
||
} else if (tableCount === 0) {
|
||
htmlTag = 'no_torrents_table';
|
||
} else {
|
||
htmlTag = 'ok';
|
||
}
|
||
// 候选选择器命中数: 定位 carpt 种子列表的真实容器 (标准 nexusphp 是 table.torrents, 自定义皮站常不同)
|
||
var selCands = ['table.torrents tr', 'table[class*="torrent"] tr', '#torrents tr', 'tr.rowfollow', '.torrent_name', '.torrenttitle', 'td.rowfollow', 'table tr'];
|
||
var selHits = [];
|
||
for (var si = 0; si < selCands.length; si++) {
|
||
try { selHits.push(selCands[si] + '=' + doc.queryList(selCands[si]).length); } catch (e) { selHits.push(selCands[si] + '=ERR'); }
|
||
}
|
||
// 原始响应 HTML 关键片段: 后端会将其落盘为快照文件 (search-snapshots/) 便于调试, 此处保留完整内容.
|
||
// ---- 诊断结束 ----
|
||
|
||
// 取所有表格行, 再由 td.rowfollow + details 链接 过滤出真实种子行.
|
||
// 注意: 不能用 table.torrents —— 不少自定义皮站 (如 carpt) 种子列表父表 class 不是 torrents.
|
||
// 侧边搜索/筛选区同样含 td.rowfollow, 故必须额外要求含 details* 链接才算种子行.
|
||
var rows = doc.queryList('table tr');
|
||
var results = [];
|
||
for (var i = 0; i < rows.length; i++) {
|
||
var row = rows[i];
|
||
// 仅取直接子 td：carpt 等站点在标题单元格内嵌套了 torrentname 子表，
|
||
// 用 'td' 会把后代 td(embedded) 一并计入，导致列索引错位、体积/做种数全变 0.
|
||
var tds = row.queryList(':scope > td');
|
||
// 种子行判定：含 td.rowfollow(排除搜索/筛选/表头) 且拥有多个直接子 td(标准约 9 列)。
|
||
// 仅 1 个直接子 td 的 tr 多为置顶/推荐/嵌套区块的重复条目(数字无法解析)，直接跳过.
|
||
if (tds.length < 6) continue;
|
||
if (row.queryList('td.rowfollow').length === 0) continue;
|
||
var titleEl = row.querySelector('a[href*="details"]');
|
||
if (!titleEl) continue;
|
||
var title = titleEl.text();
|
||
if (!title) continue;
|
||
|
||
var href = titleEl.attr('href');
|
||
var detailUrl = null;
|
||
if (href) {
|
||
detailUrl = (href.indexOf('http') === 0) ? href : (base + (href[0] === '/' ? href : '/' + href));
|
||
}
|
||
|
||
// 行内下载链接: 下载图标指向 download.php?… (passkey 站点已带 passkey/uid 等完整参数, 如 pttime;
|
||
// 仅需 Cookie 的站点如 carpt 则不带额外参数). 下载时直接复用该链接, 无需再拼 passkey.
|
||
var downloadUrl = null;
|
||
var dlLink = row.querySelector('a[href*="download.php"]');
|
||
if (dlLink) {
|
||
var dhref = dlLink.attr('href') || '';
|
||
if (dhref) {
|
||
downloadUrl = (dhref.indexOf('http') === 0) ? dhref : (base + (dhref[0] === '/' ? dhref : '/' + dhref));
|
||
}
|
||
}
|
||
|
||
var size = 0, seeders = 0, leechers = 0, completed = 0;
|
||
if (tds.length > 7) {
|
||
size = parseSizeText(tds[4] ? tds[4].text() : '');
|
||
seeders = toInt(tds[5] ? tds[5].text() : '');
|
||
leechers = toInt(tds[6] ? tds[6].text() : '');
|
||
completed = toInt(tds[7] ? tds[7].text() : '');
|
||
}
|
||
|
||
var yearM = title.match(/\b(19|20)\d{2}\b/);
|
||
var year = yearM ? parseInt(yearM[0], 10) : null;
|
||
|
||
// 促销/免费/H&R 标记：图标位于标题单元格内 (img.pro_free / img.pro_50pctdown / img.hitandrun / alt)
|
||
var isFree = false, isHr = false, freeDeadline = null, promo = null;
|
||
var imgs = row.queryList('img');
|
||
for (var pi = 0; pi < imgs.length; pi++) {
|
||
var pcls = imgs[pi].attr('class') || '';
|
||
var palt = imgs[pi].attr('alt') || '';
|
||
if (/pro_free|pro_100pctdown/i.test(pcls) || /\bfree\b|100%/i.test(palt)) { isFree = true; promo = 'free'; }
|
||
else if (/pro_50pctdown|pro_free2up/i.test(pcls) || /50%|2x/i.test(palt)) { promo = 'half'; }
|
||
if (/hitandrun/i.test(pcls) || /\bH&?R\b/i.test(palt)) isHr = true;
|
||
}
|
||
// 免费/促销截止时间 + 发布时间：行内真实存在的 <span title="2027-03-02 10:33:29">（"剩余时间："后），
|
||
// 取最大者(未来截止>过去添加时间)为免费截止，取最小者(最早)为种子发布时间.
|
||
var spans = row.queryList('span[title]');
|
||
var maxDt = null, minDt = null;
|
||
for (var si = 0; si < spans.length; si++) {
|
||
var sttl = spans[si].attr('title') || '';
|
||
var sdm = sttl.match(/^(\d{4}-\d{2}-\d{2}[ T]\d{2}:\d{2}:\d{2})/);
|
||
if (!sdm) continue;
|
||
if (maxDt === null || sdm[1] > maxDt) maxDt = sdm[1];
|
||
if (minDt === null || sdm[1] < minDt) minDt = sdm[1];
|
||
}
|
||
if (maxDt) freeDeadline = maxDt.replace(' ', 'T');
|
||
var uploadAt = minDt ? minDt.replace(' ', 'T') : null;
|
||
|
||
// 分类：行首分类图标 (class 以 c_ 开头, alt/title=音乐/Music 等)，取首个匹配
|
||
var category = null;
|
||
var cimgs = row.queryList('td.rowfollow img');
|
||
for (var ci = 0; ci < cimgs.length; ci++) {
|
||
var ccls = cimgs[ci].attr('class') || '';
|
||
if (/^c_/i.test(ccls)) {
|
||
category = cimgs[ci].attr('title') || cimgs[ci].attr('alt') || null;
|
||
break;
|
||
}
|
||
}
|
||
|
||
// 分辨率/来源/编码：从标题文本解析，供订阅偏好过滤 (PreferredResolution/Source/Encoding) 与质量评分
|
||
var resolution = parseMediaTag(title, /\b(\d{3,4}[pi])\b/i);
|
||
var source = parseMediaTag(title, /\b(WEB-DL|WEBRip|WEB|HDTV|Blu-?Ray|DVD|Remux|BDrip|BDRip|HDrip|DVDRip)\b/i);
|
||
var encoding = parseMediaTag(title, /\b(x264|x265|H\.?265|H\.?264|AVC|HEVC|VC-?1|XviD|AV1)\b/i);
|
||
|
||
results.push({
|
||
title: title,
|
||
detailUrl: detailUrl,
|
||
downloadUrl: downloadUrl,
|
||
size: size,
|
||
seeders: seeders,
|
||
leechers: leechers,
|
||
completed: completed,
|
||
year: year,
|
||
resolution: resolution,
|
||
source: source,
|
||
encoding: encoding,
|
||
category: category,
|
||
isFree: isFree,
|
||
isHr: isHr,
|
||
uploadAt: uploadAt,
|
||
freeDeadline: freeDeadline,
|
||
promo: promo
|
||
});
|
||
}
|
||
if (results.length === 0) {
|
||
// 无结果时附加诊断信息: 供后端日志区分登录墙/无结果页/选择器不匹配, 前端会按空 title 过滤
|
||
results.push({
|
||
title: '',
|
||
__diag: htmlTag || 'unknown',
|
||
__url: url,
|
||
__len: htmlText ? htmlText.length : 0,
|
||
__tables: tableClasses.join(','),
|
||
__details: detailLinkCount,
|
||
__sel: selHits.join(' | '),
|
||
__html: htmlText || ''
|
||
});
|
||
}
|
||
return results;
|
||
}
|
||
|
||
// ================= 能力: detail（种子详情）=================
|
||
// ctx: { baseUrl, id (或 detailId), detailUrl?(可选, 直接指定详情页地址) }
|
||
// 输出字段与能力模板 detail 一致（size/seeders 等保留页面原始文本）。
|
||
function detail(input) {
|
||
var ctx = __asCtx(input);
|
||
var base = __resolveBase(ctx);
|
||
if (!base) return { error: 'no_base_url' };
|
||
var url = ctx && ctx.detailUrl ? String(ctx.detailUrl) : '';
|
||
if (!url) {
|
||
var id = (ctx && (ctx.id || ctx.detailId)) || '';
|
||
if (!id) return { error: 'no_detail_id' };
|
||
url = base + '/details.php?id=' + encodeURIComponent(id);
|
||
}
|
||
if (url.indexOf('http') !== 0) url = base + (url[0] === '/' ? url : '/' + url);
|
||
|
||
var htmlText;
|
||
try {
|
||
htmlText = http.get(url);
|
||
} catch (e) {
|
||
return { error: String((e && e.message) || e), url: url };
|
||
}
|
||
if (!htmlText) return { error: 'empty_response', url: url };
|
||
var doc = html.parse(htmlText);
|
||
if (!doc) return { error: 'parse_failed', url: url, html: htmlText };
|
||
|
||
var out = { url: url };
|
||
|
||
var titleEl = doc.querySelector('#top h1') || doc.querySelector('h1');
|
||
if (titleEl) out.title = titleEl.text();
|
||
var subEl = doc.querySelector('td.embedded');
|
||
if (subEl) out.subtitle = subEl.text();
|
||
var descEl = doc.querySelector('#kdescr');
|
||
if (descEl) out.description = descEl.text();
|
||
|
||
// 行式字段：遍历 td.rowhead，取 label → 下一格文本（兼容表格布局）
|
||
var heads = doc.queryList('td.rowhead');
|
||
for (var i = 0; i < heads.length; i++) {
|
||
var label = (heads[i].text() || '').trim();
|
||
var next = heads[i].nextElementSibling;
|
||
var val = next ? (next.text() || '').trim() : '';
|
||
if (!val) continue;
|
||
if (out.size === undefined && /^(大小|尺寸|體積|Size)$/i.test(label)) out.size = val;
|
||
else if (out.seeders === undefined && /(做种者|做種者|当前做种|當前做種|Seeders?)/i.test(label)) out.seeders = val;
|
||
else if (out.leechers === undefined && /(下载者|下載者|当前下载|當前下載|Leechers?)/i.test(label)) out.leechers = val;
|
||
else if (out.completed === undefined && /(完成|Completed?)/i.test(label)) out.completed = val;
|
||
}
|
||
|
||
// 促销/免费标记（img.pro_free 等，同 search 判定）
|
||
var imgs = doc.queryList('img');
|
||
for (var p = 0; p < imgs.length; p++) {
|
||
var pcls = imgs[p].attr('class') || '';
|
||
var palt = imgs[p].attr('alt') || '';
|
||
if (/pro_free|pro_100pctdown/i.test(pcls) || /^free$|100%/i.test(palt)) { out.free = 'free'; break; }
|
||
if (/pro_50pctdown|pro_free2up/i.test(pcls) || /50%|2x/i.test(palt)) { out.free = '50%'; break; }
|
||
}
|
||
|
||
// 磁力链（部分站点详情页直接提供）
|
||
var links = doc.queryList('a');
|
||
for (var m = 0; m < links.length; m++) {
|
||
var mhref = links[m].attr('href') || '';
|
||
if (mhref.indexOf('magnet:') === 0) { out.magnet = mhref; break; }
|
||
}
|
||
|
||
return out;
|
||
}
|
||
|
||
// ================= 能力: download（下载种子）=================
|
||
// 优先直接使用 ctx.downloadUrl / ctx.detailUrl (搜索解析时抓取的行内下载链接, 已含 passkey/uid 等完整参数, 如 pttime);
|
||
// 无行内链接时由 ctx.id (或 detailId) 构造, 可带 ctx.passkey (仅需 Cookie 的站点如 carpt 不传即可)。
|
||
// 站点会话 (Cookie) 已在 http 客户端注入, Referer 由 Host 设置。
|
||
// 必须走脚本内 http, 不能由外部用裸链接直接 GET。
|
||
// 返回 { downloadUrl, url, bytes, size } 或 { error, url? }.
|
||
function download(input) {
|
||
var ctx = __asCtx(input);
|
||
var base = __resolveBase(ctx);
|
||
var url = null;
|
||
if (ctx.downloadUrl || ctx.detailUrl) {
|
||
url = String(ctx.downloadUrl || ctx.detailUrl);
|
||
if (url.indexOf('http') !== 0) url = base + (url[0] === '/' ? url : '/' + url);
|
||
} else {
|
||
var detailId = ctx && (ctx.id || ctx.detailId) ? String(ctx.id || ctx.detailId) : '';
|
||
if (!detailId) return { error: 'no_detail_id' };
|
||
if (!base) return { error: 'no_base_url' };
|
||
var path = ctx.downloadPath || '/download.php';
|
||
url = base + path + '?id=' + encodeURIComponent(detailId);
|
||
var passkey = ctx.passkey ? String(ctx.passkey) : '';
|
||
if (passkey) url += '&passkey=' + encodeURIComponent(passkey);
|
||
}
|
||
try {
|
||
var bytes = http.getBytes(url);
|
||
if (!bytes || bytes.length === 0) return { error: 'empty_response', url: url };
|
||
return { bytes: bytes, url: url, downloadUrl: url, size: bytes.length };
|
||
} catch (e) {
|
||
return { error: String((e && e.message) || e), url: url };
|
||
}
|
||
}
|
||
|
||
// 刷流入口：拉取站点 RSS/Atom feed 并解析为标准条目列表，供 RssDiscoveryJob 刷流。
|
||
// 必须走脚本内 http（带站点 Cookie），与 search/download 一致的会话上下文。
|
||
// ctx: { siteCode, baseUrl, feedUrl, rssPath?(可选, 默认 /torrentrss.php) }
|
||
// 优先使用 ctx.feedUrl（RssSubscription 配置的完整 RSS 链接，已含 passkey 等参数）；
|
||
// 缺省时回退到 baseUrl + rssPath 拼装。
|
||
// 返回: [{ title, detailUrl, downloadUrl, description, pubDate, guid }] 或 { error }
|
||
function rss(ctx) {
|
||
var base = String(ctx.baseUrl || '').replace(/\/+$/, '');
|
||
var url = ctx.feedUrl ? String(ctx.feedUrl) : '';
|
||
if (!url) {
|
||
var rssPath = ctx.rssPath || '/torrentrss.php';
|
||
if (!base) return { error: 'no_feed_url' };
|
||
url = base + rssPath;
|
||
}
|
||
var xml;
|
||
try {
|
||
xml = http.get(url);
|
||
} catch (e) {
|
||
return { error: String((e && e.message) || e), url: url };
|
||
}
|
||
if (!xml) return { error: 'empty_feed', url: url };
|
||
|
||
// Atom (<entry>) 与 RSS (<item>) 都支持：分别抽取后合并。
|
||
var items = [];
|
||
|
||
function tag(block, name) {
|
||
var m = block.match(new RegExp('<' + name + '[^>]*>([\\s\\S]*?)<\\/' + name + '>', 'i'));
|
||
return m ? m[1].replace(/<!\[CDATA\[([\s\S]*?)\]\]>/g, '$1').trim() : '';
|
||
}
|
||
function attr(block, name, an) {
|
||
var m = block.match(new RegExp(name + '\\s+' + an + '\\s*=\\s*["\']([^"\']*)["\']', 'i'));
|
||
return m ? m[1] : '';
|
||
}
|
||
|
||
// RSS 2.0: <item>…</item>
|
||
var blocks = xml.match(/<item[\s\S]*?<\/item>/gi);
|
||
if (blocks) {
|
||
for (var i = 0; i < blocks.length; i++) {
|
||
var b = blocks[i];
|
||
var title = tag(b, 'title');
|
||
var link = tag(b, 'link');
|
||
var desc = tag(b, 'description');
|
||
var guid = tag(b, 'guid') || link;
|
||
var pub = tag(b, 'pubDate');
|
||
var enc = (b.match(/<enclosure[^>]*>/i) || [])[0] || '';
|
||
var enclosureUrl = attr(enc, 'enclosure', 'url');
|
||
items.push({
|
||
title: title,
|
||
detailUrl: link,
|
||
downloadUrl: enclosureUrl || link,
|
||
description: desc,
|
||
pubDate: pub,
|
||
guid: guid
|
||
});
|
||
}
|
||
}
|
||
|
||
// Atom: <entry>…</entry>
|
||
var atomBlocks = xml.match(/<entry[\s\S]*?<\/entry>/gi);
|
||
if (atomBlocks) {
|
||
for (var j = 0; j < atomBlocks.length; j++) {
|
||
var a = atomBlocks[j];
|
||
var at = tag(a, 'title');
|
||
var aid = tag(a, 'id');
|
||
var updated = tag(a, 'updated') || tag(a, 'published');
|
||
var summary = tag(a, 'summary') || tag(a, 'content');
|
||
// Atom link：优先 rel="alternate" 的 href；无则取首个 href
|
||
var links = a.match(/<link[^>]*>/gi) || [];
|
||
var altHref = '', firstHref = '';
|
||
for (var k = 0; k < links.length; k++) {
|
||
var lh = attr(links[k], 'link', 'href');
|
||
if (!firstHref) firstHref = lh;
|
||
if (/rel=["']alternate["']/i.test(links[k])) altHref = lh;
|
||
}
|
||
var href = altHref || firstHref;
|
||
items.push({
|
||
title: at,
|
||
detailUrl: href,
|
||
downloadUrl: href,
|
||
description: summary,
|
||
pubDate: updated,
|
||
guid: aid
|
||
});
|
||
}
|
||
}
|
||
|
||
if (items.length === 0) {
|
||
return { error: 'no_items', url: url, __len: xml.length, __html: xml };
|
||
}
|
||
return items;
|
||
}
|