@lzwme/m3u8-dl 1.4.1 → 1.4.2

package/cjs/lib/file-download.js

@@ -39,7 +39,7 @@ async function fileDownload(u, opts) {
             force: options.force,
             requestOptions: {
                 headers: {
-                    referer: url,
+                    referer: new URL(url).origin,
                     ...(0, utils_js_1.formatHeaders)(options.headers),
                 },
                 rejectUnauthorized: false,

package/cjs/lib/getM3u8Urls.d.ts

@@ -1,3 +1,11 @@
 import type { OutgoingHttpHeaders } from 'node:http';
+export interface GetM3u8UrlsOption {
+    url: string;
+    /** 播放子页面 URL 特征规则 */
+    subUrlRegex?: string | RegExp;
+    headers?: OutgoingHttpHeaders | string;
+    deep?: number;
+    visited?: Set<string>;
+}
 /** 从指定的 url 页面中提取 m3u8 播放地址。deep 指定搜索页面深度 */
-export declare function getM3u8Urls(url: string, headers?: OutgoingHttpHeaders | string, deep?: number, visited?: Set<string>): Promise<Map<string, string>>;
+export declare function getM3u8Urls(opts: GetM3u8UrlsOption): Promise<Map<string, string>>;

package/cjs/lib/getM3u8Urls.js

@@ -3,67 +3,98 @@ Object.defineProperty(exports, "__esModule", { value: true });
 exports.getM3u8Urls = getM3u8Urls;
 const fe_utils_1 = require("@lzwme/fe-utils");
 const utils_js_1 = require("./utils.js");
+function getFormatTitle(text) {
+    if (typeof text !== 'string' || !text)
+        return '';
+    if (/^\d+$/.test(text))
+        return text;
+    const match = /第(\d+)(集|期)/.exec(text);
+    if (match)
+        return match[0];
+    return '';
+}
 /** 从指定的 url 页面中提取 m3u8 播放地址。deep 指定搜索页面深度 */
-async function getM3u8Urls(url, headers = {}, deep = 2, visited = new Set()) {
-    const req = new fe_utils_1.Request({ headers: { 'content-type': 'text/html; charset=UTF-8', referer: new URL(url).origin, ...(0, utils_js_1.formatHeaders)(headers) } });
-    const { data: html, response } = await req.get(url);
+async function getM3u8Urls(opts) {
+    const options = { headers: {}, deep: 1, visited: new Set(), ...opts };
+    const baseUrl = new URL(options.url).origin;
+    const req = new fe_utils_1.Request({
+        headers: { 'content-type': 'text/html; charset=UTF-8', referer: baseUrl, ...(0, utils_js_1.formatHeaders)(options.headers) },
+        reqOptions: { rejectUnauthorized: false },
+    });
+    const { data: html, response } = await req.get(options.url);
     const m3u8Urls = new Map();
     if (!response.statusCode || response.statusCode >= 400) {
-        utils_js_1.logger.error('获取页面失败:', fe_utils_1.color.red(url), response.statusCode, response.statusMessage, html);
+        utils_js_1.logger.error('获取页面失败:', fe_utils_1.color.red(options.url), response.statusCode, response.statusMessage, html);
         return m3u8Urls;
     }
     // 从 html 中正则匹配提取 m3u8
-    const m3u8Regex = /https?:[^\s'"]+\.m3u8(\?[^\s'"]*)?/gi;
+    const m3u8Regex = /https?:[^\s'":]+\.(m3u8|mp4)(\?[^\s'"]*)?/gi;
     // 1. 直接正则匹配 m3u8 地址
     let match = m3u8Regex.exec(html);
-    const title = (/<title>([^<]+)</.exec(html)?.[1].split('-')[0] || '').replace(/在线播放|详情|介绍|《|》/g, '').trim();
+    const title = (/<title>([^<]+)</.exec(html)?.[1].split('-')[0] || '').replace(/在线播放|在线观看|详情|介绍|《|》/g, '').trim();
     while (match) {
         const href = match[0].replaceAll('\\/', '/');
         match = m3u8Regex.exec(html);
         if (!m3u8Urls.has(href))
-            m3u8Urls.set(href, title);
+            m3u8Urls.set(href, getFormatTitle(title) || title);
     }
     // 找到了多个链接，修改 title 添加序号
-    if (m3u8Urls.size > 3 && !/第.+(集|期)/.test(title)) {
+    if (m3u8Urls.size > 3 && !/第\d+(集|期)/.test(title)) {
         let idx = 1;
         for (const [key] of m3u8Urls) {
-            m3u8Urls.set(key, `${title}第${String(idx++).padStart(2, '0')}集`);
+            m3u8Urls.set(key, `${title}第${String(++idx).padStart(2, '0')}集`);
         }
     }
     // 2. 若未找到且深度大于 0，则获取所有 a 标签的 href 并递归查找
-    if (m3u8Urls.size === 0 && deep > 0) {
-        utils_js_1.logger.debug('未获取到 m3u8 地址', fe_utils_1.color.gray(url), html.length);
-        visited.add(url);
+    if (m3u8Urls.size === 0 && options.deep > 0) {
+        utils_js_1.logger.debug('未获取到 m3u8 地址', fe_utils_1.color.gray(options.url), html.length);
+        options.visited.add(options.url);
         const aTagRegex = /<a\s+[^>]*href=["']([^"']+)["'][^>]*>(.*?)<\/a>/gi;
         let aMatch = aTagRegex.exec(html);
-        const origin = new URL(url).origin;
         const subPageUrls = new Map();
         let failedSubPages = 0;
         while (aMatch) {
-            const href = aMatch[1] ? new URL(aMatch[1], origin).toString() : '';
+            const href = aMatch[1] ? new URL(aMatch[1], baseUrl).toString() : '';
             const text = aMatch[2].replace(/<[^>]+>/g, '');
             aMatch = aTagRegex.exec(html);
-            if (!href || visited.has(href) || !href.startsWith(origin))
+            if (!href || options.visited.has(href) || !href.startsWith(baseUrl))
                 continue;
-            if (!/集|期|HD|高清|抢先|BD/.test(text))
+            if (options.subUrlRegex) {
+                if (typeof options.subUrlRegex === 'string') {
+                    options.subUrlRegex = new RegExp(options.subUrlRegex.replaceAll(/\*+/g, '.+'));
+                }
+                if (!options.subUrlRegex.test(href))
+                    continue;
+            }
+            else if (!/集|期|HD|高清|抢先|BD/.test(text))
                 continue;
             subPageUrls.set(href, text);
-            utils_js_1.logger.debug(' > 提取到m3u8链接: ', fe_utils_1.color.gray(href), text);
+            utils_js_1.logger.debug(' > 提取到子页面: ', fe_utils_1.color.gray(href), text);
         }
         for (const [href, text] of subPageUrls) {
             try {
-                visited.add(href);
-                const subUrls = await getM3u8Urls(href, headers, deep - 1, visited);
-                utils_js_1.logger.debug(' > 从子页面提取: ', fe_utils_1.color.gray(href), text, subUrls.size);
+                options.visited.add(href);
+                const subUrls = await getM3u8Urls({ ...options, url: href, deep: options.deep - 1 });
+                utils_js_1.logger.debug(' > 从子页面提取: ', fe_utils_1.color.gray(href), text, subUrls);
                 if (subUrls.size === 0 && m3u8Urls.size === 0) {
                     failedSubPages++;
                     if (failedSubPages > 3) {
-                        utils_js_1.logger.warn(`连续查找 ${failedSubPages} 个子页面均未获取到，不再继续`, url, href);
+                        utils_js_1.logger.warn(`连续查找 ${failedSubPages} 个子页面均未获取到，不再继续`, options.url, href);
                         return m3u8Urls;
                     }
                 }
-                for (const [u, t] of subUrls)
-                    m3u8Urls.set(u, subUrls.size === 1 || /第.+(集|期)/.test(t) ? t : text);
+                for (const [u, t] of subUrls) {
+                    let stitle = t;
+                    for (const s of [text, t, m3u8Urls.get(u) || '']) {
+                        const ft = getFormatTitle(s);
+                        if (ft) {
+                            stitle = ft;
+                            break;
+                        }
+                    }
+                    utils_js_1.logger.debug(' > m3u8地址: ', fe_utils_1.color.gray(u), fe_utils_1.color.green(stitle));
+                    m3u8Urls.set(u, stitle.trim());
+                }
             }
             catch (err) {
                 utils_js_1.logger.warn(' > 尝试访问子页面异常: ', fe_utils_1.color.red(href), err.message);

package/cjs/server/download-server.js

@@ -55,7 +55,7 @@ class DLServer {
         cacheDir: process.env.DS_CACHE_DIR || (0, node_path_1.resolve)((0, node_os_1.homedir)(), '.m3u8-dl/cache'),
         token: process.env.DS_SECRET || process.env.DS_TOKEN || '',
         debug: process.env.DS_DEBUG === '1',
-        limitFileAccess: !['0', 'false'].includes(process.env.DS_LIMTE_FILE_ACCESS),
+        limitFileAccess: ['1', 'true'].includes(process.env.DS_LIMTE_FILE_ACCESS),
     };
     serverInfo = {
         version: '',
@@ -198,7 +198,7 @@ class DLServer {
                 if ((0, node_fs_1.existsSync)((0, node_path_1.resolve)(rootDir, 'client/local/cdn'))) {
                     indexHtml = indexHtml
                         .replaceAll('https://s4.zstatic.net/ajax/libs', 'local/cdn')
-                        .replaceAll(/integrity=".+"\n?/g, '')
+                        .replaceAll(/integrity="[^"]+"\n?/g, '')
                         .replace('https://cdn.tailwindcss.com/3.4.16', 'local/cdn/tailwindcss/3.4.16/tailwindcss.min.js');
                 }
                 res.setHeader('content-type', 'text/html').send(indexHtml);
@@ -253,7 +253,7 @@ class DLServer {
         if (!url)
             return utils_js_1.logger.error('[satartDownload]Invalid URL:', url);
         if (url.endsWith('.html')) {
-            const item = Array.from(await (0, getM3u8Urls_js_1.getM3u8Urls)(url, options.headers))[0];
+            const item = Array.from(await (0, getM3u8Urls_js_1.getM3u8Urls)({ url, headers: options.headers }))[0];
             if (!item)
                 return utils_js_1.logger.error('[startDownload]不是有效(包含)M3U8的地址:', url);
             url = item[0];
@@ -564,12 +564,12 @@ class DLServer {
             res.status(404).send({ message: 'Not Found', code: 404 });
         });
         app.post('/api/getM3u8Urls', (req, res) => {
-            const { url, headers } = req.body;
+            const { url, headers, subUrlRegex } = req.body;
             if (!url) {
                 res.json({ code: 1001, message: '无效的 url 参数' });
             }
             else {
-                (0, getM3u8Urls_js_1.getM3u8Urls)(url, headers)
+                (0, getM3u8Urls_js_1.getM3u8Urls)({ url, headers, subUrlRegex })
                     .then(d => res.json({ code: 0, data: Array.from(d) }))
                     .catch(err => res.json({ code: 401, message: err.message }));
             }

package/client/index.html

@@ -832,7 +832,7 @@ services:
             html: `
               <div class="text-left">
                 <div class="flex flex-row gap-4">
-                  <input type="text" id="playUrl" placeholder="[实验性]输入视频播放页地址，尝试提取m3u8下载链接" autocomplete="off" id="urlInput" class="flex-1 border border-gray-300 rounded px-3 py-2 focus:outline-none focus:ring-2 focus:ring-blue-400" value="">
+                  <input type="text" id="playUrl" placeholder="[实验性]输入列表页或播放页地址，提取m3u8链接" autocomplete="off" id="urlInput" class="flex-1 border border-gray-300 rounded px-3 py-2 focus:outline-none focus:ring-2 focus:ring-blue-400" value="">
                   <div class="flex flex-row gap-1">
                     <button type="button" id="getM3u8UrlsBtn" class="player-btn px-2 py-1 bg-blue-600 text-white rounded hover:bg-blue-700 focus:outline-none">
                       提取
@@ -840,6 +840,13 @@ services:
                   </div>
                 </div>
 
+                <div class="mt-4">
+                  <div class="flex flex-row gap-2 items-center">
+                    <input id="subUrlRegex" class="flex-1 p-2 border rounded-lg focus:border-blue-500" placeholder="[实验性](可选)播放页链接特征规则">
+                  </div>
+                  <p class="ml-2 mt-1 text-sm text-gray-500">用于从视频列表页准确识别播放地址。如：<code>play/845-1-</code></p>
+                </div>
+
                 <div class="mt-4">
                   <label class="block text-sm font-bold text-gray-700 mb-1">视频链接（每行一个，支持m3u8地址及抖音、微博、皮皮虾视频分享链接）</label>
                   <textarea id="downloadUrls" class="w-full p-2 border rounded-lg focus:ring-blue-500" rows="3" placeholder="格式： URL | 名称(可选)、名称 | URL"></textarea>
@@ -930,7 +937,9 @@ services:
               btn.setAttribute('disabled', 'disabled');
               btn.innerText = '解析中...';
 
-              T.post('/api/getM3u8Urls', { url, headers: document.getElementById('headers').value.trim() }).then(r => {
+              const headers = document.getElementById('headers').value.trim();
+              const subUrlRegex = document.getElementById('subUrlRegex').value.trim();
+              T.post('/api/getM3u8Urls', { url, headers, subUrlRegex }).then(r => {
                 if (Array.isArray(r.data)) {
                   document.getElementById('downloadUrls').value = r.data.map(d => d.join(' | ')).join('\n');
                   T.toast(r.message || `解析完成！获取到 ${r.data.length} 个地址`);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lzwme/m3u8-dl",
-  "version": "1.4.1",
+  "version": "1.4.2",
   "description": "Batch download of m3u8 files and convert to mp4",
   "main": "cjs/index.js",
   "types": "cjs/index.d.ts",
@@ -44,24 +44,24 @@
     "registry": "https://registry.npmjs.com"
   },
   "devDependencies": {
-    "@biomejs/biome": "^2.1.1",
-    "@eslint/js": "^9.30.1",
+    "@biomejs/biome": "^2.1.2",
+    "@eslint/js": "^9.31.0",
     "@lzwme/fed-lint-helper": "^2.6.6",
     "@types/express": "^5.0.3",
     "@types/m3u8-parser": "^7.2.2",
-    "@types/node": "^24.0.12",
+    "@types/node": "^24.0.15",
     "@types/ws": "^8.18.1",
-    "@typescript-eslint/eslint-plugin": "^8.36.0",
-    "@typescript-eslint/parser": "^8.36.0",
-    "eslint": "^9.30.1",
-    "eslint-config-prettier": "^10.1.5",
-    "eslint-plugin-prettier": "^5.5.1",
+    "@typescript-eslint/eslint-plugin": "^8.37.0",
+    "@typescript-eslint/parser": "^8.37.0",
+    "eslint": "^9.31.0",
+    "eslint-config-prettier": "^10.1.8",
+    "eslint-plugin-prettier": "^5.5.3",
     "express": "^5.1.0",
     "husky": "^9.1.7",
     "prettier": "^3.6.2",
     "standard-version": "^9.5.0",
     "typescript": "^5.8.3",
-    "typescript-eslint": "^8.36.0",
+    "typescript-eslint": "^8.37.0",
     "ws": "^8.18.3"
   },
   "dependencies": {