@lzwme/m3u8-dl/cjs/lib/getM3u8Urls.js

A free, open-source, and powerful m3u8 video batch downloader with multi-threaded downloading, play-while-downloading, WebUI management, video parsing, and more.

"use strict";
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+$/.test(text))
        return text;
    const match = /第(\d+)(集|期)/.exec(text);
    if (match)
        return match[0];
    return '';
}
/** 从指定的 url 页面中提取 m3u8 播放地址。deep 指定搜索页面深度 */
async function getM3u8Urls(opts) {
    const options = {
        headers: {},
        deep: 1,
        visited: new Set(),
        maxUrls: 3000,
        maxHtmlSize: 5 * 1024 * 1024, // 5MB
        ...opts,
    };
    const baseUrl = new URL(options.url).origin;
    const req = new fe_utils_1.Request({
        headers: { 'content-type': 'text/html; charset=UTF-8', referer: baseUrl, ...(0, utils_js_1.formatHeaders)(options.headers) },
        reqOptions: { rejectUnauthorized: false },
    });
    const { data: html, response } = await req.get(options.url);
    const m3u8Urls = new Map();
    if (!response.statusCode || response.statusCode >= 400) {
        utils_js_1.logger.error('获取页面失败:', fe_utils_1.color.red(options.url), response.statusCode, response.statusMessage, html);
        return m3u8Urls;
    }
    // 检查 HTML 大小
    if (options.maxHtmlSize && html.length > options.maxHtmlSize) {
        utils_js_1.logger.warn('HTML 内容过大，跳过处理:', fe_utils_1.color.yellow(options.url), `${(html.length / 1024 / 1024).toFixed(2)}MB`, `(限制: ${(options.maxHtmlSize / 1024 / 1024).toFixed(2)}MB)`);
        return m3u8Urls;
    }
    // 从 html 中正则匹配提取 m3u8
    const m3u8Regex = /https?:[^\s'":]+\.(m3u8|mp4)(\?[^\s'"]*)?/gi;
    // 1. 直接正则匹配 m3u8 地址
    let match = m3u8Regex.exec(html);
    const title = (/<title>([^<]+)</.exec(html)?.[1].split('-')[0] || '').replace(/在线播放|在线观看|详情|介绍|《|》/g, '').trim();
    while (match) {
        const href = match[0].replaceAll('\\/', '/');
        match = m3u8Regex.exec(html);
        if (!m3u8Urls.has(href))
            m3u8Urls.set(href, getFormatTitle(title) || title);
    }
    // 找到了多个链接，修改 title 添加序号
    if (m3u8Urls.size > 3 && !/第\d+(集|期)/.test(title)) {
        let idx = 1;
        for (const [key] of m3u8Urls) {
            m3u8Urls.set(key, `${title}第${String(++idx).padStart(2, '0')}集`);
        }
    }
    // 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 && 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 subPageUrls = new Map();
        let failedSubPages = 0;
        while (aMatch) {
            const href = aMatch[1] ? new URL(aMatch[1], baseUrl).toString() : '';
            const text = aMatch[2].replace(/<[^>]+>/g, '');
            aMatch = aTagRegex.exec(html);
            if (!href || options.visited.has(href) || !href.startsWith(baseUrl))
                continue;
            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(' > 提取到子页面: ', fe_utils_1.color.gray(href), text);
        }
        const taskList = Array.from(subPageUrls.entries()).map(([href, text]) => {
            return async () => {
                // 达到最大 URL 数量限制，不再继续
                if (options.maxUrls && m3u8Urls.size >= options.maxUrls)
                    return;
                try {
                    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`, options.url, href);
                            return;
                        }
                    }
                    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);
                }
            };
        });
        await (0, fe_utils_1.concurrency)(taskList, Math.max(1, +options.concurrency || 3));
    }
    return m3u8Urls;
}
// logger.updateOptions({ levelType: 'debug' });
// getM3u8Urls({ url: process.argv.slice(2)[0] }).then(d => console.log(d));