樱&花动漫接口逆向

Dituon

想实现番剧下载的爬虫，花了两天时间逆向某知名动画网站的接口

yhmgo.com

https://github.com/Dituon/yhmgo-parser

注意：悄悄滴使用，声张滴不要

Mr.喜

此回复已被删除！

Dituon

更新: 爬虫示例代码

export const config = {
    fetchHeader: {
        "Sec-Ch-Ua": '"Not_A Brand";v="8", "Chromium";v="120", "Microsoft Edge";v="120"',
        "Dnt": '1',
        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36 Edg/120.0.0.0"
    }
}

export interface VideoInfo {
    url: string
    // type: string
}

export abstract class Parser {
    public abstract parse(raw: string): Promise<VideoInfo>
}

import fetch from 'node-fetch'
import * as cookie from 'cookie'

const exp = /www\.yhmgo\.com\/vp\/(\d+)-(\d+)-(\d+)/
const API_BASE = 'https://www.yhmgo.com/playurl'
const PAGE_BASE = 'https://www.yhmgo.com/vp'

export class YhmgoParser extends Parser {
    async parse(url: string): Promise<VideoInfo> {
        const [match, id, channel, ep] = url.match(exp)
        if (!match) {
            throw new Error('Invalid url')
        }
        const pageUrl = `${PAGE_BASE}/${id}-${channel}-${ep}.html`
        const page = await fetch(pageUrl)

        let {t1, k1} = page.headers.raw()['set-cookie'].reduce((obj, str) => {
            return {...cookie.parse(str), ...obj}
        }, {})
        let cookies = parseCookie(t1, k1)

        const raw = await fetch(
            `${API_BASE}?aid=${id}&playindex=${channel}&epindex=${ep}&r=${Math.random()}`,
            {
                headers: {
                    ...config.fetchHeader,
                    referer: pageUrl,
                    cookie: Object.entries(cookies)
                        .map(c => cookie.serialize(c[0], c[1]))
                        .join('; ')
                }
            }
        )
        const {vurl} = parseResult(await raw.text())


        return {
            url: decodeURIComponent(vurl),
            // type: 'm3u8',
        }
    }
}

export function parseCookie(t1: number, k1: number) {
    let pos = Math.floor(t1 / 0x3e8) >> 0x5
    let offset = 0x89a4
    let k2 = '' + (((((pos * (pos % 0x100) + 0x1) + offset) * ((pos % 0x80) + 0x1)) * ((pos % 0x10) + 0x1)) + pos)

    let t2 = ''
    while (true) {
        t2 = '' + Math.floor(new Date().getTime())
        let tOffset = t2.slice(t2.length - 0x3)
        let kOffset = k2.slice(k2.length - 0x1)
        if (tOffset.indexOf(kOffset) >= 0x0) {
            break
        }
    }

    let start = 0x0
    let timeOffset = Math.floor(new Date().getTime() / 0x3e8) >> (0x11 + start)
    let m2t = ((((timeOffset * 0x15 + 0x9a) * (((timeOffset % 0x40) + 0xd)) * ((timeOffset % 0x20) + 0x22) * ((timeOffset % 0x10) + 0x57)) * ((timeOffset % 0x8) + 0x41)) + 0x2ef)

    return {
        t1, k1, t2, k2, m2t
    }
}

export function parseResult(raw: string): {
    purl: string,
    vurl: string,
    inv: string
} {
    let result = ''
    const offset = 0x619
    const rawLength = raw.length
    for (let i = 0x0; i < rawLength; i += 0x2) {
        let temp = parseInt(raw[i] + raw[i + 1], 0x10)
        temp = (
            (temp + 0x100000 - offset) - (((rawLength / 0x2) - 0x1) - (i / 0x2))
        ) % 0x100
        result = String.fromCharCode(temp) + result
    }
    return JSON.parse(result)
}

使用

import {YhmgoParser} from "./parser/yhmgo";

const parser = new YhmgoParser()
const video = await parser.parse('https://www.yhmgo.com/vp/11141-2-0.html')
console.log(video)

// 输出: { url: 'https://vip.lz-cdn16.com/20230317/13224_75d9217b/index.m3u8' }