背景：

客户端上报md5校验失败，想复现，但是又不能一条一条的手工验证cdn资源

目标：

给定m3u8文件地址，自动下载与校验全部分片的md5

代码：

# coding=utf-8

import sys
import requests
import hashlib

def checkCdnFile(url):
    """
    检测一个segment
    """
    # url = 'http://valipl.cp31.ott.cibntv.net/67756D6080932713CFC02204E/03000500005DBC8F1334A405C99AC17FE460FC-F6E0-4DC0-8AE4-579B006E3EC6-00143.ts?ccode=01010201&duration=2647&expire=18000&psid=56bfdf60419bdb7f42e668114df58a1341cb3&ups_client_netip=ddc2f33f&ups_ts=1593570410&ups_userid=&utid=WkJ1nK5mJecDAE9Wggk%2FsUOv&vid=XMjQyODU1MDMy&sm=1&operate_type=1&dre=u29&si=79&eo=1&dst=1&iv=0&s=cbffd1c2962411de83b1&type=flvhdv3&bc=2&vkey=Bc2a976f747039b61fc2b55a717af1867'
    requestHeaders = {'User-Agent': 'Android'}
    response = requests.get(url)

    headers = response.headers

    md5 = headers.get('ETag')
    md5 = md5[1: (len(md5) - 1)]
    md5 = md5.lower()

    print("headerMd5:" + md5)

    with open('temp.ts', 'wb') as tempTs:
        tempTs.write(response.content)

    with open('temp.ts', 'rb') as tempTs:
        data = tempTs.read()
        realMd5 = hashlib.md5(data).hexdigest()
        print("realMd5:" + realMd5)
        if (md5 == realMd5):
            print("correct_segment")
        else:
            print("error_segment：" + url)
            print("error_header_md5" + md5)
            print("error_real_md5" + realMd5)
            sys.exit(1)


# checkCdnFile('http://m-vali.cp31.ott.cibntv.net/6581031CC4F03E719C05A73654/03000C14125CF5672A664BC457BC38537F0F2F-887E-447C-9791-5CA361FBA9C8.mp4?ccode=01010201&duration=131&expire=18000&psid=37ccbe5449a577cf9e11a940f90cdd1f41cb3&ups_client_netip=758850ca&ups_ts=1593577163&ups_userid=1865339496&utid=WNdzjS7%2Bx8oDAJ2fy%2F0l6haR&vid=XNDIxNjUyNDQ4OA&vkey=Babcac71d12d0f156b6435e8ad9b1eb49&iv=1&eo=0&bc=2&dre=u13&si=43&dst=1')

# 根据url下载m3u8文件.当然也可以注释掉这个下载
m3u8_url = 'http://valipl.cp31.ott.cibntv.net/657227C075630713D75A441E8/05000A00005EBBBA588BB780000000690827C6-432D-4F9E-855E-A710EB3F0518.m3u8?ccode=01010201&duration=2744&expire=18000&psid=4b9f64c3c529489de9dbf10de7cc77bb41cb3&ups_client_netip=df659a7b&ups_ts=1593576619&ups_userid=1434082693&utid=XjtxFE23ve8DAF7F2Cm1G2Yf&vid=XNDU0MTc1ODE0NA&vkey=B9b36304de1830bfadec05c7525f1cded&sm=1&operate_type=1&dre=u29&si=79&eo=1&dst=1&iv=0&s=e585299e47d911e6abda&type=mp5hd2v3&bc=2'
response = requests.get(m3u8_url)
with open('youku.m3u8', 'wb') as tempTs:
    tempTs.write(response.content)

# 读取m3u8每行文本
with open('youku.m3u8', 'r') as m3u8:
    segmentCount = 0
    for line in m3u8:
        line = line.strip('\n')
        if line.startswith("http"):
            segmentCount += 1
            print('no:%d'%(segmentCount))
            checkCdnFile(line)

效果：

image.png