python 批量下载m3u8的视频

方法：

解析m3u8，获取其中的ts列表，多线程下载所有ts文件。

全部下完之后，用ffmpeg合并成mp4

代码：

import requests
import os
import threading

tnum = 64

class Downloader(threading.Thread):
    def __init__(self, id, url, ts_list, file_path):
        threading.Thread.__init__(self)
        self.id = id
        self.url = url
        self.ts_list = ts_list
        self.file_path = file_path

    def run(self):
        for i in range(self.id, len(self.ts_list), tnum):
            ts_url = self.ts_list[i]
            r = requests.get(ts_url, stream=True)
            if r.status_code == 200:
                with open(self.file_path + f'_{i}.ts', 'wb') as f:
                    for chunk in r.iter_content(chunk_size=1024):
                        if chunk:
                            f.write(chunk)


def download_m3u8_video(url, file_path):
    r = requests.get(url)
    if r.status_code != 200:
        print('m3u8视频下载链接无效')
        return False

    m3u8_list = r.text.split('\n')
    m3u8_list = [i for i in m3u8_list if i and i[0] != '#']

    ts_list = []
    for ts_url in m3u8_list:
        ts_url = url.rsplit('/', 1)[0] + '/' + ts_url
        ts_list.append(ts_url)

    f = open(file_path+'file_list.txt', 'w+')
    for i in range(len(ts_list)):
        f.write('file \'' + f'_{i}.ts' + "'\n")
    f.close()

    threads = []
    for i in range(0, tnum):
        thread = Downloader(i, url, ts_list, file_path.rsplit('.', 1)[0])
        thread.start()
        threads.append(thread)

    for thread in threads:
        thread.join()

    return not endflag

def merge_file(path, save_path):
    ffmpeg_bin_dic = 'C:/ffmpeg-2023-08-07-git-d295b6b693-full_build/bin/'
    os.system(ffmpeg_bin_dic+'ffmpeg -f concat -safe 0 -i '+path+'file_list.txt'+' -c '+ ' copy ' +save_path)


def down(url):
    name = url[-41:-5]
    print(name)
    ts_file_path = 'D:/v/' + name+'/'
    os.makedirs(ts_file_path)
    mp4_file_path = 'D:/v2/'+name+'.mp4'
    global endflag
    endflag = False
    if download_m3u8_video(url, ts_file_path):
        print('m3u8视频下载完成')
        merge_file(ts_file_path, mp4_file_path)
    else :
        print('error!!!!')
        print(url)

url_list = [
    'https://sth.com/videos/202305091/d63b006e-c6fb-4997-8d43-7ebd086e9c75.m3u8',
    'https://sth.com/videos/202305081/2c9fa41b-e25b-4371-908f-d246628d7bed.m3u8',
]

for url in url_list:
    down(url)