python 玩轉m3u8視頻

最近在搗鼓視頻轉碼，所謂好記性不如爛筆頭，索性把想到的知識點都記錄在案吧

現在大多數網站都使用m3u8格式的視頻文件，一是體積小，傳輸快；二是可以切片，提高用戶體驗。

好了，廢話不多說，開始今天的主題:

假如我們在網站上看到一個好的視頻，我們想下載下來，怎么辦？？那這篇文章可以幫助到你（不過你要會使用pthon啊，騷年~~）

先放一張流程圖：

 

具體就分這4步，概括一下就是在瀏覽器中找到這個m3u8文件，下載里面所有的ts文件，最后將所有的ts文件合並為mp4視頻。

完整上碼

import re
import requests, os
project_path = os.path.abspath(os.path.dirname(__file__))
video_path = os.path.abspath(os.path.join(os.path.dirname(__file__), 'video'))


class VideoSynthesis:
    def __init__(self):
        self.header = {
            'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/71.0.3578.98 Safari/537.36'
        }
        self.m3u8_link = 'https://v.pinimg.com/videos/v2/hls/49/3e/6b/493e6bb3f832caf2bef872d648698ff4_hls480w.m3u8'
        self.base_url = 'https://v.pinimg.com/videos/v2/hls/49/3e/6b/'

    def get_m3u8_link(self):
        # 下載m3u8視頻
        m3u8_response = requests.get(url=self.m3u8_link, headers=self.header, verify=False)
        print(m3u8_response.text)
        return m3u8_response

    def handle_m3u8_response(self, m3u8_response):
        # 處理m3u8響應
        ts_filename = re.findall('.*?EXTINF:.*?,\n(.*?)\n', m3u8_response.replace(' ', '').replace(r'\n', ''))
        print(f'tslist: {ts_filename}')
        ts_link_list = []
        for item in ts_filename:
            result_url = self.base_url + item
            ts_link_list.append(result_url)
        print(f'ts_link_list: {ts_link_list}')
        return ts_link_list, ts_filename

    def download_ts_video(self, ts_url_list, ts_filename):
        # 通過ts鏈接下載ts文件
        for i in range(len(ts_url_list)):
            ts_url = ts_url_list[i]
            ts_name = ts_filename[i]
            try:
                response = requests.get(ts_url, stream=True, verify=False)
            except Exception as e:
                print("異常請求：%s" % e.args)
                return

            # 在當前目錄下創建個video文件夾
            ts_path = project_path + r"\video\{}".format(ts_name)
            with open(ts_path, "wb+") as file:
                for chunk in response.iter_content(chunk_size=1024):
                    if chunk:
                        file.write(chunk)
        print("TS文件下載完畢！！")

    def heBingTsVideo(self):
        # 視頻合成
        hebing_path = video_path + r"\m3u8_video.mp4"
        print(hebing_path)
        all_ts = os.listdir(video_path)
        with open(hebing_path, 'wb+') as f:
            for i in range(len(all_ts)):
                ts_video_path = os.path.join(video_path, all_ts[i])
                f.write(open(ts_video_path, 'rb').read())
        print("合並完成！！")

    def run(self):
        m3u8_response = self.get_m3u8_link()
        ts_link_list, ts_filename = self.handle_m3u8_response(m3u8_response)
        
        # ts_link_list = ['https://v.pinimg.com/videos/v2/hls/49/3e/6b/493e6bb3f832caf2bef872d648698ff4_hls480w00000.ts', 'https://v.pinimg.com/videos/v2/hls/49/3e/6b/493e6bb3f832caf2bef872d648698ff4_hls480w00001.ts', 'https://v.pinimg.com/videos/v2/hls/49/3e/6b/493e6bb3f832caf2bef872d648698ff4_hls480w00002.ts', 'https://v.pinimg.com/videos/v2/hls/49/3e/6b/493e6bb3f832caf2bef872d648698ff4_hls480w00003.ts']
        # ts_filename = ['493e6bb3f832caf2bef872d648698ff4_hls480w00000.ts', '493e6bb3f832caf2bef872d648698ff4_hls480w00001.ts' ,'493e6bb3f832caf2bef872d648698ff4_hls480w00002.ts', '493e6bb3f832caf2bef872d648698ff4_hls480w00003.ts']
        
        self.download_ts_video(ts_link_list, ts_filename)
        self.heBingTsVideo()


if __name__ == '__main__':
    video_synthesis = VideoSynthesis()
    video_synthesis.run()

 

視頻下載完以后，可能還會遇到壓縮，合並問題，那這時候就要用到ffmpeg 這個神器了，直接編寫一個腳本就完事了~

 

# coding=utf-8

import os

NEW_RESOLUTION = "640x480"  # 目標分辨率，常量
NEW_FPS = 12  # 目標幀率，常量

curpath = os.getcwd()  # 獲取當前路徑
input_dir = os.path.join(curpath, "Input_Video")
output_dir = os.path.join(curpath, "Output_Video")
input_video_list = os.listdir(input_dir)  # 獲取視頻列表

# 如果沒有Output_Video這個文件夾，則創建這個文件夾
if not os.path.exists(output_dir):
    os.mkdir(output_dir)

# 開始批量二次編碼壓縮視頻轉碼
for each_video in input_video_list:
    video_name, _ = os.path.splitext(each_video)  # _是沒意義，就只是一個無用代號，占個坑而已
    ffmpeg_command = ("ffmpeg -i %s%s%s -s %s -r %s -y %s%s%s_c.mp4" % (
        input_dir, os.sep, each_video, NEW_RESOLUTION, NEW_FPS, output_dir, os.sep, video_name))
    print(ffmpeg_command)
    os.system(ffmpeg_command)

os.system("pause")

注意：這里調用的是系統指令，所以務必使你的ffmpeg.exe執行程序能夠被訪問到，放到系統全局變量路徑下面即可。

 

視頻合並

ffmpeg -f concat -i filelist.txt -c copy output.mp4

filelist.txt為ffmpeg同級目錄下新建記事本，名字隨意，而里面的內容長這個樣子

file 's1-1.mp4'
file 's1-2.mp4'
file 's1-3.mp4'

注意，如果用這種方式，那么這3個文件也必須存放在同級目錄下。如果想使用全路徑的形式，比如：file D:\s1\2.mp4'

則應使用下面這條命令

ffmpeg -f concat -safe 0 -i filelist.txt -c copy output.mp4　　-- -safe 0 授權進入

 

 

參考文章　　https://blog.csdn.net/weixin_38640052/article/details/119680697

　　　　　　https://blog.csdn.net/human_soul/article/details/103263573

　　　　　　https://www.cnblogs.com/zipython/p/12920095.html

　　　　　　https://jingyan.baidu.com/article/2a138328517dce464b134f24.html

　　　　　　https://blog.csdn.net/winniezhang/article/details/89260841