下载 Youtube 视频和字幕的脚本

Youtube-downloader.py

# 需要下载
# https://github.com/iawia002/annie
# 将 annie 可执行文件放在本脚本同目录或者加入环境变量
# 或者自行修改相对路径
# 需要自行使用 IFTTT 或者其他手段创建 Youtube 视频地址列表
# 详情请参阅
# https://blog.tookdes.org/archives/download-youtube-subscribe-video-with-srt.html

# -*- coding: utf-8 -*-

import subprocess
import requests
import json
import time
import datetime

srt_api_key = u"a2d09c7d76fced01f8be4b1f4cce8bec" #测试用 API，也可以自行更改
url_file = "C:\YourPath\Subscribe.txt"#定义好自己文件路径

url_list = []
video_id_list = []

with open(url_file) as f:
    url = f.read().splitlines()
    url_list.extend(url)
#print(url_list)

for i in range(0,len(url_list)):
        video_id_list.append(url_list[i].replace('https://youtu.be/', ''))
#print(video_id_list)

def get_vid(url):
    exe = "annie.exe -s 127.0.0.1:1080 " + url #自行更改 socks5 代理地址[doge]，详见 annie 说明
    subprocess.call(exe)

def get_srt(video_id, url):
    params_dict = {"api-key":srt_api_key}
    exe = "annie.exe -s 127.0.0.1:1080 -j " + url #自行更改代理地址
    video_title = json.loads(subprocess.Popen(exe, shell=True,\
                                       stdout=subprocess.PIPE,\
                                       stderr=subprocess.STDOUT).stdout.read())\
                                       ["Title"]
# API 获得的 title 并不一样，不推荐使用
#    video_title = json.loads(requests.get("https://api.zhuwei.me/v1/captions/"+\
#                                        video_id,\
#                                        params = params_dict).text)\
#                                        ["response"]["captions"]["title"]
    srt_get = json.loads(requests.get("https://api.zhuwei.me/v1/captions/"+\
                                        video_id,\
                                        params = params_dict).text)
#    print(srt_get)
    if srt_get["response"]["captions"]["available_captions_count"] == 0:
        pass
    else:
        for i in range(0,srt_get["response"]["captions"]["available_captions_count"]):
            srt_lang = srt_get["response"]["captions"]["available_captions"][i]["language"]
            srt_url = srt_get["response"]["captions"]["available_captions"][i]["caption_content_url"]
            srt_data = json.loads(requests.get(srt_url, params = params_dict).text)["contents"]["content"]
            print(video_id + " " + srt_lang + " srt done.")
            srt_file = open(video_title + "." + srt_lang +".srt","w",encoding='utf-8')#如果不转编码可能会导致中文字幕写入失败
            srt_file.write(srt_data)
            srt_file.close

def main():
    for i in range(0,len(url_list)):
        get_vid(url_list[i])
        get_srt(video_id_list[i], url_list[i])
    with open(url_file,"w+") as f:
            f.truncate()
            f.close

main() #如果采用外部定时运行
exit()

##如果后台连续运行
##while True:
##    now = datetime.datetime.now()
##    if now.hour==14: #and now.minute==10: #自己定义运行脚本的时间
##        main()
##        time.sleep(3600) #自己定制