ytb_dlp源码解析

源代码

# -*- coding:utf-8 -*-
from yt_dlp import YoutubeDL
import pickle,os
os.environ["http_proxy"] = "http://127.0.0.1:10809"
os.environ["https_proxy"] = "http://127.0.0.1:10809"
with open('../urls.pkl','rb')as f:
    urls_list=pickle.load(f)
# youtube测试网址
# https://www.youtube.com/watch?v=BaW_jenozKc

urls_list=[ "https://www.youtube.com/watch?v="+url for url in urls_list]
# urls_list = ['https://www.youtube.com/watch?v=BaW_jenozKc']
# print(urls_list)
start_time = '00:00:5'  # start time in HH:MM:SS format
end_time = '00:00:10'
yld_opts={
    'format': 'bestvideo[height<=360]',
    'noplaylist': True,
    'writesubtitles': False,
    'audio': False,
    'outtmpl':'../origin_data/%(title)s_%(id)s.%(ext)s',
    'concurrent_fragment_downloads': 4,
    'limit_rate': None,
    # 'download_sections': f"*{start_time}-{end_time}"
}
x=1
with YoutubeDL(yld_opts) as ydl:
    ydl.download(urls_list[50:100])

进入YoutubeDL.py

1. __init__最后一行,返回self.archive=None

...
            except OSError as ioe:
                if ioe.errno != errno.ENOENT:
                    raise
            return archive
self.archive = preload_download_archive(self.params.get('download_archive'))
...

2. download()

        for url in url_list:
            self.__download_wrapper(self.extract_info)(
                url, force_generic_extractor=self.params.get('force_generic_extractor', False))

        return self._download_retcode

force_generic_extractor:False
3. extract_info

        if ie_key:
            ies = {ie_key: self._ies[ie_key]} if ie_key in self._ies else {}
        else:
            ies = self._ies

ies=self._ies