yt_dlp/extractor/blogger.py

   1 import re
   2
   3 from ..utils import (
   4     mimetype2ext,
   5     parse_duration,
   6     parse_qs,
   7     str_or_none,
   8     traverse_obj,
   9 )
  10 from .common import InfoExtractor
  11
  12
  13 class BloggerIE(InfoExtractor):
  14     IE_NAME = 'blogger.com'
  15     _VALID_URL = r'https?://(?:www\.)?blogger\.com/video\.g\?token=(?P<id>.+)'
  16     _VALID_EMBED = r'''<iframe[^>]+src=["']((?:https?:)?//(?:www\.)?blogger\.com/video\.g\?token=[^"']+)["']'''
  17     _TESTS = [{
  18         'url': 'https://www.blogger.com/video.g?token=AD6v5dzEe9hfcARr5Hlq1WTkYy6t-fXH3BBahVhGvVHe5szdEUBEloSEDSTA8-b111089KbfWuBvTN7fnbxMtymsHhXAXwVvyzHH4Qch2cfLQdGxKQrrEuFpC1amSl_9GuLWODjPgw',
  19         'md5': 'f1bc19b6ea1b0fd1d81e84ca9ec467ac',
  20         'info_dict': {
  21             'id': 'BLOGGER-video-3c740e3a49197e16-796',
  22             'title': 'BLOGGER-video-3c740e3a49197e16-796',
  23             'ext': 'mp4',
  24             'thumbnail': r're:^https?://.*',
  25             'duration': 76.068,
  26         }
  27     }]
  28
  29     @staticmethod
  30     def _extract_urls(webpage):
  31         return re.findall(BloggerIE._VALID_EMBED, webpage)
  32
  33     def _real_extract(self, url):
  34         token_id = self._match_id(url)
  35         webpage = self._download_webpage(url, token_id)
  36         data_json = self._search_regex(r'var\s+VIDEO_CONFIG\s*=\s*(\{.*)', webpage, 'JSON data')
  37         data = self._parse_json(data_json.encode('utf-8').decode('unicode_escape'), token_id)
  38         streams = data['streams']
  39         formats = [{
  40             'ext': mimetype2ext(traverse_obj(parse_qs(stream['play_url']), ('mime', 0))),
  41             'url': stream['play_url'],
  42             'format_id': str_or_none(stream.get('format_id')),
  43         } for stream in streams]
  44
  45         return {
  46             'id': data.get('iframe_id', token_id),
  47             'title': data.get('iframe_id', token_id),
  48             'formats': formats,
  49             'thumbnail': data.get('thumbnail'),
  50             'duration': parse_duration(traverse_obj(parse_qs(streams[0]['play_url']), ('dur', 0))),
  51         }