yt_dlp/extractor/dailymail.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     determine_protocol,
   4     int_or_none,
   5     try_get,
   6     unescapeHTML,
   7 )
   8
   9
  10 class DailyMailIE(InfoExtractor):
  11     _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/(?:video/[^/]+/video-|embed/video/)(?P<id>[0-9]+)'
  12     _EMBED_REGEX = [r'<iframe\b[^>]+\bsrc=["\'](?P<url>(?:https?:)?//(?:www\.)?dailymail\.co\.uk/embed/video/\d+\.html)']
  13     _TESTS = [{
  14         'url': 'http://www.dailymail.co.uk/video/tvshowbiz/video-1295863/The-Mountain-appears-sparkling-water-ad-Heavy-Bubbles.html',
  15         'md5': 'f6129624562251f628296c3a9ffde124',
  16         'info_dict': {
  17             'id': '1295863',
  18             'ext': 'mp4',
  19             'title': 'The Mountain appears in sparkling water ad for \'Heavy Bubbles\'',
  20             'description': 'md5:a93d74b6da172dd5dc4d973e0b766a84',
  21         },
  22     }, {
  23         'url': 'http://www.dailymail.co.uk/embed/video/1295863.html',
  24         'only_matching': True,
  25     }]
  26
  27     def _real_extract(self, url):
  28         video_id = self._match_id(url)
  29         webpage = self._download_webpage(url, video_id)
  30         video_data = self._parse_json(self._search_regex(
  31             r"data-opts='({.+?})'", webpage, 'video data'), video_id)
  32         title = unescapeHTML(video_data['title'])
  33
  34         sources_url = (try_get(
  35             video_data,
  36             (lambda x: x['plugins']['sources']['url'],
  37              lambda x: x['sources']['url']), str)
  38             or f'http://www.dailymail.co.uk/api/player/{video_id}/video-sources.json')
  39
  40         video_sources = self._download_json(sources_url, video_id)
  41         body = video_sources.get('body')
  42         if body:
  43             video_sources = body
  44
  45         formats = []
  46         for rendition in video_sources['renditions']:
  47             rendition_url = rendition.get('url')
  48             if not rendition_url:
  49                 continue
  50             tbr = int_or_none(rendition.get('encodingRate'), 1000)
  51             container = rendition.get('videoContainer')
  52             is_hls = container == 'M2TS'
  53             protocol = 'm3u8_native' if is_hls else determine_protocol({'url': rendition_url})
  54             formats.append({
  55                 'format_id': ('hls' if is_hls else protocol) + (f'-{tbr}' if tbr else ''),
  56                 'url': rendition_url,
  57                 'width': int_or_none(rendition.get('frameWidth')),
  58                 'height': int_or_none(rendition.get('frameHeight')),
  59                 'tbr': tbr,
  60                 'vcodec': rendition.get('videoCodec'),
  61                 'container': container,
  62                 'protocol': protocol,
  63                 'ext': 'mp4' if is_hls else None,
  64             })
  65
  66         return {
  67             'id': video_id,
  68             'title': title,
  69             'description': unescapeHTML(video_data.get('descr')),
  70             'thumbnail': video_data.get('poster') or video_data.get('thumbnail'),
  71             'formats': formats,
  72         }