yt_dlp/extractor/dailymail.py

   1 from .common import InfoExtractor
   2 from ..compat import compat_str
   3 from ..utils import (
   4     determine_protocol,
   5     int_or_none,
   6     try_get,
   7     unescapeHTML,
   8 )
   9
  10
  11 class DailyMailIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/(?:video/[^/]+/video-|embed/video/)(?P<id>[0-9]+)'
  13     _EMBED_REGEX = [r'<iframe\b[^>]+\bsrc=["\'](?P<url>(?:https?:)?//(?:www\.)?dailymail\.co\.uk/embed/video/\d+\.html)']
  14     _TESTS = [{
  15         'url': 'http://www.dailymail.co.uk/video/tvshowbiz/video-1295863/The-Mountain-appears-sparkling-water-ad-Heavy-Bubbles.html',
  16         'md5': 'f6129624562251f628296c3a9ffde124',
  17         'info_dict': {
  18             'id': '1295863',
  19             'ext': 'mp4',
  20             'title': 'The Mountain appears in sparkling water ad for \'Heavy Bubbles\'',
  21             'description': 'md5:a93d74b6da172dd5dc4d973e0b766a84',
  22         }
  23     }, {
  24         'url': 'http://www.dailymail.co.uk/embed/video/1295863.html',
  25         'only_matching': True,
  26     }]
  27
  28     def _real_extract(self, url):
  29         video_id = self._match_id(url)
  30         webpage = self._download_webpage(url, video_id)
  31         video_data = self._parse_json(self._search_regex(
  32             r"data-opts='({.+?})'", webpage, 'video data'), video_id)
  33         title = unescapeHTML(video_data['title'])
  34
  35         sources_url = (try_get(
  36             video_data,
  37             (lambda x: x['plugins']['sources']['url'],
  38              lambda x: x['sources']['url']), compat_str)
  39             or 'http://www.dailymail.co.uk/api/player/%s/video-sources.json' % video_id)
  40
  41         video_sources = self._download_json(sources_url, video_id)
  42         body = video_sources.get('body')
  43         if body:
  44             video_sources = body
  45
  46         formats = []
  47         for rendition in video_sources['renditions']:
  48             rendition_url = rendition.get('url')
  49             if not rendition_url:
  50                 continue
  51             tbr = int_or_none(rendition.get('encodingRate'), 1000)
  52             container = rendition.get('videoContainer')
  53             is_hls = container == 'M2TS'
  54             protocol = 'm3u8_native' if is_hls else determine_protocol({'url': rendition_url})
  55             formats.append({
  56                 'format_id': ('hls' if is_hls else protocol) + ('-%d' % tbr if tbr else ''),
  57                 'url': rendition_url,
  58                 'width': int_or_none(rendition.get('frameWidth')),
  59                 'height': int_or_none(rendition.get('frameHeight')),
  60                 'tbr': tbr,
  61                 'vcodec': rendition.get('videoCodec'),
  62                 'container': container,
  63                 'protocol': protocol,
  64                 'ext': 'mp4' if is_hls else None,
  65             })
  66
  67         return {
  68             'id': video_id,
  69             'title': title,
  70             'description': unescapeHTML(video_data.get('descr')),
  71             'thumbnail': video_data.get('poster') or video_data.get('thumbnail'),
  72             'formats': formats,
  73         }