yt_dlp/extractor/dailymail.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..compat import compat_str
   5 from ..utils import (
   6     int_or_none,
   7     determine_protocol,
   8     try_get,
   9     unescapeHTML,
  10 )
  11
  12
  13 class DailyMailIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/(?:video/[^/]+/video-|embed/video/)(?P<id>[0-9]+)'
  15     _TESTS = [{
  16         'url': 'http://www.dailymail.co.uk/video/tvshowbiz/video-1295863/The-Mountain-appears-sparkling-water-ad-Heavy-Bubbles.html',
  17         'md5': 'f6129624562251f628296c3a9ffde124',
  18         'info_dict': {
  19             'id': '1295863',
  20             'ext': 'mp4',
  21             'title': 'The Mountain appears in sparkling water ad for \'Heavy Bubbles\'',
  22             'description': 'md5:a93d74b6da172dd5dc4d973e0b766a84',
  23         }
  24     }, {
  25         'url': 'http://www.dailymail.co.uk/embed/video/1295863.html',
  26         'only_matching': True,
  27     }]
  28
  29     @staticmethod
  30     def _extract_urls(webpage):
  31         return re.findall(
  32             r'<iframe\b[^>]+\bsrc=["\'](?P<url>(?:https?:)?//(?:www\.)?dailymail\.co\.uk/embed/video/\d+\.html)',
  33             webpage)
  34
  35     def _real_extract(self, url):
  36         video_id = self._match_id(url)
  37         webpage = self._download_webpage(url, video_id)
  38         video_data = self._parse_json(self._search_regex(
  39             r"data-opts='({.+?})'", webpage, 'video data'), video_id)
  40         title = unescapeHTML(video_data['title'])
  41
  42         sources_url = (try_get(
  43             video_data,
  44             (lambda x: x['plugins']['sources']['url'],
  45              lambda x: x['sources']['url']), compat_str)
  46             or 'http://www.dailymail.co.uk/api/player/%s/video-sources.json' % video_id)
  47
  48         video_sources = self._download_json(sources_url, video_id)
  49         body = video_sources.get('body')
  50         if body:
  51             video_sources = body
  52
  53         formats = []
  54         for rendition in video_sources['renditions']:
  55             rendition_url = rendition.get('url')
  56             if not rendition_url:
  57                 continue
  58             tbr = int_or_none(rendition.get('encodingRate'), 1000)
  59             container = rendition.get('videoContainer')
  60             is_hls = container == 'M2TS'
  61             protocol = 'm3u8_native' if is_hls else determine_protocol({'url': rendition_url})
  62             formats.append({
  63                 'format_id': ('hls' if is_hls else protocol) + ('-%d' % tbr if tbr else ''),
  64                 'url': rendition_url,
  65                 'width': int_or_none(rendition.get('frameWidth')),
  66                 'height': int_or_none(rendition.get('frameHeight')),
  67                 'tbr': tbr,
  68                 'vcodec': rendition.get('videoCodec'),
  69                 'container': container,
  70                 'protocol': protocol,
  71                 'ext': 'mp4' if is_hls else None,
  72             })
  73         self._sort_formats(formats)
  74
  75         return {
  76             'id': video_id,
  77             'title': title,
  78             'description': unescapeHTML(video_data.get('descr')),
  79             'thumbnail': video_data.get('poster') or video_data.get('thumbnail'),
  80             'formats': formats,
  81         }