yt_dlp/extractor/malltv.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     clean_html,
   4     dict_get,
   5     float_or_none,
   6     int_or_none,
   7     merge_dicts,
   8     parse_duration,
   9     try_get,
  10 )
  11
  12
  13 class MallTVIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:(?:www|sk)\.)?mall\.tv/(?:[^/]+/)*(?P<id>[^/?#&]+)'
  15     _TESTS = [{
  16         'url': 'https://www.mall.tv/18-miliard-pro-neziskovky-opravdu-jsou-sportovci-nebo-clovek-v-tisni-pijavice',
  17         'md5': '1c4a37f080e1f3023103a7b43458e518',
  18         'info_dict': {
  19             'id': 't0zzt0',
  20             'display_id': '18-miliard-pro-neziskovky-opravdu-jsou-sportovci-nebo-clovek-v-tisni-pijavice',
  21             'ext': 'mp4',
  22             'title': '18 miliard pro neziskovky. Opravdu jsou sportovci nebo Člověk v tísni pijavice?',
  23             'description': 'md5:db7d5744a4bd4043d9d98324aa72ab35',
  24             'duration': 216,
  25             'timestamp': 1538870400,
  26             'upload_date': '20181007',
  27             'view_count': int,
  28         }
  29     }, {
  30         'url': 'https://www.mall.tv/kdo-to-plati/18-miliard-pro-neziskovky-opravdu-jsou-sportovci-nebo-clovek-v-tisni-pijavice',
  31         'only_matching': True,
  32     }, {
  33         'url': 'https://sk.mall.tv/gejmhaus/reklamacia-nehreje-vyrobnik-tepla-alebo-spekacka',
  34         'only_matching': True,
  35     }]
  36
  37     def _real_extract(self, url):
  38         display_id = self._match_id(url)
  39
  40         webpage = self._download_webpage(
  41             url, display_id, headers=self.geo_verification_headers())
  42
  43         video = self._parse_json(self._search_regex(
  44             r'videoObject\s*=\s*JSON\.parse\(JSON\.stringify\(({.+?})\)\);',
  45             webpage, 'video object'), display_id)
  46         video_source = video['VideoSource']
  47         video_id = self._search_regex(
  48             r'/([\da-z]+)/index\b', video_source, 'video id')
  49
  50         formats = self._extract_m3u8_formats(
  51             video_source + '.m3u8', video_id, 'mp4', 'm3u8_native')
  52         self._sort_formats(formats)
  53
  54         subtitles = {}
  55         for s in (video.get('Subtitles') or {}):
  56             s_url = s.get('Url')
  57             if not s_url:
  58                 continue
  59             subtitles.setdefault(s.get('Language') or 'cz', []).append({
  60                 'url': s_url,
  61             })
  62
  63         entity_counts = video.get('EntityCounts') or {}
  64
  65         def get_count(k):
  66             v = entity_counts.get(k + 's') or {}
  67             return int_or_none(dict_get(v, ('Count', 'StrCount')))
  68
  69         info = self._search_json_ld(webpage, video_id, default={})
  70
  71         return merge_dicts({
  72             'id': video_id,
  73             'display_id': display_id,
  74             'title': video.get('Title'),
  75             'description': clean_html(video.get('Description')),
  76             'thumbnail': video.get('ThumbnailUrl'),
  77             'formats': formats,
  78             'subtitles': subtitles,
  79             'duration': int_or_none(video.get('DurationSeconds')) or parse_duration(video.get('Duration')),
  80             'view_count': get_count('View'),
  81             'like_count': get_count('Like'),
  82             'dislike_count': get_count('Dislike'),
  83             'average_rating': float_or_none(try_get(video, lambda x: x['EntityRating']['AvarageRate'])),
  84             'comment_count': get_count('Comment'),
  85         }, info)