yt_dlp/extractor/streamcz.py

   1 import json
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     float_or_none,
   6     int_or_none,
   7     parse_codecs,
   8     traverse_obj,
   9     urljoin,
  10 )
  11
  12
  13 class StreamCZIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?(?:stream|televizeseznam)\.cz/[^?#]+/(?P<display_id>[^?#]+)-(?P<id>[0-9]+)'
  15     _TESTS = [{
  16         'url': 'https://www.televizeseznam.cz/video/lajna/buh-57953890',
  17         'md5': '40c41ade1464a390a0b447e333df4239',
  18         'info_dict': {
  19             'id': '57953890',
  20             'ext': 'mp4',
  21             'title': 'Bůh',
  22             'display_id': 'buh',
  23             'description': 'md5:8f5f09b9b7bc67df910486cdd88f7165',
  24             'duration': 1369.6,
  25             'view_count': int,
  26         }
  27     }, {
  28         'url': 'https://www.stream.cz/kdo-to-mluvi/kdo-to-mluvi-velke-odhaleni-prinasi-novy-porad-uz-od-25-srpna-64087937',
  29         'md5': '41fd358000086a1ccdb068c77809b158',
  30         'info_dict': {
  31             'id': '64087937',
  32             'ext': 'mp4',
  33             'title': 'Kdo to mluví? Velké odhalení přináší nový pořad už od 25. srpna',
  34             'display_id': 'kdo-to-mluvi-velke-odhaleni-prinasi-novy-porad-uz-od-25-srpna',
  35             'description': 'md5:97a811000a6460266029d6c1c2ebcd59',
  36             'duration': 50.2,
  37             'view_count': int,
  38         }
  39     }, {
  40         'url': 'https://www.stream.cz/tajemno/znicehonic-jim-skrz-strechu-prolitnul-zahadny-predmet-badatele-vse-objasnili-64147267',
  41         'md5': '3ee4d0be040e8f4a543e67e509d55e3f',
  42         'info_dict': {
  43             'id': '64147267',
  44             'ext': 'mp4',
  45             'title': 'Zničehonic jim skrz střechu prolítnul záhadný předmět. Badatelé vše objasnili',
  46             'display_id': 'znicehonic-jim-skrz-strechu-prolitnul-zahadny-predmet-badatele-vse-objasnili',
  47             'description': 'md5:4b8ada6718d34bb011c4e04ca4bc19bf',
  48             'duration': 442.84,
  49             'view_count': int,
  50         }
  51     }]
  52
  53     def _extract_formats(self, spl_url, video):
  54         for ext, pref, streams in (
  55                 ('ts', -1, traverse_obj(video, ('http_stream', 'qualities')) or {}),
  56                 ('mp4', 1, video.get('mp4') or {})):
  57             for format_id, stream in streams.items():
  58                 if not stream.get('url'):
  59                     continue
  60                 yield {
  61                     'format_id': f'{format_id}-{ext}',
  62                     'ext': ext,
  63                     'source_preference': pref,
  64                     'url': urljoin(spl_url, stream['url']),
  65                     'tbr': float_or_none(stream.get('bandwidth'), scale=1000),
  66                     'duration': float_or_none(stream.get('duration'), scale=1000),
  67                     'width': traverse_obj(stream, ('resolution', 0)),
  68                     'height': traverse_obj(stream, ('resolution', 1)) or int_or_none(format_id.replace('p', '')),
  69                     **parse_codecs(stream.get('codec')),
  70                 }
  71
  72     def _real_extract(self, url):
  73         display_id, video_id = self._match_valid_url(url).groups()
  74
  75         data = self._download_json(
  76             'https://www.televizeseznam.cz/api/graphql', video_id, 'Downloading GraphQL result',
  77             data=json.dumps({
  78                 'variables': {'urlName': video_id},
  79                 'query': '''
  80                     query LoadEpisode($urlName : String){ episode(urlName: $urlName){ ...VideoDetailFragmentOnEpisode } }
  81                     fragment VideoDetailFragmentOnEpisode on Episode {
  82                         id
  83                         spl
  84                         urlName
  85                         name
  86                         perex
  87                         duration
  88                         views
  89                     }'''
  90             }).encode('utf-8'),
  91             headers={'Content-Type': 'application/json;charset=UTF-8'}
  92         )['data']['episode']
  93
  94         spl_url = data['spl'] + 'spl2,3'
  95         metadata = self._download_json(spl_url, video_id, 'Downloading playlist')
  96         if 'Location' in metadata and 'data' not in metadata:
  97             spl_url = metadata['Location']
  98             metadata = self._download_json(spl_url, video_id, 'Downloading redirected playlist')
  99         video = metadata['data']
 100
 101         subtitles = {}
 102         for subs in video.get('subtitles', {}).values():
 103             if not subs.get('language'):
 104                 continue
 105             for ext, sub_url in subs.get('urls').items():
 106                 subtitles.setdefault(subs['language'], []).append({
 107                     'ext': ext,
 108                     'url': urljoin(spl_url, sub_url)
 109                 })
 110
 111         formats = list(self._extract_formats(spl_url, video))
 112
 113         return {
 114             'id': video_id,
 115             'display_id': display_id,
 116             'title': data.get('name'),
 117             'description': data.get('perex'),
 118             'duration': float_or_none(data.get('duration')),
 119             'view_count': int_or_none(data.get('views')),
 120             'formats': formats,
 121             'subtitles': subtitles,
 122         }