yt_dlp/extractor/ceskatelevize.py

   1 import re
   2 import urllib.parse
   3
   4 from .common import InfoExtractor
   5 from ..networking import Request
   6 from ..utils import (
   7     ExtractorError,
   8     float_or_none,
   9     str_or_none,
  10     traverse_obj,
  11     urlencode_postdata,
  12 )
  13
  14 USER_AGENTS = {
  15     'Safari': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) AppleWebKit/533.20.25 (KHTML, like Gecko) Version/5.0.4 Safari/533.20.27',
  16 }
  17
  18
  19 class CeskaTelevizeIE(InfoExtractor):
  20     _VALID_URL = r'https?://(?:www\.)?ceskatelevize\.cz/(?:ivysilani|porady|zive)/(?:[^/?#&]+/)*(?P<id>[^/#?]+)'
  21     _TESTS = [{
  22         'url': 'http://www.ceskatelevize.cz/ivysilani/10441294653-hyde-park-civilizace/215411058090502/bonus/20641-bonus-01-en',
  23         'info_dict': {
  24             'id': '61924494877028507',
  25             'ext': 'mp4',
  26             'title': 'Bonus 01 - En - Hyde Park Civilizace',
  27             'description': 'English Subtittles',
  28             'thumbnail': r're:^https?://.*\.jpg',
  29             'duration': 81.3,
  30         },
  31         'params': {
  32             # m3u8 download
  33             'skip_download': True,
  34         },
  35     }, {
  36         # live stream
  37         'url': 'http://www.ceskatelevize.cz/zive/ct1/',
  38         'info_dict': {
  39             'id': '102',
  40             'ext': 'mp4',
  41             'title': r'ČT1 - živé vysílání online',
  42             'description': 'Sledujte živé vysílání kanálu ČT1 online. Vybírat si můžete i z dalších kanálů České televize na kterémkoli z vašich zařízení.',
  43             'is_live': True,
  44         },
  45         'params': {
  46             # m3u8 download
  47             'skip_download': True,
  48         },
  49     }, {
  50         # another
  51         'url': 'http://www.ceskatelevize.cz/ivysilani/zive/ct4/',
  52         'only_matching': True,
  53         'info_dict': {
  54             'id': '402',
  55             'ext': 'mp4',
  56             'title': r're:^ČT Sport \d{4}-\d{2}-\d{2} \d{2}:\d{2}$',
  57             'is_live': True,
  58         },
  59         # 'skip': 'Georestricted to Czech Republic',
  60     }, {
  61         'url': 'http://www.ceskatelevize.cz/ivysilani/embed/iFramePlayer.php?hash=d6a3e1370d2e4fa76296b90bad4dfc19673b641e&IDEC=217 562 22150/0004&channelID=1&width=100%25',
  62         'only_matching': True,
  63     }, {
  64         # video with 18+ caution trailer
  65         'url': 'http://www.ceskatelevize.cz/porady/10520528904-queer/215562210900007-bogotart/',
  66         'info_dict': {
  67             'id': '215562210900007-bogotart',
  68             'title': 'Bogotart - Queer',
  69             'description': 'Hlavní město Kolumbie v doprovodu queer umělců. Vroucí svět plný vášně, sebevědomí, ale i násilí a bolesti',
  70         },
  71         'playlist': [{
  72             'info_dict': {
  73                 'id': '61924494877311053',
  74                 'ext': 'mp4',
  75                 'title': 'Bogotart - Queer (Varování 18+)',
  76                 'duration': 11.9,
  77             },
  78         }, {
  79             'info_dict': {
  80                 'id': '61924494877068022',
  81                 'ext': 'mp4',
  82                 'title': 'Bogotart - Queer (Queer)',
  83                 'thumbnail': r're:^https?://.*\.jpg',
  84                 'duration': 1558.3,
  85             },
  86         }],
  87         'params': {
  88             # m3u8 download
  89             'skip_download': True,
  90         },
  91     }, {
  92         # iframe embed
  93         'url': 'http://www.ceskatelevize.cz/porady/10614999031-neviditelni/21251212048/',
  94         'only_matching': True,
  95     }]
  96
  97     def _real_extract(self, url):
  98         playlist_id = self._match_id(url)
  99         webpage, urlh = self._download_webpage_handle(url, playlist_id)
 100         parsed_url = urllib.parse.urlparse(urlh.url)
 101         site_name = self._og_search_property('site_name', webpage, fatal=False, default='Česká televize')
 102         playlist_title = self._og_search_title(webpage, default=None)
 103         if site_name and playlist_title:
 104             playlist_title = re.split(rf'\s*[—|]\s*{site_name}', playlist_title, maxsplit=1)[0]
 105         playlist_description = self._og_search_description(webpage, default=None)
 106         if playlist_description:
 107             playlist_description = playlist_description.replace('\xa0', ' ')
 108
 109         type_ = 'IDEC'
 110         if re.search(r'(^/porady|/zive)/', parsed_url.path):
 111             next_data = self._search_nextjs_data(webpage, playlist_id)
 112             if '/zive/' in parsed_url.path:
 113                 idec = traverse_obj(next_data, ('props', 'pageProps', 'data', 'liveBroadcast', 'current', 'idec'), get_all=False)
 114             else:
 115                 idec = traverse_obj(next_data, ('props', 'pageProps', 'data', ('show', 'mediaMeta'), 'idec'), get_all=False)
 116                 if not idec:
 117                     idec = traverse_obj(next_data, ('props', 'pageProps', 'data', 'videobonusDetail', 'bonusId'), get_all=False)
 118                     if idec:
 119                         type_ = 'bonus'
 120             if not idec:
 121                 raise ExtractorError('Failed to find IDEC id')
 122             iframe_hash = self._download_webpage(
 123                 'https://www.ceskatelevize.cz/v-api/iframe-hash/',
 124                 playlist_id, note='Getting IFRAME hash')
 125             query = {'hash': iframe_hash, 'origin': 'iVysilani', 'autoStart': 'true', type_: idec}
 126             webpage = self._download_webpage(
 127                 'https://www.ceskatelevize.cz/ivysilani/embed/iFramePlayer.php',
 128                 playlist_id, note='Downloading player', query=query)
 129
 130         NOT_AVAILABLE_STRING = 'This content is not available at your territory due to limited copyright.'
 131         if f'{NOT_AVAILABLE_STRING}</p>' in webpage:
 132             self.raise_geo_restricted(NOT_AVAILABLE_STRING)
 133         if any(not_found in webpage for not_found in ('Neplatný parametr pro videopřehrávač', 'IDEC nebyl nalezen')):
 134             raise ExtractorError('no video with IDEC available', video_id=idec, expected=True)
 135
 136         type_ = None
 137         episode_id = None
 138
 139         playlist = self._parse_json(
 140             self._search_regex(
 141                 r'getPlaylistUrl\(\[({.+?})\]', webpage, 'playlist',
 142                 default='{}'), playlist_id)
 143         if playlist:
 144             type_ = playlist.get('type')
 145             episode_id = playlist.get('id')
 146
 147         if not type_:
 148             type_ = self._html_search_regex(
 149                 r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],',
 150                 webpage, 'type')
 151         if not episode_id:
 152             episode_id = self._html_search_regex(
 153                 r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],',
 154                 webpage, 'episode_id')
 155
 156         data = {
 157             'playlist[0][type]': type_,
 158             'playlist[0][id]': episode_id,
 159             'requestUrl': parsed_url.path,
 160             'requestSource': 'iVysilani',
 161         }
 162
 163         entries = []
 164
 165         for user_agent in (None, USER_AGENTS['Safari']):
 166             req = Request(
 167                 'https://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist/',
 168                 data=urlencode_postdata(data))
 169
 170             req.headers['Content-type'] = 'application/x-www-form-urlencoded'
 171             req.headers['x-addr'] = '127.0.0.1'
 172             req.headers['X-Requested-With'] = 'XMLHttpRequest'
 173             if user_agent:
 174                 req.headers['User-Agent'] = user_agent
 175             req.headers['Referer'] = url
 176
 177             playlistpage = self._download_json(req, playlist_id, fatal=False)
 178
 179             if not playlistpage:
 180                 continue
 181
 182             playlist_url = playlistpage['url']
 183             if playlist_url == 'error_region':
 184                 raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
 185
 186             req = Request(urllib.parse.unquote(playlist_url))
 187             req.headers['Referer'] = url
 188
 189             playlist = self._download_json(req, playlist_id, fatal=False)
 190             if not playlist:
 191                 continue
 192
 193             playlist = playlist.get('playlist')
 194             if not isinstance(playlist, list):
 195                 continue
 196
 197             playlist_len = len(playlist)
 198
 199             for num, item in enumerate(playlist):
 200                 is_live = item.get('type') == 'LIVE'
 201                 formats = []
 202                 for format_id, stream_url in item.get('streamUrls', {}).items():
 203                     if 'playerType=flash' in stream_url:
 204                         stream_formats = self._extract_m3u8_formats(
 205                             stream_url, playlist_id, 'mp4', 'm3u8_native',
 206                             m3u8_id=f'hls-{format_id}', fatal=False)
 207                     else:
 208                         stream_formats = self._extract_mpd_formats(
 209                             stream_url, playlist_id,
 210                             mpd_id=f'dash-{format_id}', fatal=False)
 211                     if 'drmOnly=true' in stream_url:
 212                         for f in stream_formats:
 213                             f['has_drm'] = True
 214                     # See https://github.com/ytdl-org/youtube-dl/issues/12119#issuecomment-280037031
 215                     if format_id == 'audioDescription':
 216                         for f in stream_formats:
 217                             f['source_preference'] = -10
 218                     formats.extend(stream_formats)
 219
 220                 if user_agent and len(entries) == playlist_len:
 221                     entries[num]['formats'].extend(formats)
 222                     continue
 223
 224                 item_id = str_or_none(item.get('id') or item['assetId'])
 225                 title = item['title']
 226
 227                 duration = float_or_none(item.get('duration'))
 228                 thumbnail = item.get('previewImageUrl')
 229
 230                 subtitles = {}
 231                 if item.get('type') == 'VOD':
 232                     subs = item.get('subtitles')
 233                     if subs:
 234                         subtitles = self.extract_subtitles(episode_id, subs)
 235
 236                 if playlist_len == 1:
 237                     final_title = playlist_title or title
 238                 else:
 239                     final_title = f'{playlist_title} ({title})'
 240
 241                 entries.append({
 242                     'id': item_id,
 243                     'title': final_title,
 244                     'description': playlist_description if playlist_len == 1 else None,
 245                     'thumbnail': thumbnail,
 246                     'duration': duration,
 247                     'formats': formats,
 248                     'subtitles': subtitles,
 249                     'is_live': is_live,
 250                 })
 251
 252         if len(entries) == 1:
 253             return entries[0]
 254         return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
 255
 256     def _get_subtitles(self, episode_id, subs):
 257         original_subtitles = self._download_webpage(
 258             subs[0]['url'], episode_id, 'Downloading subtitles')
 259         srt_subs = self._fix_subtitles(original_subtitles)
 260         return {
 261             'cs': [{
 262                 'ext': 'srt',
 263                 'data': srt_subs,
 264             }],
 265         }
 266
 267     @staticmethod
 268     def _fix_subtitles(subtitles):
 269         """ Convert millisecond-based subtitles to SRT """
 270
 271         def _msectotimecode(msec):
 272             """ Helper utility to convert milliseconds to timecode """
 273             components = []
 274             for divider in [1000, 60, 60, 100]:
 275                 components.append(msec % divider)
 276                 msec //= divider
 277             return '{3:02}:{2:02}:{1:02},{0:03}'.format(*components)
 278
 279         def _fix_subtitle(subtitle):
 280             for line in subtitle.splitlines():
 281                 m = re.match(r'^\s*([0-9]+);\s*([0-9]+)\s+([0-9]+)\s*$', line)
 282                 if m:
 283                     yield m.group(1)
 284                     start, stop = (_msectotimecode(int(t)) for t in m.groups()[1:])
 285                     yield f'{start} --> {stop}'
 286                 else:
 287                     yield line
 288
 289         return '\r\n'.join(_fix_subtitle(subtitles))