_ Git - youtube-dl/blob - youtube_dl/extractor/ceskatelevize.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_urllib_parse_unquote,
   9     compat_urllib_parse_urlparse,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     float_or_none,
  14     sanitized_Request,
  15     urlencode_postdata,
  16     USER_AGENTS,
  17 )
  18
  19
  20 class CeskaTelevizeIE(InfoExtractor):
  21     _VALID_URL = r'https?://(?:www\.)?ceskatelevize\.cz/ivysilani/(?:[^/?#&]+/)*(?P<id>[^/#?]+)'
  22     _TESTS = [{
  23         'url': 'http://www.ceskatelevize.cz/ivysilani/ivysilani/10441294653-hyde-park-civilizace/214411058091220',
  24         'info_dict': {
  25             'id': '61924494877246241',
  26             'ext': 'mp4',
  27             'title': 'Hyde Park Civilizace: Život v Grónsku',
  28             'description': 'md5:3fec8f6bb497be5cdb0c9e8781076626',
  29             'thumbnail': r're:^https?://.*\.jpg',
  30             'duration': 3350,
  31         },
  32         'params': {
  33             # m3u8 download
  34             'skip_download': True,
  35         },
  36     }, {
  37         'url': 'http://www.ceskatelevize.cz/ivysilani/10441294653-hyde-park-civilizace/215411058090502/bonus/20641-bonus-01-en',
  38         'info_dict': {
  39             'id': '61924494877028507',
  40             'ext': 'mp4',
  41             'title': 'Hyde Park Civilizace: Bonus 01 - En',
  42             'description': 'English Subtittles',
  43             'thumbnail': r're:^https?://.*\.jpg',
  44             'duration': 81.3,
  45         },
  46         'params': {
  47             # m3u8 download
  48             'skip_download': True,
  49         },
  50     }, {
  51         # live stream
  52         'url': 'http://www.ceskatelevize.cz/ivysilani/zive/ct4/',
  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         'params': {
  60             # m3u8 download
  61             'skip_download': True,
  62         },
  63         'skip': 'Georestricted to Czech Republic',
  64     }, {
  65         'url': 'http://www.ceskatelevize.cz/ivysilani/embed/iFramePlayer.php?hash=d6a3e1370d2e4fa76296b90bad4dfc19673b641e&IDEC=217 562 22150/0004&channelID=1&width=100%25',
  66         'only_matching': True,
  67     }]
  68
  69     def _real_extract(self, url):
  70         playlist_id = self._match_id(url)
  71
  72         webpage = self._download_webpage(url, playlist_id)
  73
  74         NOT_AVAILABLE_STRING = 'This content is not available at your territory due to limited copyright.'
  75         if '%s</p>' % NOT_AVAILABLE_STRING in webpage:
  76             raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
  77
  78         type_ = None
  79         episode_id = None
  80
  81         playlist = self._parse_json(
  82             self._search_regex(
  83                 r'getPlaylistUrl\(\[({.+?})\]', webpage, 'playlist',
  84                 default='{}'), playlist_id)
  85         if playlist:
  86             type_ = playlist.get('type')
  87             episode_id = playlist.get('id')
  88
  89         if not type_:
  90             type_ = self._html_search_regex(
  91                 r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],',
  92                 webpage, 'type')
  93         if not episode_id:
  94             episode_id = self._html_search_regex(
  95                 r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],',
  96                 webpage, 'episode_id')
  97
  98         data = {
  99             'playlist[0][type]': type_,
 100             'playlist[0][id]': episode_id,
 101             'requestUrl': compat_urllib_parse_urlparse(url).path,
 102             'requestSource': 'iVysilani',
 103         }
 104
 105         entries = []
 106
 107         for user_agent in (None, USER_AGENTS['Safari']):
 108             req = sanitized_Request(
 109                 'http://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist',
 110                 data=urlencode_postdata(data))
 111
 112             req.add_header('Content-type', 'application/x-www-form-urlencoded')
 113             req.add_header('x-addr', '127.0.0.1')
 114             req.add_header('X-Requested-With', 'XMLHttpRequest')
 115             if user_agent:
 116                 req.add_header('User-Agent', user_agent)
 117             req.add_header('Referer', url)
 118
 119             playlistpage = self._download_json(req, playlist_id, fatal=False)
 120
 121             if not playlistpage:
 122                 continue
 123
 124             playlist_url = playlistpage['url']
 125             if playlist_url == 'error_region':
 126                 raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
 127
 128             req = sanitized_Request(compat_urllib_parse_unquote(playlist_url))
 129             req.add_header('Referer', url)
 130
 131             playlist_title = self._og_search_title(webpage, default=None)
 132             playlist_description = self._og_search_description(webpage, default=None)
 133
 134             playlist = self._download_json(req, playlist_id, fatal=False)
 135             if not playlist:
 136                 continue
 137
 138             playlist = playlist.get('playlist')
 139             if not isinstance(playlist, list):
 140                 continue
 141
 142             playlist_len = len(playlist)
 143
 144             for num, item in enumerate(playlist):
 145                 is_live = item.get('type') == 'LIVE'
 146                 formats = []
 147                 for format_id, stream_url in item.get('streamUrls', {}).items():
 148                     if 'playerType=flash' in stream_url:
 149                         stream_formats = self._extract_m3u8_formats(
 150                             stream_url, playlist_id, 'mp4', 'm3u8_native',
 151                             m3u8_id='hls-%s' % format_id, fatal=False)
 152                     else:
 153                         stream_formats = self._extract_mpd_formats(
 154                             stream_url, playlist_id,
 155                             mpd_id='dash-%s' % format_id, fatal=False)
 156                     # See https://github.com/rg3/youtube-dl/issues/12119#issuecomment-280037031
 157                     if format_id == 'audioDescription':
 158                         for f in stream_formats:
 159                             f['source_preference'] = -10
 160                     formats.extend(stream_formats)
 161
 162                 if user_agent and len(entries) == playlist_len:
 163                     entries[num]['formats'].extend(formats)
 164                     continue
 165
 166                 item_id = item.get('id') or item['assetId']
 167                 title = item['title']
 168
 169                 duration = float_or_none(item.get('duration'))
 170                 thumbnail = item.get('previewImageUrl')
 171
 172                 subtitles = {}
 173                 if item.get('type') == 'VOD':
 174                     subs = item.get('subtitles')
 175                     if subs:
 176                         subtitles = self.extract_subtitles(episode_id, subs)
 177
 178                 if playlist_len == 1:
 179                     final_title = playlist_title or title
 180                     if is_live:
 181                         final_title = self._live_title(final_title)
 182                 else:
 183                     final_title = '%s (%s)' % (playlist_title, title)
 184
 185                 entries.append({
 186                     'id': item_id,
 187                     'title': final_title,
 188                     'description': playlist_description if playlist_len == 1 else None,
 189                     'thumbnail': thumbnail,
 190                     'duration': duration,
 191                     'formats': formats,
 192                     'subtitles': subtitles,
 193                     'is_live': is_live,
 194                 })
 195
 196         for e in entries:
 197             self._sort_formats(e['formats'])
 198
 199         return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
 200
 201     def _get_subtitles(self, episode_id, subs):
 202         original_subtitles = self._download_webpage(
 203             subs[0]['url'], episode_id, 'Downloading subtitles')
 204         srt_subs = self._fix_subtitles(original_subtitles)
 205         return {
 206             'cs': [{
 207                 'ext': 'srt',
 208                 'data': srt_subs,
 209             }]
 210         }
 211
 212     @staticmethod
 213     def _fix_subtitles(subtitles):
 214         """ Convert millisecond-based subtitles to SRT """
 215
 216         def _msectotimecode(msec):
 217             """ Helper utility to convert milliseconds to timecode """
 218             components = []
 219             for divider in [1000, 60, 60, 100]:
 220                 components.append(msec % divider)
 221                 msec //= divider
 222             return '{3:02}:{2:02}:{1:02},{0:03}'.format(*components)
 223
 224         def _fix_subtitle(subtitle):
 225             for line in subtitle.splitlines():
 226                 m = re.match(r'^\s*([0-9]+);\s*([0-9]+)\s+([0-9]+)\s*$', line)
 227                 if m:
 228                     yield m.group(1)
 229                     start, stop = (_msectotimecode(int(t)) for t in m.groups()[1:])
 230                     yield '{0} --> {1}'.format(start, stop)
 231                 else:
 232                     yield line
 233
 234         return '\r\n'.join(_fix_subtitle(subtitles))