Raphaël G. Git Repositories - youtubedl/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     unescapeHTML,
  16     urlencode_postdata,
  17     USER_AGENTS,
  18 )
  19
  20
  21 class CeskaTelevizeIE(InfoExtractor):
  22     _VALID_URL = r'https?://(?:www\.)?ceskatelevize\.cz/ivysilani/(?:[^/?#&]+/)*(?P<id>[^/#?]+)'
  23     _TESTS = [{
  24         'url': 'http://www.ceskatelevize.cz/ivysilani/ivysilani/10441294653-hyde-park-civilizace/214411058091220',
  25         'info_dict': {
  26             'id': '61924494877246241',
  27             'ext': 'mp4',
  28             'title': 'Hyde Park Civilizace: Život v Grónsku',
  29             'description': 'md5:3fec8f6bb497be5cdb0c9e8781076626',
  30             'thumbnail': r're:^https?://.*\.jpg',
  31             'duration': 3350,
  32         },
  33         'params': {
  34             # m3u8 download
  35             'skip_download': True,
  36         },
  37     }, {
  38         'url': 'http://www.ceskatelevize.cz/ivysilani/10441294653-hyde-park-civilizace/215411058090502/bonus/20641-bonus-01-en',
  39         'info_dict': {
  40             'id': '61924494877028507',
  41             'ext': 'mp4',
  42             'title': 'Hyde Park Civilizace: Bonus 01 - En',
  43             'description': 'English Subtittles',
  44             'thumbnail': r're:^https?://.*\.jpg',
  45             'duration': 81.3,
  46         },
  47         'params': {
  48             # m3u8 download
  49             'skip_download': True,
  50         },
  51     }, {
  52         # live stream
  53         'url': 'http://www.ceskatelevize.cz/ivysilani/zive/ct4/',
  54         'info_dict': {
  55             'id': 402,
  56             'ext': 'mp4',
  57             'title': r're:^ČT Sport \d{4}-\d{2}-\d{2} \d{2}:\d{2}$',
  58             'is_live': True,
  59         },
  60         'params': {
  61             # m3u8 download
  62             'skip_download': True,
  63         },
  64         'skip': 'Georestricted to Czech Republic',
  65     }, {
  66         'url': 'http://www.ceskatelevize.cz/ivysilani/embed/iFramePlayer.php?hash=d6a3e1370d2e4fa76296b90bad4dfc19673b641e&IDEC=217 562 22150/0004&channelID=1&width=100%25',
  67         'only_matching': True,
  68     }]
  69
  70     def _real_extract(self, url):
  71         playlist_id = self._match_id(url)
  72
  73         webpage = self._download_webpage(url, playlist_id)
  74
  75         NOT_AVAILABLE_STRING = 'This content is not available at your territory due to limited copyright.'
  76         if '%s</p>' % NOT_AVAILABLE_STRING in webpage:
  77             raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
  78
  79         type_ = None
  80         episode_id = None
  81
  82         playlist = self._parse_json(
  83             self._search_regex(
  84                 r'getPlaylistUrl\(\[({.+?})\]', webpage, 'playlist',
  85                 default='{}'), playlist_id)
  86         if playlist:
  87             type_ = playlist.get('type')
  88             episode_id = playlist.get('id')
  89
  90         if not type_:
  91             type_ = self._html_search_regex(
  92                 r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],',
  93                 webpage, 'type')
  94         if not episode_id:
  95             episode_id = self._html_search_regex(
  96                 r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],',
  97                 webpage, 'episode_id')
  98
  99         data = {
 100             'playlist[0][type]': type_,
 101             'playlist[0][id]': episode_id,
 102             'requestUrl': compat_urllib_parse_urlparse(url).path,
 103             'requestSource': 'iVysilani',
 104         }
 105
 106         entries = []
 107
 108         for user_agent in (None, USER_AGENTS['Safari']):
 109             req = sanitized_Request(
 110                 'http://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist',
 111                 data=urlencode_postdata(data))
 112
 113             req.add_header('Content-type', 'application/x-www-form-urlencoded')
 114             req.add_header('x-addr', '127.0.0.1')
 115             req.add_header('X-Requested-With', 'XMLHttpRequest')
 116             if user_agent:
 117                 req.add_header('User-Agent', user_agent)
 118             req.add_header('Referer', url)
 119
 120             playlistpage = self._download_json(req, playlist_id, fatal=False)
 121
 122             if not playlistpage:
 123                 continue
 124
 125             playlist_url = playlistpage['url']
 126             if playlist_url == 'error_region':
 127                 raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
 128
 129             req = sanitized_Request(compat_urllib_parse_unquote(playlist_url))
 130             req.add_header('Referer', url)
 131
 132             playlist_title = self._og_search_title(webpage, default=None)
 133             playlist_description = self._og_search_description(webpage, default=None)
 134
 135             playlist = self._download_json(req, playlist_id, fatal=False)
 136             if not playlist:
 137                 continue
 138
 139             playlist = playlist.get('playlist')
 140             if not isinstance(playlist, list):
 141                 continue
 142
 143             playlist_len = len(playlist)
 144
 145             for num, item in enumerate(playlist):
 146                 is_live = item.get('type') == 'LIVE'
 147                 formats = []
 148                 for format_id, stream_url in item.get('streamUrls', {}).items():
 149                     if 'playerType=flash' in stream_url:
 150                         stream_formats = self._extract_m3u8_formats(
 151                             stream_url, playlist_id, 'mp4', 'm3u8_native',
 152                             m3u8_id='hls-%s' % format_id, fatal=False)
 153                     else:
 154                         stream_formats = self._extract_mpd_formats(
 155                             stream_url, playlist_id,
 156                             mpd_id='dash-%s' % format_id, fatal=False)
 157                     # See https://github.com/rg3/youtube-dl/issues/12119#issuecomment-280037031
 158                     if format_id == 'audioDescription':
 159                         for f in stream_formats:
 160                             f['source_preference'] = -10
 161                     formats.extend(stream_formats)
 162
 163                 if user_agent and len(entries) == playlist_len:
 164                     entries[num]['formats'].extend(formats)
 165                     continue
 166
 167                 item_id = item.get('id') or item['assetId']
 168                 title = item['title']
 169
 170                 duration = float_or_none(item.get('duration'))
 171                 thumbnail = item.get('previewImageUrl')
 172
 173                 subtitles = {}
 174                 if item.get('type') == 'VOD':
 175                     subs = item.get('subtitles')
 176                     if subs:
 177                         subtitles = self.extract_subtitles(episode_id, subs)
 178
 179                 if playlist_len == 1:
 180                     final_title = playlist_title or title
 181                     if is_live:
 182                         final_title = self._live_title(final_title)
 183                 else:
 184                     final_title = '%s (%s)' % (playlist_title, title)
 185
 186                 entries.append({
 187                     'id': item_id,
 188                     'title': final_title,
 189                     'description': playlist_description if playlist_len == 1 else None,
 190                     'thumbnail': thumbnail,
 191                     'duration': duration,
 192                     'formats': formats,
 193                     'subtitles': subtitles,
 194                     'is_live': is_live,
 195                 })
 196
 197         for e in entries:
 198             self._sort_formats(e['formats'])
 199
 200         return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
 201
 202     def _get_subtitles(self, episode_id, subs):
 203         original_subtitles = self._download_webpage(
 204             subs[0]['url'], episode_id, 'Downloading subtitles')
 205         srt_subs = self._fix_subtitles(original_subtitles)
 206         return {
 207             'cs': [{
 208                 'ext': 'srt',
 209                 'data': srt_subs,
 210             }]
 211         }
 212
 213     @staticmethod
 214     def _fix_subtitles(subtitles):
 215         """ Convert millisecond-based subtitles to SRT """
 216
 217         def _msectotimecode(msec):
 218             """ Helper utility to convert milliseconds to timecode """
 219             components = []
 220             for divider in [1000, 60, 60, 100]:
 221                 components.append(msec % divider)
 222                 msec //= divider
 223             return '{3:02}:{2:02}:{1:02},{0:03}'.format(*components)
 224
 225         def _fix_subtitle(subtitle):
 226             for line in subtitle.splitlines():
 227                 m = re.match(r'^\s*([0-9]+);\s*([0-9]+)\s+([0-9]+)\s*$', line)
 228                 if m:
 229                     yield m.group(1)
 230                     start, stop = (_msectotimecode(int(t)) for t in m.groups()[1:])
 231                     yield '{0} --> {1}'.format(start, stop)
 232                 else:
 233                     yield line
 234
 235         return '\r\n'.join(_fix_subtitle(subtitles))
 236
 237
 238 class CeskaTelevizePoradyIE(InfoExtractor):
 239     _VALID_URL = r'https?://(?:www\.)?ceskatelevize\.cz/porady/(?:[^/?#&]+/)*(?P<id>[^/#?]+)'
 240     _TESTS = [{
 241         # video with 18+ caution trailer
 242         'url': 'http://www.ceskatelevize.cz/porady/10520528904-queer/215562210900007-bogotart/',
 243         'info_dict': {
 244             'id': '215562210900007-bogotart',
 245             'title': 'Queer: Bogotart',
 246             'description': 'Alternativní průvodce současným queer světem',
 247         },
 248         'playlist': [{
 249             'info_dict': {
 250                 'id': '61924494876844842',
 251                 'ext': 'mp4',
 252                 'title': 'Queer: Bogotart (Varování 18+)',
 253                 'duration': 10.2,
 254             },
 255         }, {
 256             'info_dict': {
 257                 'id': '61924494877068022',
 258                 'ext': 'mp4',
 259                 'title': 'Queer: Bogotart (Queer)',
 260                 'thumbnail': r're:^https?://.*\.jpg',
 261                 'duration': 1558.3,
 262             },
 263         }],
 264         'params': {
 265             # m3u8 download
 266             'skip_download': True,
 267         },
 268     }]
 269
 270     def _real_extract(self, url):
 271         video_id = self._match_id(url)
 272
 273         webpage = self._download_webpage(url, video_id)
 274
 275         data_url = unescapeHTML(self._search_regex(
 276             r'<span[^>]*\bdata-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
 277             webpage, 'iframe player url', group='url'))
 278
 279         return self.url_result(data_url, ie=CeskaTelevizeIE.ie_key())