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     update_url_query,
  17     urlencode_postdata,
  18     USER_AGENTS,
  19 )
  20
  21
  22 class CeskaTelevizeIE(InfoExtractor):
  23     _VALID_URL = r'https?://(?:www\.)?ceskatelevize\.cz/ivysilani/(?:[^/?#&]+/)*(?P<id>[^/#?]+)'
  24     _TESTS = [{
  25         'url': 'http://www.ceskatelevize.cz/ivysilani/ivysilani/10441294653-hyde-park-civilizace/214411058091220',
  26         'info_dict': {
  27             'id': '61924494877246241',
  28             'ext': 'mp4',
  29             'title': 'Hyde Park Civilizace: Život v Grónsku',
  30             'description': 'md5:3fec8f6bb497be5cdb0c9e8781076626',
  31             'thumbnail': r're:^https?://.*\.jpg',
  32             'duration': 3350,
  33         },
  34         'params': {
  35             # m3u8 download
  36             'skip_download': True,
  37         },
  38     }, {
  39         'url': 'http://www.ceskatelevize.cz/ivysilani/10441294653-hyde-park-civilizace/215411058090502/bonus/20641-bonus-01-en',
  40         'info_dict': {
  41             'id': '61924494877028507',
  42             'ext': 'mp4',
  43             'title': 'Hyde Park Civilizace: Bonus 01 - En',
  44             'description': 'English Subtittles',
  45             'thumbnail': r're:^https?://.*\.jpg',
  46             'duration': 81.3,
  47         },
  48         'params': {
  49             # m3u8 download
  50             'skip_download': True,
  51         },
  52     }, {
  53         # live stream
  54         'url': 'http://www.ceskatelevize.cz/ivysilani/zive/ct4/',
  55         'info_dict': {
  56             'id': 402,
  57             'ext': 'mp4',
  58             'title': r're:^ČT Sport \d{4}-\d{2}-\d{2} \d{2}:\d{2}$',
  59             'is_live': True,
  60         },
  61         'params': {
  62             # m3u8 download
  63             'skip_download': True,
  64         },
  65         'skip': 'Georestricted to Czech Republic',
  66     }, {
  67         'url': 'http://www.ceskatelevize.cz/ivysilani/embed/iFramePlayer.php?hash=d6a3e1370d2e4fa76296b90bad4dfc19673b641e&IDEC=217 562 22150/0004&channelID=1&width=100%25',
  68         'only_matching': True,
  69     }]
  70
  71     def _real_extract(self, url):
  72         playlist_id = self._match_id(url)
  73
  74         webpage = self._download_webpage(url, playlist_id)
  75
  76         NOT_AVAILABLE_STRING = 'This content is not available at your territory due to limited copyright.'
  77         if '%s</p>' % NOT_AVAILABLE_STRING in webpage:
  78             raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
  79
  80         type_ = None
  81         episode_id = None
  82
  83         playlist = self._parse_json(
  84             self._search_regex(
  85                 r'getPlaylistUrl\(\[({.+?})\]', webpage, 'playlist',
  86                 default='{}'), playlist_id)
  87         if playlist:
  88             type_ = playlist.get('type')
  89             episode_id = playlist.get('id')
  90
  91         if not type_:
  92             type_ = self._html_search_regex(
  93                 r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],',
  94                 webpage, 'type')
  95         if not episode_id:
  96             episode_id = self._html_search_regex(
  97                 r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],',
  98                 webpage, 'episode_id')
  99
 100         data = {
 101             'playlist[0][type]': type_,
 102             'playlist[0][id]': episode_id,
 103             'requestUrl': compat_urllib_parse_urlparse(url).path,
 104             'requestSource': 'iVysilani',
 105         }
 106
 107         entries = []
 108
 109         for user_agent in (None, USER_AGENTS['Safari']):
 110             req = sanitized_Request(
 111                 'http://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist',
 112                 data=urlencode_postdata(data))
 113
 114             req.add_header('Content-type', 'application/x-www-form-urlencoded')
 115             req.add_header('x-addr', '127.0.0.1')
 116             req.add_header('X-Requested-With', 'XMLHttpRequest')
 117             if user_agent:
 118                 req.add_header('User-Agent', user_agent)
 119             req.add_header('Referer', url)
 120
 121             playlistpage = self._download_json(req, playlist_id, fatal=False)
 122
 123             if not playlistpage:
 124                 continue
 125
 126             playlist_url = playlistpage['url']
 127             if playlist_url == 'error_region':
 128                 raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
 129
 130             req = sanitized_Request(compat_urllib_parse_unquote(playlist_url))
 131             req.add_header('Referer', url)
 132
 133             playlist_title = self._og_search_title(webpage, default=None)
 134             playlist_description = self._og_search_description(webpage, default=None)
 135
 136             playlist = self._download_json(req, playlist_id, fatal=False)
 137             if not playlist:
 138                 continue
 139
 140             playlist = playlist.get('playlist')
 141             if not isinstance(playlist, list):
 142                 continue
 143
 144             playlist_len = len(playlist)
 145
 146             for num, item in enumerate(playlist):
 147                 is_live = item.get('type') == 'LIVE'
 148                 formats = []
 149                 for format_id, stream_url in item.get('streamUrls', {}).items():
 150                     if 'playerType=flash' in stream_url:
 151                         stream_formats = self._extract_m3u8_formats(
 152                             stream_url, playlist_id, 'mp4', 'm3u8_native',
 153                             m3u8_id='hls-%s' % format_id, fatal=False)
 154                     else:
 155                         stream_formats = self._extract_mpd_formats(
 156                             stream_url, playlist_id,
 157                             mpd_id='dash-%s' % format_id, fatal=False)
 158                     # See https://github.com/rg3/youtube-dl/issues/12119#issuecomment-280037031
 159                     if format_id == 'audioDescription':
 160                         for f in stream_formats:
 161                             f['source_preference'] = -10
 162                     formats.extend(stream_formats)
 163
 164                 if user_agent and len(entries) == playlist_len:
 165                     entries[num]['formats'].extend(formats)
 166                     continue
 167
 168                 item_id = item.get('id') or item['assetId']
 169                 title = item['title']
 170
 171                 duration = float_or_none(item.get('duration'))
 172                 thumbnail = item.get('previewImageUrl')
 173
 174                 subtitles = {}
 175                 if item.get('type') == 'VOD':
 176                     subs = item.get('subtitles')
 177                     if subs:
 178                         subtitles = self.extract_subtitles(episode_id, subs)
 179
 180                 if playlist_len == 1:
 181                     final_title = playlist_title or title
 182                     if is_live:
 183                         final_title = self._live_title(final_title)
 184                 else:
 185                     final_title = '%s (%s)' % (playlist_title, title)
 186
 187                 entries.append({
 188                     'id': item_id,
 189                     'title': final_title,
 190                     'description': playlist_description if playlist_len == 1 else None,
 191                     'thumbnail': thumbnail,
 192                     'duration': duration,
 193                     'formats': formats,
 194                     'subtitles': subtitles,
 195                     'is_live': is_live,
 196                 })
 197
 198         for e in entries:
 199             self._sort_formats(e['formats'])
 200
 201         return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
 202
 203     def _get_subtitles(self, episode_id, subs):
 204         original_subtitles = self._download_webpage(
 205             subs[0]['url'], episode_id, 'Downloading subtitles')
 206         srt_subs = self._fix_subtitles(original_subtitles)
 207         return {
 208             'cs': [{
 209                 'ext': 'srt',
 210                 'data': srt_subs,
 211             }]
 212         }
 213
 214     @staticmethod
 215     def _fix_subtitles(subtitles):
 216         """ Convert millisecond-based subtitles to SRT """
 217
 218         def _msectotimecode(msec):
 219             """ Helper utility to convert milliseconds to timecode """
 220             components = []
 221             for divider in [1000, 60, 60, 100]:
 222                 components.append(msec % divider)
 223                 msec //= divider
 224             return '{3:02}:{2:02}:{1:02},{0:03}'.format(*components)
 225
 226         def _fix_subtitle(subtitle):
 227             for line in subtitle.splitlines():
 228                 m = re.match(r'^\s*([0-9]+);\s*([0-9]+)\s+([0-9]+)\s*$', line)
 229                 if m:
 230                     yield m.group(1)
 231                     start, stop = (_msectotimecode(int(t)) for t in m.groups()[1:])
 232                     yield '{0} --> {1}'.format(start, stop)
 233                 else:
 234                     yield line
 235
 236         return '\r\n'.join(_fix_subtitle(subtitles))
 237
 238
 239 class CeskaTelevizePoradyIE(InfoExtractor):
 240     _VALID_URL = r'https?://(?:www\.)?ceskatelevize\.cz/porady/(?:[^/?#&]+/)*(?P<id>[^/#?]+)'
 241     _TESTS = [{
 242         # video with 18+ caution trailer
 243         'url': 'http://www.ceskatelevize.cz/porady/10520528904-queer/215562210900007-bogotart/',
 244         'info_dict': {
 245             'id': '215562210900007-bogotart',
 246             'title': 'Queer: Bogotart',
 247             'description': 'Alternativní průvodce současným queer světem',
 248         },
 249         'playlist': [{
 250             'info_dict': {
 251                 'id': '61924494876844842',
 252                 'ext': 'mp4',
 253                 'title': 'Queer: Bogotart (Varování 18+)',
 254                 'duration': 10.2,
 255             },
 256         }, {
 257             'info_dict': {
 258                 'id': '61924494877068022',
 259                 'ext': 'mp4',
 260                 'title': 'Queer: Bogotart (Queer)',
 261                 'thumbnail': r're:^https?://.*\.jpg',
 262                 'duration': 1558.3,
 263             },
 264         }],
 265         'params': {
 266             # m3u8 download
 267             'skip_download': True,
 268         },
 269     }, {
 270         # iframe embed
 271         'url': 'http://www.ceskatelevize.cz/porady/10614999031-neviditelni/21251212048/',
 272         'only_matching': True,
 273     }]
 274
 275     def _real_extract(self, url):
 276         video_id = self._match_id(url)
 277
 278         webpage = self._download_webpage(url, video_id)
 279
 280         data_url = update_url_query(unescapeHTML(self._search_regex(
 281             (r'<span[^>]*\bdata-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
 282              r'<iframe[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//(?:www\.)?ceskatelevize\.cz/ivysilani/embed/iFramePlayer\.php.*?)\1'),
 283             webpage, 'iframe player url', group='url')), query={
 284                 'autoStart': 'true',
 285         })
 286
 287         return self.url_result(data_url, ie=CeskaTelevizeIE.ie_key())