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

   1 from __future__ import unicode_literals
   2
   3 import xml.etree.ElementTree
   4
   5 from .subtitles import SubtitlesInfoExtractor
   6 from ..utils import ExtractorError
   7 from ..compat import compat_HTTPError
   8
   9
  10 class BBCCoUkIE(SubtitlesInfoExtractor):
  11     IE_NAME = 'bbc.co.uk'
  12     IE_DESC = 'BBC iPlayer'
  13     _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:programmes|iplayer/(?:episode|playlist))/(?P<id>[\da-z]{8})'
  14
  15     _TESTS = [
  16         {
  17             'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
  18             'info_dict': {
  19                 'id': 'b039d07m',
  20                 'ext': 'flv',
  21                 'title': 'Kaleidoscope: Leonard Cohen',
  22                 'description': 'md5:db4755d7a665ae72343779f7dacb402c',
  23                 'duration': 1740,
  24             },
  25             'params': {
  26                 # rtmp download
  27                 'skip_download': True,
  28             }
  29         },
  30         {
  31             'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
  32             'info_dict': {
  33                 'id': 'b00yng1d',
  34                 'ext': 'flv',
  35                 'title': 'The Man in Black: Series 3: The Printed Name',
  36                 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
  37                 'duration': 1800,
  38             },
  39             'params': {
  40                 # rtmp download
  41                 'skip_download': True,
  42             },
  43             'skip': 'Episode is no longer available on BBC iPlayer Radio',
  44         },
  45         {
  46             'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
  47             'info_dict': {
  48                 'id': 'b00yng1d',
  49                 'ext': 'flv',
  50                 'title': 'The Voice UK: Series 3: Blind Auditions 5',
  51                 'description': "Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.",
  52                 'duration': 5100,
  53             },
  54             'params': {
  55                 # rtmp download
  56                 'skip_download': True,
  57             },
  58             'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
  59         },
  60         {
  61             'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
  62             'info_dict': {
  63                 'id': 'b03k3pb7',
  64                 'ext': 'flv',
  65                 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
  66                 'description': '2. Invasion',
  67                 'duration': 3600,
  68             },
  69             'params': {
  70                 # rtmp download
  71                 'skip_download': True,
  72             },
  73             'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
  74         }, {
  75             'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
  76             'info_dict': {
  77                 'id': 'b04v209v',
  78                 'ext': 'flv',
  79                 'title': 'Pete Tong, The Essential New Tune Special',
  80                 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
  81                 'duration': 10800,
  82             },
  83             'params': {
  84                 # rtmp download
  85                 'skip_download': True,
  86             }
  87         }, {
  88             'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
  89             'only_matching': True,
  90         }
  91     ]
  92
  93     def _extract_asx_playlist(self, connection, programme_id):
  94         asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
  95         return [ref.get('href') for ref in asx.findall('./Entry/ref')]
  96
  97     def _extract_connection(self, connection, programme_id):
  98         formats = []
  99         protocol = connection.get('protocol')
 100         supplier = connection.get('supplier')
 101         if protocol == 'http':
 102             href = connection.get('href')
 103             # ASX playlist
 104             if supplier == 'asx':
 105                 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
 106                     formats.append({
 107                         'url': ref,
 108                         'format_id': 'ref%s_%s' % (i, supplier),
 109                     })
 110             # Direct link
 111             else:
 112                 formats.append({
 113                     'url': href,
 114                     'format_id': supplier,
 115                 })
 116         elif protocol == 'rtmp':
 117             application = connection.get('application', 'ondemand')
 118             auth_string = connection.get('authString')
 119             identifier = connection.get('identifier')
 120             server = connection.get('server')
 121             formats.append({
 122                 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
 123                 'play_path': identifier,
 124                 'app': '%s?%s' % (application, auth_string),
 125                 'page_url': 'http://www.bbc.co.uk',
 126                 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
 127                 'rtmp_live': False,
 128                 'ext': 'flv',
 129                 'format_id': supplier,
 130             })
 131         return formats
 132
 133     def _extract_items(self, playlist):
 134         return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
 135
 136     def _extract_medias(self, media_selection):
 137         error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error')
 138         if error is not None:
 139             raise ExtractorError(
 140                 '%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True)
 141         return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
 142
 143     def _extract_connections(self, media):
 144         return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
 145
 146     def _extract_video(self, media, programme_id):
 147         formats = []
 148         vbr = int(media.get('bitrate'))
 149         vcodec = media.get('encoding')
 150         service = media.get('service')
 151         width = int(media.get('width'))
 152         height = int(media.get('height'))
 153         file_size = int(media.get('media_file_size'))
 154         for connection in self._extract_connections(media):
 155             conn_formats = self._extract_connection(connection, programme_id)
 156             for format in conn_formats:
 157                 format.update({
 158                     'format_id': '%s_%s' % (service, format['format_id']),
 159                     'width': width,
 160                     'height': height,
 161                     'vbr': vbr,
 162                     'vcodec': vcodec,
 163                     'filesize': file_size,
 164                 })
 165             formats.extend(conn_formats)
 166         return formats
 167
 168     def _extract_audio(self, media, programme_id):
 169         formats = []
 170         abr = int(media.get('bitrate'))
 171         acodec = media.get('encoding')
 172         service = media.get('service')
 173         for connection in self._extract_connections(media):
 174             conn_formats = self._extract_connection(connection, programme_id)
 175             for format in conn_formats:
 176                 format.update({
 177                     'format_id': '%s_%s' % (service, format['format_id']),
 178                     'abr': abr,
 179                     'acodec': acodec,
 180                 })
 181             formats.extend(conn_formats)
 182         return formats
 183
 184     def _extract_captions(self, media, programme_id):
 185         subtitles = {}
 186         for connection in self._extract_connections(media):
 187             captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
 188             lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
 189             ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}'))
 190             srt = ''
 191             for pos, p in enumerate(ps):
 192                 srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'),
 193                                                           p.text.strip() if p.text is not None else '')
 194             subtitles[lang] = srt
 195         return subtitles
 196
 197     def _download_media_selector(self, programme_id):
 198         try:
 199             media_selection = self._download_xml(
 200                 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s' % programme_id,
 201                 programme_id, 'Downloading media selection XML')
 202         except ExtractorError as ee:
 203             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 204                 media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().encode('utf-8'))
 205             else:
 206                 raise
 207
 208         formats = []
 209         subtitles = None
 210
 211         for media in self._extract_medias(media_selection):
 212             kind = media.get('kind')
 213             if kind == 'audio':
 214                 formats.extend(self._extract_audio(media, programme_id))
 215             elif kind == 'video':
 216                 formats.extend(self._extract_video(media, programme_id))
 217             elif kind == 'captions':
 218                 subtitles = self._extract_captions(media, programme_id)
 219
 220         return formats, subtitles
 221
 222     def _download_playlist(self, playlist_id):
 223         try:
 224             playlist = self._download_json(
 225                 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
 226                 playlist_id, 'Downloading playlist JSON')
 227
 228             version = playlist.get('defaultAvailableVersion')
 229             if version:
 230                 smp_config = version['smpConfig']
 231                 title = smp_config['title']
 232                 description = smp_config['summary']
 233                 for item in smp_config['items']:
 234                     kind = item['kind']
 235                     if kind != 'programme' and kind != 'radioProgramme':
 236                         continue
 237                     programme_id = item.get('vpid')
 238                     duration = int(item.get('duration'))
 239                     formats, subtitles = self._download_media_selector(programme_id)
 240                 return programme_id, title, description, duration, formats, subtitles
 241         except ExtractorError as ee:
 242             if not isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404:
 243                 raise
 244
 245         # fallback to legacy playlist
 246         playlist = self._download_xml(
 247             'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id,
 248             playlist_id, 'Downloading legacy playlist XML')
 249
 250         no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
 251         if no_items is not None:
 252             reason = no_items.get('reason')
 253             if reason == 'preAvailability':
 254                 msg = 'Episode %s is not yet available' % playlist_id
 255             elif reason == 'postAvailability':
 256                 msg = 'Episode %s is no longer available' % playlist_id
 257             elif reason == 'noMedia':
 258                 msg = 'Episode %s is not currently available' % playlist_id
 259             else:
 260                 msg = 'Episode %s is not available: %s' % (playlist_id, reason)
 261             raise ExtractorError(msg, expected=True)
 262
 263         for item in self._extract_items(playlist):
 264             kind = item.get('kind')
 265             if kind != 'programme' and kind != 'radioProgramme':
 266                 continue
 267             title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
 268             description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
 269             programme_id = item.get('identifier')
 270             duration = int(item.get('duration'))
 271             formats, subtitles = self._download_media_selector(programme_id)
 272
 273         return programme_id, title, description, duration, formats, subtitles
 274
 275     def _real_extract(self, url):
 276         group_id = self._match_id(url)
 277
 278         webpage = self._download_webpage(url, group_id, 'Downloading video page')
 279
 280         programme_id = self._search_regex(
 281             r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None)
 282         if programme_id:
 283             player = self._download_json(
 284                 'http://www.bbc.co.uk/iplayer/episode/%s.json' % group_id,
 285                 group_id)['jsConf']['player']
 286             title = player['title']
 287             description = player['subtitle']
 288             duration = player['duration']
 289             formats, subtitles = self._download_media_selector(programme_id)
 290         else:
 291             programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
 292
 293         if self._downloader.params.get('listsubtitles', False):
 294             self._list_available_subtitles(programme_id, subtitles)
 295             return
 296
 297         self._sort_formats(formats)
 298
 299         return {
 300             'id': programme_id,
 301             'title': title,
 302             'description': description,
 303             'duration': duration,
 304             'formats': formats,
 305             'subtitles': subtitles,
 306         }