Raphaël G. Git Repositories - youtubedl/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))/)|music/clips[/#])(?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': 'The Canadian poet and songwriter reflects on his musical career.',
  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/music/clips/p02frcc3',
  89             'note': 'Audio',
  90             'info_dict': {
  91                 'id': 'p02frcch',
  92                 'ext': 'flv',
  93                 'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix',
  94                 'description': 'French house superstar Madeon takes us out of the club and onto the after party.',
  95                 'duration': 3507,
  96             },
  97             'params': {
  98                 # rtmp download
  99                 'skip_download': True,
 100             }
 101         }, {
 102             'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
 103             'note': 'Video',
 104             'info_dict': {
 105                 'id': 'p025c103',
 106                 'ext': 'flv',
 107                 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
 108                 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
 109                 'duration': 226,
 110             },
 111             'params': {
 112                 # rtmp download
 113                 'skip_download': True,
 114             }
 115         }, {
 116             'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
 117             'only_matching': True,
 118         }, {
 119             'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
 120             'only_matching': True,
 121         }
 122     ]
 123
 124     def _extract_asx_playlist(self, connection, programme_id):
 125         asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
 126         return [ref.get('href') for ref in asx.findall('./Entry/ref')]
 127
 128     def _extract_connection(self, connection, programme_id):
 129         formats = []
 130         protocol = connection.get('protocol')
 131         supplier = connection.get('supplier')
 132         if protocol == 'http':
 133             href = connection.get('href')
 134             # ASX playlist
 135             if supplier == 'asx':
 136                 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
 137                     formats.append({
 138                         'url': ref,
 139                         'format_id': 'ref%s_%s' % (i, supplier),
 140                     })
 141             # Direct link
 142             else:
 143                 formats.append({
 144                     'url': href,
 145                     'format_id': supplier,
 146                 })
 147         elif protocol == 'rtmp':
 148             application = connection.get('application', 'ondemand')
 149             auth_string = connection.get('authString')
 150             identifier = connection.get('identifier')
 151             server = connection.get('server')
 152             formats.append({
 153                 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
 154                 'play_path': identifier,
 155                 'app': '%s?%s' % (application, auth_string),
 156                 'page_url': 'http://www.bbc.co.uk',
 157                 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
 158                 'rtmp_live': False,
 159                 'ext': 'flv',
 160                 'format_id': supplier,
 161             })
 162         return formats
 163
 164     def _extract_items(self, playlist):
 165         return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
 166
 167     def _extract_medias(self, media_selection):
 168         error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error')
 169         if error is not None:
 170             raise ExtractorError(
 171                 '%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True)
 172         return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
 173
 174     def _extract_connections(self, media):
 175         return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
 176
 177     def _extract_video(self, media, programme_id):
 178         formats = []
 179         vbr = int(media.get('bitrate'))
 180         vcodec = media.get('encoding')
 181         service = media.get('service')
 182         width = int(media.get('width'))
 183         height = int(media.get('height'))
 184         file_size = int(media.get('media_file_size'))
 185         for connection in self._extract_connections(media):
 186             conn_formats = self._extract_connection(connection, programme_id)
 187             for format in conn_formats:
 188                 format.update({
 189                     'format_id': '%s_%s' % (service, format['format_id']),
 190                     'width': width,
 191                     'height': height,
 192                     'vbr': vbr,
 193                     'vcodec': vcodec,
 194                     'filesize': file_size,
 195                 })
 196             formats.extend(conn_formats)
 197         return formats
 198
 199     def _extract_audio(self, media, programme_id):
 200         formats = []
 201         abr = int(media.get('bitrate'))
 202         acodec = media.get('encoding')
 203         service = media.get('service')
 204         for connection in self._extract_connections(media):
 205             conn_formats = self._extract_connection(connection, programme_id)
 206             for format in conn_formats:
 207                 format.update({
 208                     'format_id': '%s_%s' % (service, format['format_id']),
 209                     'abr': abr,
 210                     'acodec': acodec,
 211                 })
 212             formats.extend(conn_formats)
 213         return formats
 214
 215     def _extract_captions(self, media, programme_id):
 216         subtitles = {}
 217         for connection in self._extract_connections(media):
 218             captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
 219             lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
 220             ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}'))
 221             srt = ''
 222             for pos, p in enumerate(ps):
 223                 srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'),
 224                                                           p.text.strip() if p.text is not None else '')
 225             subtitles[lang] = srt
 226         return subtitles
 227
 228     def _download_media_selector(self, programme_id):
 229         try:
 230             media_selection = self._download_xml(
 231                 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s' % programme_id,
 232                 programme_id, 'Downloading media selection XML')
 233         except ExtractorError as ee:
 234             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 235                 media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().encode('utf-8'))
 236             else:
 237                 raise
 238
 239         formats = []
 240         subtitles = None
 241
 242         for media in self._extract_medias(media_selection):
 243             kind = media.get('kind')
 244             if kind == 'audio':
 245                 formats.extend(self._extract_audio(media, programme_id))
 246             elif kind == 'video':
 247                 formats.extend(self._extract_video(media, programme_id))
 248             elif kind == 'captions':
 249                 subtitles = self._extract_captions(media, programme_id)
 250
 251         return formats, subtitles
 252
 253     def _download_playlist(self, playlist_id):
 254         try:
 255             playlist = self._download_json(
 256                 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
 257                 playlist_id, 'Downloading playlist JSON')
 258
 259             version = playlist.get('defaultAvailableVersion')
 260             if version:
 261                 smp_config = version['smpConfig']
 262                 title = smp_config['title']
 263                 description = smp_config['summary']
 264                 for item in smp_config['items']:
 265                     kind = item['kind']
 266                     if kind != 'programme' and kind != 'radioProgramme':
 267                         continue
 268                     programme_id = item.get('vpid')
 269                     duration = int(item.get('duration'))
 270                     formats, subtitles = self._download_media_selector(programme_id)
 271                 return programme_id, title, description, duration, formats, subtitles
 272         except ExtractorError as ee:
 273             if not isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404:
 274                 raise
 275
 276         # fallback to legacy playlist
 277         playlist = self._download_xml(
 278             'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id,
 279             playlist_id, 'Downloading legacy playlist XML')
 280
 281         no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
 282         if no_items is not None:
 283             reason = no_items.get('reason')
 284             if reason == 'preAvailability':
 285                 msg = 'Episode %s is not yet available' % playlist_id
 286             elif reason == 'postAvailability':
 287                 msg = 'Episode %s is no longer available' % playlist_id
 288             elif reason == 'noMedia':
 289                 msg = 'Episode %s is not currently available' % playlist_id
 290             else:
 291                 msg = 'Episode %s is not available: %s' % (playlist_id, reason)
 292             raise ExtractorError(msg, expected=True)
 293
 294         for item in self._extract_items(playlist):
 295             kind = item.get('kind')
 296             if kind != 'programme' and kind != 'radioProgramme':
 297                 continue
 298             title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
 299             description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
 300             programme_id = item.get('identifier')
 301             duration = int(item.get('duration'))
 302             formats, subtitles = self._download_media_selector(programme_id)
 303
 304         return programme_id, title, description, duration, formats, subtitles
 305
 306     def _real_extract(self, url):
 307         group_id = self._match_id(url)
 308
 309         webpage = self._download_webpage(url, group_id, 'Downloading video page')
 310
 311         programme_id = self._search_regex(
 312             r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None)
 313         if programme_id:
 314             player = self._download_json(
 315                 'http://www.bbc.co.uk/iplayer/episode/%s.json' % group_id,
 316                 group_id)['jsConf']['player']
 317             title = player['title']
 318             description = player['subtitle']
 319             duration = player['duration']
 320             formats, subtitles = self._download_media_selector(programme_id)
 321         else:
 322             programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
 323
 324         if self._downloader.params.get('listsubtitles', False):
 325             self._list_available_subtitles(programme_id, subtitles)
 326             return
 327
 328         self._sort_formats(formats)
 329
 330         return {
 331             'id': programme_id,
 332             'title': title,
 333             'description': description,
 334             'duration': duration,
 335             'formats': formats,
 336             'subtitles': subtitles,
 337         }