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

   1 from __future__ import unicode_literals
   2
   3 import xml.etree.ElementTree
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8     int_or_none,
   9 )
  10 from ..compat import compat_HTTPError
  11
  12
  13 class BBCCoUkIE(InfoExtractor):
  14     IE_NAME = 'bbc.co.uk'
  15     IE_DESC = 'BBC iPlayer'
  16     _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})'
  17
  18     mediaselector_url = 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s'
  19
  20     _TESTS = [
  21         {
  22             'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
  23             'info_dict': {
  24                 'id': 'b039d07m',
  25                 'ext': 'flv',
  26                 'title': 'Kaleidoscope, Leonard Cohen',
  27                 'description': 'The Canadian poet and songwriter reflects on his musical career.',
  28                 'duration': 1740,
  29             },
  30             'params': {
  31                 # rtmp download
  32                 'skip_download': True,
  33             }
  34         },
  35         {
  36             'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
  37             'info_dict': {
  38                 'id': 'b00yng1d',
  39                 'ext': 'flv',
  40                 'title': 'The Man in Black: Series 3: The Printed Name',
  41                 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
  42                 'duration': 1800,
  43             },
  44             'params': {
  45                 # rtmp download
  46                 'skip_download': True,
  47             },
  48             'skip': 'Episode is no longer available on BBC iPlayer Radio',
  49         },
  50         {
  51             'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
  52             'info_dict': {
  53                 'id': 'b00yng1d',
  54                 'ext': 'flv',
  55                 'title': 'The Voice UK: Series 3: Blind Auditions 5',
  56                 '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.",
  57                 'duration': 5100,
  58             },
  59             'params': {
  60                 # rtmp download
  61                 'skip_download': True,
  62             },
  63             'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
  64         },
  65         {
  66             'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
  67             'info_dict': {
  68                 'id': 'b03k3pb7',
  69                 'ext': 'flv',
  70                 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
  71                 'description': '2. Invasion',
  72                 'duration': 3600,
  73             },
  74             'params': {
  75                 # rtmp download
  76                 'skip_download': True,
  77             },
  78             'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
  79         }, {
  80             'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
  81             'info_dict': {
  82                 'id': 'b04v209v',
  83                 'ext': 'flv',
  84                 'title': 'Pete Tong, The Essential New Tune Special',
  85                 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
  86                 'duration': 10800,
  87             },
  88             'params': {
  89                 # rtmp download
  90                 'skip_download': True,
  91             }
  92         }, {
  93             'url': 'http://www.bbc.co.uk/music/clips/p02frcc3',
  94             'note': 'Audio',
  95             'info_dict': {
  96                 'id': 'p02frcch',
  97                 'ext': 'flv',
  98                 'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix',
  99                 'description': 'French house superstar Madeon takes us out of the club and onto the after party.',
 100                 'duration': 3507,
 101             },
 102             'params': {
 103                 # rtmp download
 104                 'skip_download': True,
 105             }
 106         }, {
 107             'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
 108             'note': 'Video',
 109             'info_dict': {
 110                 'id': 'p025c103',
 111                 'ext': 'flv',
 112                 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
 113                 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
 114                 'duration': 226,
 115             },
 116             'params': {
 117                 # rtmp download
 118                 'skip_download': True,
 119             }
 120         }, {
 121             'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
 122             'info_dict': {
 123                 'id': 'p02n76xf',
 124                 'ext': 'flv',
 125                 'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
 126                 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
 127                 'duration': 3540,
 128             },
 129             'params': {
 130                 # rtmp download
 131                 'skip_download': True,
 132             },
 133             'skip': 'geolocation',
 134         }, {
 135             'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
 136             'info_dict': {
 137                 'id': 'b05zmgw1',
 138                 'ext': 'flv',
 139                 'description': 'Kirsty Wark and Morgan Quaintance visit the Royal Academy as it prepares for its annual artistic extravaganza, meeting people who have come together to make the show unique.',
 140                 'title': 'Royal Academy Summer Exhibition',
 141                 'duration': 3540,
 142             },
 143             'params': {
 144                 # rtmp download
 145                 'skip_download': True,
 146             },
 147             'skip': 'geolocation',
 148         }, {
 149             'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
 150             'only_matching': True,
 151         }, {
 152             'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
 153             'only_matching': True,
 154         }, {
 155             'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
 156             'only_matching': True,
 157         }
 158     ]
 159
 160     def _extract_asx_playlist(self, connection, programme_id):
 161         asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
 162         return [ref.get('href') for ref in asx.findall('./Entry/ref')]
 163
 164     def _extract_connection(self, connection, programme_id):
 165         formats = []
 166         protocol = connection.get('protocol')
 167         supplier = connection.get('supplier')
 168         if protocol == 'http':
 169             href = connection.get('href')
 170             # ASX playlist
 171             if supplier == 'asx':
 172                 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
 173                     formats.append({
 174                         'url': ref,
 175                         'format_id': 'ref%s_%s' % (i, supplier),
 176                     })
 177             # Direct link
 178             else:
 179                 formats.append({
 180                     'url': href,
 181                     'format_id': supplier,
 182                 })
 183         elif protocol == 'rtmp':
 184             application = connection.get('application', 'ondemand')
 185             auth_string = connection.get('authString')
 186             identifier = connection.get('identifier')
 187             server = connection.get('server')
 188             formats.append({
 189                 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
 190                 'play_path': identifier,
 191                 'app': '%s?%s' % (application, auth_string),
 192                 'page_url': 'http://www.bbc.co.uk',
 193                 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
 194                 'rtmp_live': False,
 195                 'ext': 'flv',
 196                 'format_id': supplier,
 197             })
 198         return formats
 199
 200     def _extract_items(self, playlist):
 201         return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
 202
 203     def _extract_medias(self, media_selection):
 204         error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error')
 205         if error is not None:
 206             raise ExtractorError(
 207                 '%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True)
 208         return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
 209
 210     def _extract_connections(self, media):
 211         return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
 212
 213     def _extract_video(self, media, programme_id):
 214         formats = []
 215         vbr = int(media.get('bitrate'))
 216         vcodec = media.get('encoding')
 217         service = media.get('service')
 218         width = int(media.get('width'))
 219         height = int(media.get('height'))
 220         file_size = int(media.get('media_file_size'))
 221         for connection in self._extract_connections(media):
 222             conn_formats = self._extract_connection(connection, programme_id)
 223             for format in conn_formats:
 224                 format.update({
 225                     'format_id': '%s_%s' % (service, format['format_id']),
 226                     'width': width,
 227                     'height': height,
 228                     'vbr': vbr,
 229                     'vcodec': vcodec,
 230                     'filesize': file_size,
 231                 })
 232             formats.extend(conn_formats)
 233         return formats
 234
 235     def _extract_audio(self, media, programme_id):
 236         formats = []
 237         abr = int(media.get('bitrate'))
 238         acodec = media.get('encoding')
 239         service = media.get('service')
 240         for connection in self._extract_connections(media):
 241             conn_formats = self._extract_connection(connection, programme_id)
 242             for format in conn_formats:
 243                 format.update({
 244                     'format_id': '%s_%s' % (service, format['format_id']),
 245                     'abr': abr,
 246                     'acodec': acodec,
 247                 })
 248             formats.extend(conn_formats)
 249         return formats
 250
 251     def _get_subtitles(self, media, programme_id):
 252         subtitles = {}
 253         for connection in self._extract_connections(media):
 254             captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
 255             lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
 256             ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}'))
 257             srt = ''
 258
 259             def _extract_text(p):
 260                 if p.text is not None:
 261                     stripped_text = p.text.strip()
 262                     if stripped_text:
 263                         return stripped_text
 264                 return ' '.join(span.text.strip() for span in p.findall('{http://www.w3.org/2006/10/ttaf1}span'))
 265             for pos, p in enumerate(ps):
 266                 srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'), _extract_text(p))
 267             subtitles[lang] = [
 268                 {
 269                     'url': connection.get('href'),
 270                     'ext': 'ttml',
 271                 },
 272                 {
 273                     'data': srt,
 274                     'ext': 'srt',
 275                 },
 276             ]
 277         return subtitles
 278
 279     def _download_media_selector(self, programme_id):
 280         try:
 281             media_selection = self._download_xml(
 282                 self.mediaselector_url % programme_id,
 283                 programme_id, 'Downloading media selection XML')
 284         except ExtractorError as ee:
 285             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 286                 media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().decode('utf-8'))
 287             else:
 288                 raise
 289
 290         formats = []
 291         subtitles = None
 292
 293         for media in self._extract_medias(media_selection):
 294             kind = media.get('kind')
 295             if kind == 'audio':
 296                 formats.extend(self._extract_audio(media, programme_id))
 297             elif kind == 'video':
 298                 formats.extend(self._extract_video(media, programme_id))
 299             elif kind == 'captions':
 300                 subtitles = self.extract_subtitles(media, programme_id)
 301
 302         return formats, subtitles
 303
 304     def _download_playlist(self, playlist_id):
 305         try:
 306             playlist = self._download_json(
 307                 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
 308                 playlist_id, 'Downloading playlist JSON')
 309
 310             version = playlist.get('defaultAvailableVersion')
 311             if version:
 312                 smp_config = version['smpConfig']
 313                 title = smp_config['title']
 314                 description = smp_config['summary']
 315                 for item in smp_config['items']:
 316                     kind = item['kind']
 317                     if kind != 'programme' and kind != 'radioProgramme':
 318                         continue
 319                     programme_id = item.get('vpid')
 320                     duration = int(item.get('duration'))
 321                     formats, subtitles = self._download_media_selector(programme_id)
 322                 return programme_id, title, description, duration, formats, subtitles
 323         except ExtractorError as ee:
 324             if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
 325                 raise
 326
 327         # fallback to legacy playlist
 328         playlist = self._download_xml(
 329             'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id,
 330             playlist_id, 'Downloading legacy playlist XML')
 331
 332         no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
 333         if no_items is not None:
 334             reason = no_items.get('reason')
 335             if reason == 'preAvailability':
 336                 msg = 'Episode %s is not yet available' % playlist_id
 337             elif reason == 'postAvailability':
 338                 msg = 'Episode %s is no longer available' % playlist_id
 339             elif reason == 'noMedia':
 340                 msg = 'Episode %s is not currently available' % playlist_id
 341             else:
 342                 msg = 'Episode %s is not available: %s' % (playlist_id, reason)
 343             raise ExtractorError(msg, expected=True)
 344
 345         for item in self._extract_items(playlist):
 346             kind = item.get('kind')
 347             if kind != 'programme' and kind != 'radioProgramme':
 348                 continue
 349             title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
 350             description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
 351             programme_id = item.get('identifier')
 352             duration = int(item.get('duration'))
 353             formats, subtitles = self._download_media_selector(programme_id)
 354
 355         return programme_id, title, description, duration, formats, subtitles
 356
 357     def _real_extract(self, url):
 358         group_id = self._match_id(url)
 359
 360         webpage = self._download_webpage(url, group_id, 'Downloading video page')
 361
 362         programme_id = None
 363
 364         tviplayer = self._search_regex(
 365             r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
 366             webpage, 'player', default=None)
 367
 368         if tviplayer:
 369             player = self._parse_json(tviplayer, group_id).get('player', {})
 370             duration = int_or_none(player.get('duration'))
 371             programme_id = player.get('vpid')
 372
 373         if not programme_id:
 374             programme_id = self._search_regex(
 375                 r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None)
 376
 377         if programme_id:
 378             formats, subtitles = self._download_media_selector(programme_id)
 379             title = self._og_search_title(webpage)
 380             description = self._search_regex(
 381                 r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
 382                 webpage, 'description', fatal=False)
 383         else:
 384             programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
 385
 386         self._sort_formats(formats)
 387
 388         return {
 389             'id': programme_id,
 390             'title': title,
 391             'description': description,
 392             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 393             'duration': duration,
 394             'formats': formats,
 395             'subtitles': subtitles,
 396         }