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