_ 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/playlist/p01dvks4',
 120             'only_matching': True,
 121         }, {
 122             'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
 123             'only_matching': True,
 124         }, {
 125             'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
 126             'only_matching': True,
 127         }
 128     ]
 129
 130     def _extract_asx_playlist(self, connection, programme_id):
 131         asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
 132         return [ref.get('href') for ref in asx.findall('./Entry/ref')]
 133
 134     def _extract_connection(self, connection, programme_id):
 135         formats = []
 136         protocol = connection.get('protocol')
 137         supplier = connection.get('supplier')
 138         if protocol == 'http':
 139             href = connection.get('href')
 140             # ASX playlist
 141             if supplier == 'asx':
 142                 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
 143                     formats.append({
 144                         'url': ref,
 145                         'format_id': 'ref%s_%s' % (i, supplier),
 146                     })
 147             # Direct link
 148             else:
 149                 formats.append({
 150                     'url': href,
 151                     'format_id': supplier,
 152                 })
 153         elif protocol == 'rtmp':
 154             application = connection.get('application', 'ondemand')
 155             auth_string = connection.get('authString')
 156             identifier = connection.get('identifier')
 157             server = connection.get('server')
 158             formats.append({
 159                 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
 160                 'play_path': identifier,
 161                 'app': '%s?%s' % (application, auth_string),
 162                 'page_url': 'http://www.bbc.co.uk',
 163                 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
 164                 'rtmp_live': False,
 165                 'ext': 'flv',
 166                 'format_id': supplier,
 167             })
 168         return formats
 169
 170     def _extract_items(self, playlist):
 171         return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
 172
 173     def _extract_medias(self, media_selection):
 174         error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error')
 175         if error is not None:
 176             raise ExtractorError(
 177                 '%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True)
 178         return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
 179
 180     def _extract_connections(self, media):
 181         return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
 182
 183     def _extract_video(self, media, programme_id):
 184         formats = []
 185         vbr = int(media.get('bitrate'))
 186         vcodec = media.get('encoding')
 187         service = media.get('service')
 188         width = int(media.get('width'))
 189         height = int(media.get('height'))
 190         file_size = int(media.get('media_file_size'))
 191         for connection in self._extract_connections(media):
 192             conn_formats = self._extract_connection(connection, programme_id)
 193             for format in conn_formats:
 194                 format.update({
 195                     'format_id': '%s_%s' % (service, format['format_id']),
 196                     'width': width,
 197                     'height': height,
 198                     'vbr': vbr,
 199                     'vcodec': vcodec,
 200                     'filesize': file_size,
 201                 })
 202             formats.extend(conn_formats)
 203         return formats
 204
 205     def _extract_audio(self, media, programme_id):
 206         formats = []
 207         abr = int(media.get('bitrate'))
 208         acodec = media.get('encoding')
 209         service = media.get('service')
 210         for connection in self._extract_connections(media):
 211             conn_formats = self._extract_connection(connection, programme_id)
 212             for format in conn_formats:
 213                 format.update({
 214                     'format_id': '%s_%s' % (service, format['format_id']),
 215                     'abr': abr,
 216                     'acodec': acodec,
 217                 })
 218             formats.extend(conn_formats)
 219         return formats
 220
 221     def _get_subtitles(self, media, programme_id):
 222         subtitles = {}
 223         for connection in self._extract_connections(media):
 224             captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
 225             lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
 226             ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}'))
 227             srt = ''
 228
 229             def _extract_text(p):
 230                 if p.text is not None:
 231                     stripped_text = p.text.strip()
 232                     if stripped_text:
 233                         return stripped_text
 234                 return ' '.join(span.text.strip() for span in p.findall('{http://www.w3.org/2006/10/ttaf1}span'))
 235             for pos, p in enumerate(ps):
 236                 srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'), _extract_text(p))
 237             subtitles[lang] = [
 238                 {
 239                     'url': connection.get('href'),
 240                     'ext': 'ttml',
 241                 },
 242                 {
 243                     'data': srt,
 244                     'ext': 'srt',
 245                 },
 246             ]
 247         return subtitles
 248
 249     def _download_media_selector(self, programme_id):
 250         try:
 251             media_selection = self._download_xml(
 252                 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s' % programme_id,
 253                 programme_id, 'Downloading media selection XML')
 254         except ExtractorError as ee:
 255             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 256                 media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().encode('utf-8'))
 257             else:
 258                 raise
 259
 260         formats = []
 261         subtitles = None
 262
 263         for media in self._extract_medias(media_selection):
 264             kind = media.get('kind')
 265             if kind == 'audio':
 266                 formats.extend(self._extract_audio(media, programme_id))
 267             elif kind == 'video':
 268                 formats.extend(self._extract_video(media, programme_id))
 269             elif kind == 'captions':
 270                 subtitles = self.extract_subtitles(media, programme_id)
 271
 272         return formats, subtitles
 273
 274     def _download_playlist(self, playlist_id):
 275         try:
 276             playlist = self._download_json(
 277                 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
 278                 playlist_id, 'Downloading playlist JSON')
 279
 280             version = playlist.get('defaultAvailableVersion')
 281             if version:
 282                 smp_config = version['smpConfig']
 283                 title = smp_config['title']
 284                 description = smp_config['summary']
 285                 for item in smp_config['items']:
 286                     kind = item['kind']
 287                     if kind != 'programme' and kind != 'radioProgramme':
 288                         continue
 289                     programme_id = item.get('vpid')
 290                     duration = int(item.get('duration'))
 291                     formats, subtitles = self._download_media_selector(programme_id)
 292                 return programme_id, title, description, duration, formats, subtitles
 293         except ExtractorError as ee:
 294             if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
 295                 raise
 296
 297         # fallback to legacy playlist
 298         playlist = self._download_xml(
 299             'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id,
 300             playlist_id, 'Downloading legacy playlist XML')
 301
 302         no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
 303         if no_items is not None:
 304             reason = no_items.get('reason')
 305             if reason == 'preAvailability':
 306                 msg = 'Episode %s is not yet available' % playlist_id
 307             elif reason == 'postAvailability':
 308                 msg = 'Episode %s is no longer available' % playlist_id
 309             elif reason == 'noMedia':
 310                 msg = 'Episode %s is not currently available' % playlist_id
 311             else:
 312                 msg = 'Episode %s is not available: %s' % (playlist_id, reason)
 313             raise ExtractorError(msg, expected=True)
 314
 315         for item in self._extract_items(playlist):
 316             kind = item.get('kind')
 317             if kind != 'programme' and kind != 'radioProgramme':
 318                 continue
 319             title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
 320             description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
 321             programme_id = item.get('identifier')
 322             duration = int(item.get('duration'))
 323             formats, subtitles = self._download_media_selector(programme_id)
 324
 325         return programme_id, title, description, duration, formats, subtitles
 326
 327     def _real_extract(self, url):
 328         group_id = self._match_id(url)
 329
 330         webpage = self._download_webpage(url, group_id, 'Downloading video page')
 331
 332         thumbnail = self._og_search_thumbnail(webpage)
 333
 334         programme_id = None
 335
 336         tviplayer = self._search_regex(
 337             r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
 338             webpage, 'player', default=None)
 339
 340         if tviplayer:
 341             player = self._parse_json(tviplayer, group_id).get('player', {})
 342             duration = int_or_none(player.get('duration'))
 343             programme_id = player.get('vpid')
 344
 345         if not programme_id:
 346             programme_id = self._search_regex(
 347                 r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None)
 348
 349         if programme_id:
 350             formats, subtitles = self._download_media_selector(programme_id)
 351             title = self._og_search_title(webpage)
 352             description = self._search_regex(
 353                 r'<p class="medium-description">([^<]+)</p>',
 354                 webpage, 'description', fatal=False)
 355         else:
 356             programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
 357
 358         self._sort_formats(formats)
 359
 360         return {
 361             'id': programme_id,
 362             'title': title,
 363             'description': description,
 364             'thumbnail': thumbnail,
 365             'duration': duration,
 366             'formats': formats,
 367             'subtitles': subtitles,
 368         }