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 .common import InfoExtractor
   6 from ..utils import ExtractorError
   7 from ..compat import compat_HTTPError
   8
   9
  10 class BBCCoUkIE(InfoExtractor):
  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             'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
 123             'only_matching': True,
 124         }
 125     ]
 126
 127     def _extract_asx_playlist(self, connection, programme_id):
 128         asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
 129         return [ref.get('href') for ref in asx.findall('./Entry/ref')]
 130
 131     def _extract_connection(self, connection, programme_id):
 132         formats = []
 133         protocol = connection.get('protocol')
 134         supplier = connection.get('supplier')
 135         if protocol == 'http':
 136             href = connection.get('href')
 137             # ASX playlist
 138             if supplier == 'asx':
 139                 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
 140                     formats.append({
 141                         'url': ref,
 142                         'format_id': 'ref%s_%s' % (i, supplier),
 143                     })
 144             # Direct link
 145             else:
 146                 formats.append({
 147                     'url': href,
 148                     'format_id': supplier,
 149                 })
 150         elif protocol == 'rtmp':
 151             application = connection.get('application', 'ondemand')
 152             auth_string = connection.get('authString')
 153             identifier = connection.get('identifier')
 154             server = connection.get('server')
 155             formats.append({
 156                 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
 157                 'play_path': identifier,
 158                 'app': '%s?%s' % (application, auth_string),
 159                 'page_url': 'http://www.bbc.co.uk',
 160                 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
 161                 'rtmp_live': False,
 162                 'ext': 'flv',
 163                 'format_id': supplier,
 164             })
 165         return formats
 166
 167     def _extract_items(self, playlist):
 168         return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
 169
 170     def _extract_medias(self, media_selection):
 171         error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error')
 172         if error is not None:
 173             raise ExtractorError(
 174                 '%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True)
 175         return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
 176
 177     def _extract_connections(self, media):
 178         return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
 179
 180     def _extract_video(self, media, programme_id):
 181         formats = []
 182         vbr = int(media.get('bitrate'))
 183         vcodec = media.get('encoding')
 184         service = media.get('service')
 185         width = int(media.get('width'))
 186         height = int(media.get('height'))
 187         file_size = int(media.get('media_file_size'))
 188         for connection in self._extract_connections(media):
 189             conn_formats = self._extract_connection(connection, programme_id)
 190             for format in conn_formats:
 191                 format.update({
 192                     'format_id': '%s_%s' % (service, format['format_id']),
 193                     'width': width,
 194                     'height': height,
 195                     'vbr': vbr,
 196                     'vcodec': vcodec,
 197                     'filesize': file_size,
 198                 })
 199             formats.extend(conn_formats)
 200         return formats
 201
 202     def _extract_audio(self, media, programme_id):
 203         formats = []
 204         abr = int(media.get('bitrate'))
 205         acodec = media.get('encoding')
 206         service = media.get('service')
 207         for connection in self._extract_connections(media):
 208             conn_formats = self._extract_connection(connection, programme_id)
 209             for format in conn_formats:
 210                 format.update({
 211                     'format_id': '%s_%s' % (service, format['format_id']),
 212                     'abr': abr,
 213                     'acodec': acodec,
 214                 })
 215             formats.extend(conn_formats)
 216         return formats
 217
 218     def _get_subtitles(self, media, programme_id):
 219         subtitles = {}
 220         for connection in self._extract_connections(media):
 221             captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
 222             lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
 223             ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}'))
 224             srt = ''
 225
 226             def _extract_text(p):
 227                 if p.text is not None:
 228                     stripped_text = p.text.strip()
 229                     if stripped_text:
 230                         return stripped_text
 231                 return ' '.join(span.text.strip() for span in p.findall('{http://www.w3.org/2006/10/ttaf1}span'))
 232             for pos, p in enumerate(ps):
 233                 srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'), _extract_text(p))
 234             subtitles[lang] = [
 235                 {
 236                     'url': connection.get('href'),
 237                     'ext': 'ttml',
 238                 },
 239                 {
 240                     'data': srt,
 241                     'ext': 'srt',
 242                 },
 243             ]
 244         return subtitles
 245
 246     def _download_media_selector(self, programme_id):
 247         try:
 248             media_selection = self._download_xml(
 249                 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s' % programme_id,
 250                 programme_id, 'Downloading media selection XML')
 251         except ExtractorError as ee:
 252             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 253                 media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().encode('utf-8'))
 254             else:
 255                 raise
 256
 257         formats = []
 258         subtitles = None
 259
 260         for media in self._extract_medias(media_selection):
 261             kind = media.get('kind')
 262             if kind == 'audio':
 263                 formats.extend(self._extract_audio(media, programme_id))
 264             elif kind == 'video':
 265                 formats.extend(self._extract_video(media, programme_id))
 266             elif kind == 'captions':
 267                 subtitles = self.extract_subtitles(media, programme_id)
 268
 269         return formats, subtitles
 270
 271     def _download_playlist(self, playlist_id):
 272         try:
 273             playlist = self._download_json(
 274                 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
 275                 playlist_id, 'Downloading playlist JSON')
 276
 277             version = playlist.get('defaultAvailableVersion')
 278             if version:
 279                 smp_config = version['smpConfig']
 280                 title = smp_config['title']
 281                 description = smp_config['summary']
 282                 for item in smp_config['items']:
 283                     kind = item['kind']
 284                     if kind != 'programme' and kind != 'radioProgramme':
 285                         continue
 286                     programme_id = item.get('vpid')
 287                     duration = int(item.get('duration'))
 288                     formats, subtitles = self._download_media_selector(programme_id)
 289                 return programme_id, title, description, duration, formats, subtitles
 290         except ExtractorError as ee:
 291             if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
 292                 raise
 293
 294         # fallback to legacy playlist
 295         playlist = self._download_xml(
 296             'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id,
 297             playlist_id, 'Downloading legacy playlist XML')
 298
 299         no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
 300         if no_items is not None:
 301             reason = no_items.get('reason')
 302             if reason == 'preAvailability':
 303                 msg = 'Episode %s is not yet available' % playlist_id
 304             elif reason == 'postAvailability':
 305                 msg = 'Episode %s is no longer available' % playlist_id
 306             elif reason == 'noMedia':
 307                 msg = 'Episode %s is not currently available' % playlist_id
 308             else:
 309                 msg = 'Episode %s is not available: %s' % (playlist_id, reason)
 310             raise ExtractorError(msg, expected=True)
 311
 312         for item in self._extract_items(playlist):
 313             kind = item.get('kind')
 314             if kind != 'programme' and kind != 'radioProgramme':
 315                 continue
 316             title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
 317             description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
 318             programme_id = item.get('identifier')
 319             duration = int(item.get('duration'))
 320             formats, subtitles = self._download_media_selector(programme_id)
 321
 322         return programme_id, title, description, duration, formats, subtitles
 323
 324     def _real_extract(self, url):
 325         group_id = self._match_id(url)
 326
 327         webpage = self._download_webpage(url, group_id, 'Downloading video page')
 328
 329         programme_id = self._search_regex(
 330             r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None)
 331         if programme_id:
 332             player = self._download_json(
 333                 'http://www.bbc.co.uk/iplayer/episode/%s.json' % group_id,
 334                 group_id)['jsConf']['player']
 335             title = player['title']
 336             description = player['subtitle']
 337             duration = player['duration']
 338             formats, subtitles = self._download_media_selector(programme_id)
 339         else:
 340             programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
 341
 342         self._sort_formats(formats)
 343
 344         return {
 345             'id': programme_id,
 346             'title': title,
 347             'description': description,
 348             'duration': duration,
 349             'formats': formats,
 350             'subtitles': subtitles,
 351         }