youtube_dl/extractor/bbc.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     parse_duration,
   9     int_or_none,
  10 )
  11 from ..compat import compat_HTTPError
  12 import re
  13
  14
  15 class BBCCoUkIE(InfoExtractor):
  16     IE_NAME = 'bbc.co.uk'
  17     IE_DESC = 'BBC iPlayer'
  18     _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})'
  19
  20     mediaselector_url = 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s'
  21
  22     _TESTS = [
  23         {
  24             'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
  25             'info_dict': {
  26                 'id': 'b039d07m',
  27                 'ext': 'flv',
  28                 'title': 'Kaleidoscope, Leonard Cohen',
  29                 'description': 'The Canadian poet and songwriter reflects on his musical career.',
  30                 'duration': 1740,
  31             },
  32             'params': {
  33                 # rtmp download
  34                 'skip_download': True,
  35             }
  36         },
  37         {
  38             'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
  39             'info_dict': {
  40                 'id': 'b00yng1d',
  41                 'ext': 'flv',
  42                 'title': 'The Man in Black: Series 3: The Printed Name',
  43                 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
  44                 'duration': 1800,
  45             },
  46             'params': {
  47                 # rtmp download
  48                 'skip_download': True,
  49             },
  50             'skip': 'Episode is no longer available on BBC iPlayer Radio',
  51         },
  52         {
  53             'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
  54             'info_dict': {
  55                 'id': 'b00yng1d',
  56                 'ext': 'flv',
  57                 'title': 'The Voice UK: Series 3: Blind Auditions 5',
  58                 '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.",
  59                 'duration': 5100,
  60             },
  61             'params': {
  62                 # rtmp download
  63                 'skip_download': True,
  64             },
  65             'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
  66         },
  67         {
  68             'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
  69             'info_dict': {
  70                 'id': 'b03k3pb7',
  71                 'ext': 'flv',
  72                 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
  73                 'description': '2. Invasion',
  74                 'duration': 3600,
  75             },
  76             'params': {
  77                 # rtmp download
  78                 'skip_download': True,
  79             },
  80             'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
  81         }, {
  82             'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
  83             'info_dict': {
  84                 'id': 'b04v209v',
  85                 'ext': 'flv',
  86                 'title': 'Pete Tong, The Essential New Tune Special',
  87                 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
  88                 'duration': 10800,
  89             },
  90             'params': {
  91                 # rtmp download
  92                 'skip_download': True,
  93             }
  94         }, {
  95             'url': 'http://www.bbc.co.uk/music/clips/p02frcc3',
  96             'note': 'Audio',
  97             'info_dict': {
  98                 'id': 'p02frcch',
  99                 'ext': 'flv',
 100                 'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix',
 101                 'description': 'French house superstar Madeon takes us out of the club and onto the after party.',
 102                 'duration': 3507,
 103             },
 104             'params': {
 105                 # rtmp download
 106                 'skip_download': True,
 107             }
 108         }, {
 109             'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
 110             'note': 'Video',
 111             'info_dict': {
 112                 'id': 'p025c103',
 113                 'ext': 'flv',
 114                 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
 115                 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
 116                 'duration': 226,
 117             },
 118             'params': {
 119                 # rtmp download
 120                 'skip_download': True,
 121             }
 122         }, {
 123             'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
 124             'info_dict': {
 125                 'id': 'p02n76xf',
 126                 'ext': 'flv',
 127                 'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
 128                 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
 129                 'duration': 3540,
 130             },
 131             'params': {
 132                 # rtmp download
 133                 'skip_download': True,
 134             },
 135             'skip': 'geolocation',
 136         }, {
 137             'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
 138             'info_dict': {
 139                 'id': 'b05zmgw1',
 140                 'ext': 'flv',
 141                 '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.',
 142                 'title': 'Royal Academy Summer Exhibition',
 143                 'duration': 3540,
 144             },
 145             'params': {
 146                 # rtmp download
 147                 'skip_download': True,
 148             },
 149             'skip': 'geolocation',
 150         }, {
 151             'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
 152             'only_matching': True,
 153         }, {
 154             'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
 155             'only_matching': True,
 156         }, {
 157             'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
 158             'only_matching': True,
 159         }
 160     ]
 161
 162     def _extract_asx_playlist(self, connection, programme_id):
 163         asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
 164         return [ref.get('href') for ref in asx.findall('./Entry/ref')]
 165
 166     def _extract_connection(self, connection, programme_id):
 167         formats = []
 168         protocol = connection.get('protocol')
 169         supplier = connection.get('supplier')
 170         if protocol == 'http':
 171             href = connection.get('href')
 172             # ASX playlist
 173             if supplier == 'asx':
 174                 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
 175                     formats.append({
 176                         'url': ref,
 177                         'format_id': 'ref%s_%s' % (i, supplier),
 178                     })
 179             # Direct link
 180             else:
 181                 formats.append({
 182                     'url': href,
 183                     'format_id': supplier,
 184                 })
 185         elif protocol == 'rtmp':
 186             application = connection.get('application', 'ondemand')
 187             auth_string = connection.get('authString')
 188             identifier = connection.get('identifier')
 189             server = connection.get('server')
 190             formats.append({
 191                 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
 192                 'play_path': identifier,
 193                 'app': '%s?%s' % (application, auth_string),
 194                 'page_url': 'http://www.bbc.co.uk',
 195                 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
 196                 'rtmp_live': False,
 197                 'ext': 'flv',
 198                 'format_id': supplier,
 199             })
 200         return formats
 201
 202     def _extract_items(self, playlist):
 203         return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
 204
 205     def _extract_medias(self, media_selection):
 206         error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error')
 207         if error is not None:
 208             raise ExtractorError(
 209                 '%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True)
 210         return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
 211
 212     def _extract_connections(self, media):
 213         return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
 214
 215     def _extract_video(self, media, programme_id):
 216         formats = []
 217         vbr = int(media.get('bitrate'))
 218         vcodec = media.get('encoding')
 219         service = media.get('service')
 220         width = int(media.get('width'))
 221         height = int(media.get('height'))
 222         file_size = int(media.get('media_file_size'))
 223         for connection in self._extract_connections(media):
 224             conn_formats = self._extract_connection(connection, programme_id)
 225             for format in conn_formats:
 226                 format.update({
 227                     'format_id': '%s_%s' % (service, format['format_id']),
 228                     'width': width,
 229                     'height': height,
 230                     'vbr': vbr,
 231                     'vcodec': vcodec,
 232                     'filesize': file_size,
 233                 })
 234             formats.extend(conn_formats)
 235         return formats
 236
 237     def _extract_audio(self, media, programme_id):
 238         formats = []
 239         abr = int(media.get('bitrate'))
 240         acodec = media.get('encoding')
 241         service = media.get('service')
 242         for connection in self._extract_connections(media):
 243             conn_formats = self._extract_connection(connection, programme_id)
 244             for format in conn_formats:
 245                 format.update({
 246                     'format_id': '%s_%s' % (service, format['format_id']),
 247                     'abr': abr,
 248                     'acodec': acodec,
 249                 })
 250             formats.extend(conn_formats)
 251         return formats
 252
 253     def _get_subtitles(self, media, programme_id):
 254         subtitles = {}
 255         for connection in self._extract_connections(media):
 256             captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
 257             lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
 258             ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}'))
 259             srt = ''
 260
 261             def _extract_text(p):
 262                 if p.text is not None:
 263                     stripped_text = p.text.strip()
 264                     if stripped_text:
 265                         return stripped_text
 266                 return ' '.join(span.text.strip() for span in p.findall('{http://www.w3.org/2006/10/ttaf1}span'))
 267             for pos, p in enumerate(ps):
 268                 srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'), _extract_text(p))
 269             subtitles[lang] = [
 270                 {
 271                     'url': connection.get('href'),
 272                     'ext': 'ttml',
 273                 },
 274                 {
 275                     'data': srt,
 276                     'ext': 'srt',
 277                 },
 278             ]
 279         return subtitles
 280
 281     def _download_media_selector(self, programme_id):
 282         try:
 283             media_selection = self._download_xml(
 284                 self.mediaselector_url % programme_id,
 285                 programme_id, 'Downloading media selection XML')
 286         except ExtractorError as ee:
 287             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 288                 media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().decode('utf-8'))
 289             else:
 290                 raise
 291
 292         formats = []
 293         subtitles = None
 294
 295         for media in self._extract_medias(media_selection):
 296             kind = media.get('kind')
 297             if kind == 'audio':
 298                 formats.extend(self._extract_audio(media, programme_id))
 299             elif kind == 'video':
 300                 formats.extend(self._extract_video(media, programme_id))
 301             elif kind == 'captions':
 302                 subtitles = self.extract_subtitles(media, programme_id)
 303
 304         return formats, subtitles
 305
 306     def _download_playlist(self, playlist_id):
 307         try:
 308             playlist = self._download_json(
 309                 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
 310                 playlist_id, 'Downloading playlist JSON')
 311
 312             version = playlist.get('defaultAvailableVersion')
 313             if version:
 314                 smp_config = version['smpConfig']
 315                 title = smp_config['title']
 316                 description = smp_config['summary']
 317                 for item in smp_config['items']:
 318                     kind = item['kind']
 319                     if kind != 'programme' and kind != 'radioProgramme':
 320                         continue
 321                     programme_id = item.get('vpid')
 322                     duration = int(item.get('duration'))
 323                     formats, subtitles = self._download_media_selector(programme_id)
 324                 return programme_id, title, description, duration, formats, subtitles
 325         except ExtractorError as ee:
 326             if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
 327                 raise
 328
 329         # fallback to legacy playlist
 330         playlist = self._download_xml(
 331             'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id,
 332             playlist_id, 'Downloading legacy playlist XML')
 333
 334         no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
 335         if no_items is not None:
 336             reason = no_items.get('reason')
 337             if reason == 'preAvailability':
 338                 msg = 'Episode %s is not yet available' % playlist_id
 339             elif reason == 'postAvailability':
 340                 msg = 'Episode %s is no longer available' % playlist_id
 341             elif reason == 'noMedia':
 342                 msg = 'Episode %s is not currently available' % playlist_id
 343             else:
 344                 msg = 'Episode %s is not available: %s' % (playlist_id, reason)
 345             raise ExtractorError(msg, expected=True)
 346
 347         for item in self._extract_items(playlist):
 348             kind = item.get('kind')
 349             if kind != 'programme' and kind != 'radioProgramme':
 350                 continue
 351             title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
 352             description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
 353             programme_id = item.get('identifier')
 354             duration = int(item.get('duration'))
 355             formats, subtitles = self._download_media_selector(programme_id)
 356
 357         return programme_id, title, description, duration, formats, subtitles
 358
 359     def _real_extract(self, url):
 360         group_id = self._match_id(url)
 361
 362         webpage = self._download_webpage(url, group_id, 'Downloading video page')
 363
 364         programme_id = None
 365
 366         tviplayer = self._search_regex(
 367             r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
 368             webpage, 'player', default=None)
 369
 370         if tviplayer:
 371             player = self._parse_json(tviplayer, group_id).get('player', {})
 372             duration = int_or_none(player.get('duration'))
 373             programme_id = player.get('vpid')
 374
 375         if not programme_id:
 376             programme_id = self._search_regex(
 377                 r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None)
 378
 379         if programme_id:
 380             formats, subtitles = self._download_media_selector(programme_id)
 381             title = self._og_search_title(webpage)
 382             description = self._search_regex(
 383                 r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
 384                 webpage, 'description', fatal=False)
 385         else:
 386             programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
 387
 388         self._sort_formats(formats)
 389
 390         return {
 391             'id': programme_id,
 392             'title': title,
 393             'description': description,
 394             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 395             'duration': duration,
 396             'formats': formats,
 397             'subtitles': subtitles,
 398         }
 399
 400
 401 class BBCNewsIE(BBCCoUkIE):
 402     IE_NAME = 'bbc.com'
 403     IE_DESC = 'BBC news'
 404     _VALID_URL = r'https?://(?:www\.)?bbc\.com/.+?/(?P<id>[^/]+)$'
 405
 406     mediaselector_url = 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s'
 407
 408     _TESTS = [{
 409         'url': 'http://www.bbc.com/news/world-europe-32668511',
 410         'info_dict': {
 411             'id': 'world-europe-32668511',
 412             'title': 'Russia stages massive WW2 parade despite Western boycott',
 413         },
 414         'playlist_count': 2,
 415     },{
 416         'url': 'http://www.bbc.com/news/business-28299555',
 417         'info_dict': {
 418             'id': 'business-28299555',
 419             'title': 'Farnborough Airshow: Video highlights',
 420         },
 421         'playlist_count': 9,
 422     },{
 423         'url': 'http://www.bbc.com/news/world-europe-32041533',
 424         'note': 'Video',
 425         'info_dict': {
 426             'id': 'p02mprgb',
 427             'ext': 'mp4',
 428             'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
 429             'description': 'Germanwings plane crash site in aerial video - Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
 430             'duration': 47,
 431             'upload_date': '20150324',
 432             'uploader': 'BBC News',
 433         },
 434         'params': {
 435             'skip_download': True,
 436         }
 437     },{
 438         'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
 439         'note': 'Video',
 440         'info_dict': {
 441             'id': 'NA',
 442             'ext': 'mp4',
 443             'title': 'YPG: Tel Abyad\'\u0131n tamam\u0131 kontrol\xfcm\xfczde',
 444             'description': 'YPG: Tel Abyad\'\u0131n tamam\u0131 kontrol\xfcm\xfczde',
 445             'duration': 47,
 446             'upload_date': '20150615',
 447             'uploader': 'BBC News',
 448         },
 449         'params': {
 450             'skip_download': True,
 451         }
 452     },{
 453         'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
 454         'note': 'Video',
 455         'info_dict': {
 456             'id': '39275083',
 457             'ext': 'mp4',
 458             'title': 'Honduras militariza sus hospitales por nuevo esc\xe1ndalo de corrupci\xf3n',
 459             'description': 'Honduras militariza sus hospitales por nuevo esc\xe1ndalo de corrupci\xf3n',
 460             'duration': 87,
 461             'upload_date': '20150619',
 462             'uploader': 'BBC News',
 463         },
 464         'params': {
 465             'skip_download': True,
 466         }
 467     }]
 468
 469     def _real_extract(self, url):
 470         list_id = self._match_id(url)
 471         webpage = self._download_webpage(url, list_id)
 472
 473         list_title = self._html_search_regex(r'<title>(.*?)(?:\s*-\s*BBC [^ ]+)?</title>', webpage, 'list title')
 474
 475         pubdate = self._html_search_regex(r'"datePublished":\s*"(\d+-\d+-\d+)', webpage, 'date', default=None)
 476         if pubdate:
 477            pubdate = pubdate.replace('-','')
 478
 479         ret = []
 480         jsent = []
 481
 482         # works with bbc.com/news/something-something-123456 articles
 483         jsent = map(
 484            lambda m: self._parse_json(m,list_id),
 485            re.findall(r"data-media-meta='({[^']+})'", webpage)
 486         )
 487
 488         if len(jsent) == 0:
 489            # http://www.bbc.com/news/video_and_audio/international
 490            # and single-video articles
 491            masset = self._html_search_regex(r'mediaAssetPage\.init\(\s*({.+?}), "/', webpage, 'mediaassets', default=None)
 492            if masset:
 493               jmasset = self._parse_json(masset,list_id)
 494               for key, val in jmasset.get('videos',{}).items():
 495                   for skey, sval in val.items():
 496                       sval['id'] = skey
 497                       jsent.append(sval)
 498
 499         if len(jsent) == 0:
 500            # stubbornly generic extractor for {json with "image":{allvideoshavethis},etc}
 501            # in http://www.bbc.com/news/video_and_audio/international
 502            # prone to breaking if entries have sourceFiles list
 503            jsent = map(
 504                lambda m: self._parse_json(m,list_id),
 505                re.findall(r"({[^{}]+image\":{[^}]+}[^}]+})", webpage)
 506            )
 507
 508         if len(jsent) == 0:
 509            raise ExtractorError('No video found', expected=True)
 510
 511         for jent in jsent:
 512             programme_id = jent.get('externalId')
 513             xml_url = jent.get('href')
 514
 515             title = jent.get('caption',list_title)
 516
 517             duration = parse_duration(jent.get('duration'))
 518             description = list_title
 519             if jent.get('caption'):
 520                description += ' - ' + jent.get('caption')
 521             thumbnail = None
 522             if jent.has_key('image'):
 523                thumbnail=jent['image'].get('href')
 524
 525             formats = []
 526             subtitles = []
 527
 528             if programme_id:
 529                formats, subtitles = self._download_media_selector(programme_id)
 530             elif jent.has_key('sourceFiles'):
 531                # mediaselector not used at
 532                # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu
 533                for key, val in jent['sourceFiles'].items():
 534                   formats.append( {
 535                      'ext': val.get('encoding'),
 536                      'url': val.get('url'),
 537                      'filesize': int(val.get('filesize')),
 538                      'format_id': key
 539                   } )
 540             elif xml_url:
 541                # Cheap fallback
 542                # http://playlists.bbc.co.uk/news/(list_id)[ABC..]/playlist.sxml
 543                xml = self._download_webpage(xml_url, programme_id, 'Downloading playlist.sxml for externalId (fallback)')
 544                programme_id = self._search_regex(r'<mediator [^>]*identifier="(.+?)"', xml, 'playlist.sxml (externalId fallback)')
 545                formats, subtitles = self._download_media_selector(programme_id)
 546
 547             if len(formats) == 0:
 548                raise ExtractorError('unsupported json media entry.\n    '+str(jent)+'\n')
 549
 550             self._sort_formats(formats)
 551
 552             id = jent.get('id') if programme_id == None else programme_id
 553             if id == None:
 554                id = 'NA'
 555
 556             ret.append( {
 557                 'id': id,
 558                 'uploader': 'BBC News',
 559                 'upload_date': pubdate,
 560                 'title': title,
 561                 'description': description,
 562                 'thumbnail': thumbnail,
 563                 'duration': duration,
 564                 'formats': formats,
 565                 'subtitles': subtitles,
 566             } )
 567
 568         if len(ret) > 0:
 569            return self.playlist_result(ret, list_id, list_title)
 570         raise ExtractorError('No video found', expected=True)