youtube_dl/extractor/bbccouk.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .subtitles import SubtitlesInfoExtractor
   6 from ..utils import ExtractorError
   7
   8
   9 class BBCCoUkIE(SubtitlesInfoExtractor):
  10     IE_NAME = 'bbc.co.uk'
  11     IE_DESC = 'BBC iPlayer'
  12     _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:programmes|iplayer/episode)/(?P<id>[\da-z]{8})'
  13
  14     _TESTS = [
  15         {
  16             'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
  17             'info_dict': {
  18                 'id': 'b039d07m',
  19                 'ext': 'flv',
  20                 'title': 'Kaleidoscope: Leonard Cohen',
  21                 'description': 'md5:db4755d7a665ae72343779f7dacb402c',
  22                 'duration': 1740,
  23             },
  24             'params': {
  25                 # rtmp download
  26                 'skip_download': True,
  27             }
  28         },
  29         {
  30             'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
  31             'info_dict': {
  32                 'id': 'b00yng1d',
  33                 'ext': 'flv',
  34                 'title': 'The Man in Black: Series 3: The Printed Name',
  35                 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
  36                 'duration': 1800,
  37             },
  38             'params': {
  39                 # rtmp download
  40                 'skip_download': True,
  41             },
  42             'skip': 'Episode is no longer available on BBC iPlayer Radio',
  43         },
  44         {
  45             'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
  46             'info_dict': {
  47                 'id': 'b00yng1d',
  48                 'ext': 'flv',
  49                 'title': 'The Voice UK: Series 3: Blind Auditions 5',
  50                 '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.",
  51                 'duration': 5100,
  52             },
  53             'params': {
  54                 # rtmp download
  55                 'skip_download': True,
  56             },
  57             'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
  58         }
  59     ]
  60
  61     def _extract_asx_playlist(self, connection, programme_id):
  62         asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
  63         return [ref.get('href') for ref in asx.findall('./Entry/ref')]
  64
  65     def _extract_connection(self, connection, programme_id):
  66         formats = []
  67         protocol = connection.get('protocol')
  68         supplier = connection.get('supplier')
  69         if protocol == 'http':
  70             href = connection.get('href')
  71             # ASX playlist
  72             if supplier == 'asx':
  73                 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
  74                     formats.append({
  75                         'url': ref,
  76                         'format_id': 'ref%s_%s' % (i, supplier),
  77                     })
  78             # Direct link
  79             else:
  80                 formats.append({
  81                     'url': href,
  82                     'format_id': supplier,
  83                 })
  84         elif protocol == 'rtmp':
  85             application = connection.get('application', 'ondemand')
  86             auth_string = connection.get('authString')
  87             identifier = connection.get('identifier')
  88             server = connection.get('server')
  89             formats.append({
  90                 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
  91                 'play_path': identifier,
  92                 'app': '%s?%s' % (application, auth_string),
  93                 'page_url': 'http://www.bbc.co.uk',
  94                 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
  95                 'rtmp_live': False,
  96                 'ext': 'flv',
  97                 'format_id': supplier,
  98             })
  99         return formats
 100
 101     def _extract_items(self, playlist):
 102         return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
 103
 104     def _extract_medias(self, media_selection):
 105         return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
 106
 107     def _extract_connections(self, media):
 108         return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
 109
 110     def _extract_video(self, media, programme_id):
 111         formats = []
 112         vbr = int(media.get('bitrate'))
 113         vcodec = media.get('encoding')
 114         service = media.get('service')
 115         width = int(media.get('width'))
 116         height = int(media.get('height'))
 117         file_size = int(media.get('media_file_size'))
 118         for connection in self._extract_connections(media):
 119             conn_formats = self._extract_connection(connection, programme_id)
 120             for format in conn_formats:
 121                 format.update({
 122                     'format_id': '%s_%s' % (service, format['format_id']),
 123                     'width': width,
 124                     'height': height,
 125                     'vbr': vbr,
 126                     'vcodec': vcodec,
 127                     'filesize': file_size,
 128                 })
 129             formats.extend(conn_formats)
 130         return formats
 131
 132     def _extract_audio(self, media, programme_id):
 133         formats = []
 134         abr = int(media.get('bitrate'))
 135         acodec = media.get('encoding')
 136         service = media.get('service')
 137         for connection in self._extract_connections(media):
 138             conn_formats = self._extract_connection(connection, programme_id)
 139             for format in conn_formats:
 140                 format.update({
 141                     'format_id': '%s_%s' % (service, format['format_id']),
 142                     'abr': abr,
 143                     'acodec': acodec,
 144                 })
 145             formats.extend(conn_formats)
 146         return formats
 147
 148     def _extract_captions(self, media, programme_id):
 149         subtitles = {}
 150         for connection in self._extract_connections(media):
 151             captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
 152             lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
 153             ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}'))
 154             srt = ''
 155             for pos, p in enumerate(ps):
 156                 srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'),
 157                                                           p.text.strip() if p.text is not None else '')
 158             subtitles[lang] = srt
 159         return subtitles
 160
 161     def _real_extract(self, url):
 162         mobj = re.match(self._VALID_URL, url)
 163         group_id = mobj.group('id')
 164
 165         webpage = self._download_webpage(url, group_id, 'Downloading video page')
 166         if re.search(r'id="emp-error" class="notinuk">', webpage):
 167             raise ExtractorError('Currently BBC iPlayer TV programmes are available to play in the UK only',
 168                                  expected=True)
 169
 170         playlist = self._download_xml('http://www.bbc.co.uk/iplayer/playlist/%s' % group_id, group_id,
 171                                       'Downloading playlist XML')
 172
 173         no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
 174         if no_items is not None:
 175             reason = no_items.get('reason')
 176             if reason == 'preAvailability':
 177                 msg = 'Episode %s is not yet available' % group_id
 178             elif reason == 'postAvailability':
 179                 msg = 'Episode %s is no longer available' % group_id
 180             else:
 181                 msg = 'Episode %s is not available: %s' % (group_id, reason)
 182             raise ExtractorError(msg, expected=True)
 183
 184         formats = []
 185         subtitles = None
 186
 187         for item in self._extract_items(playlist):
 188             kind = item.get('kind')
 189             if kind != 'programme' and kind != 'radioProgramme':
 190                 continue
 191             title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
 192             description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
 193
 194             programme_id = item.get('identifier')
 195             duration = int(item.get('duration'))
 196
 197             media_selection = self._download_xml(
 198                 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s' % programme_id,
 199                 programme_id, 'Downloading media selection XML')
 200
 201             for media in self._extract_medias(media_selection):
 202                 kind = media.get('kind')
 203                 if kind == 'audio':
 204                     formats.extend(self._extract_audio(media, programme_id))
 205                 elif kind == 'video':
 206                     formats.extend(self._extract_video(media, programme_id))
 207                 elif kind == 'captions':
 208                     subtitles = self._extract_captions(media, programme_id)
 209
 210         if self._downloader.params.get('listsubtitles', False):
 211             self._list_available_subtitles(programme_id, subtitles)
 212             return
 213
 214         self._sort_formats(formats)
 215
 216         return {
 217             'id': programme_id,
 218             'title': title,
 219             'description': description,
 220             'duration': duration,
 221             'formats': formats,
 222             'subtitles': subtitles,
 223         }