youtube_dl/extractor/bbccouk.py

   1 from __future__ import unicode_literals
   2
   3 import xml.etree.ElementTree
   4
   5 from .subtitles import SubtitlesInfoExtractor
   6 from ..utils import ExtractorError
   7 from ..compat import compat_HTTPError
   8
   9
  10 class BBCCoUkIE(SubtitlesInfoExtractor):
  11     IE_NAME = 'bbc.co.uk'
  12     IE_DESC = 'BBC iPlayer'
  13     _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:programmes|iplayer/episode)/(?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': 'md5:db4755d7a665ae72343779f7dacb402c',
  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     ]
  76
  77     def _extract_asx_playlist(self, connection, programme_id):
  78         asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
  79         return [ref.get('href') for ref in asx.findall('./Entry/ref')]
  80
  81     def _extract_connection(self, connection, programme_id):
  82         formats = []
  83         protocol = connection.get('protocol')
  84         supplier = connection.get('supplier')
  85         if protocol == 'http':
  86             href = connection.get('href')
  87             # ASX playlist
  88             if supplier == 'asx':
  89                 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
  90                     formats.append({
  91                         'url': ref,
  92                         'format_id': 'ref%s_%s' % (i, supplier),
  93                     })
  94             # Direct link
  95             else:
  96                 formats.append({
  97                     'url': href,
  98                     'format_id': supplier,
  99                 })
 100         elif protocol == 'rtmp':
 101             application = connection.get('application', 'ondemand')
 102             auth_string = connection.get('authString')
 103             identifier = connection.get('identifier')
 104             server = connection.get('server')
 105             formats.append({
 106                 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
 107                 'play_path': identifier,
 108                 'app': '%s?%s' % (application, auth_string),
 109                 'page_url': 'http://www.bbc.co.uk',
 110                 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
 111                 'rtmp_live': False,
 112                 'ext': 'flv',
 113                 'format_id': supplier,
 114             })
 115         return formats
 116
 117     def _extract_items(self, playlist):
 118         return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
 119
 120     def _extract_medias(self, media_selection):
 121         error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error')
 122         if error is not None:
 123             raise ExtractorError(
 124                 '%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True)
 125         return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
 126
 127     def _extract_connections(self, media):
 128         return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
 129
 130     def _extract_video(self, media, programme_id):
 131         formats = []
 132         vbr = int(media.get('bitrate'))
 133         vcodec = media.get('encoding')
 134         service = media.get('service')
 135         width = int(media.get('width'))
 136         height = int(media.get('height'))
 137         file_size = int(media.get('media_file_size'))
 138         for connection in self._extract_connections(media):
 139             conn_formats = self._extract_connection(connection, programme_id)
 140             for format in conn_formats:
 141                 format.update({
 142                     'format_id': '%s_%s' % (service, format['format_id']),
 143                     'width': width,
 144                     'height': height,
 145                     'vbr': vbr,
 146                     'vcodec': vcodec,
 147                     'filesize': file_size,
 148                 })
 149             formats.extend(conn_formats)
 150         return formats
 151
 152     def _extract_audio(self, media, programme_id):
 153         formats = []
 154         abr = int(media.get('bitrate'))
 155         acodec = media.get('encoding')
 156         service = media.get('service')
 157         for connection in self._extract_connections(media):
 158             conn_formats = self._extract_connection(connection, programme_id)
 159             for format in conn_formats:
 160                 format.update({
 161                     'format_id': '%s_%s' % (service, format['format_id']),
 162                     'abr': abr,
 163                     'acodec': acodec,
 164                 })
 165             formats.extend(conn_formats)
 166         return formats
 167
 168     def _extract_captions(self, media, programme_id):
 169         subtitles = {}
 170         for connection in self._extract_connections(media):
 171             captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
 172             lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
 173             ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}'))
 174             srt = ''
 175             for pos, p in enumerate(ps):
 176                 srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'),
 177                                                           p.text.strip() if p.text is not None else '')
 178             subtitles[lang] = srt
 179         return subtitles
 180
 181     def _download_media_selector(self, programme_id):
 182         try:
 183             media_selection = self._download_xml(
 184                 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s' % programme_id,
 185                 programme_id, 'Downloading media selection XML')
 186         except ExtractorError as ee:
 187             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 188                 media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().encode('utf-8'))
 189             else:
 190                 raise
 191
 192         formats = []
 193         subtitles = None
 194
 195         for media in self._extract_medias(media_selection):
 196             kind = media.get('kind')
 197             if kind == 'audio':
 198                 formats.extend(self._extract_audio(media, programme_id))
 199             elif kind == 'video':
 200                 formats.extend(self._extract_video(media, programme_id))
 201             elif kind == 'captions':
 202                 subtitles = self._extract_captions(media, programme_id)
 203
 204         return formats, subtitles
 205
 206     def _real_extract(self, url):
 207         group_id = self._match_id(url)
 208
 209         webpage = self._download_webpage(url, group_id, 'Downloading video page')
 210
 211         programme_id = self._search_regex(
 212             r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False)
 213         if programme_id:
 214             player = self._download_json(
 215                 'http://www.bbc.co.uk/iplayer/episode/%s.json' % group_id,
 216                 group_id)['jsConf']['player']
 217             title = player['title']
 218             description = player['subtitle']
 219             duration = player['duration']
 220             formats, subtitles = self._download_media_selector(programme_id)
 221         else:
 222             playlist = self._download_xml(
 223                 'http://www.bbc.co.uk/iplayer/playlist/%s' % group_id,
 224                 group_id, 'Downloading playlist XML')
 225
 226             no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
 227             if no_items is not None:
 228                 reason = no_items.get('reason')
 229                 if reason == 'preAvailability':
 230                     msg = 'Episode %s is not yet available' % group_id
 231                 elif reason == 'postAvailability':
 232                     msg = 'Episode %s is no longer available' % group_id
 233                 elif reason == 'noMedia':
 234                     msg = 'Episode %s is not currently available' % group_id
 235                 else:
 236                     msg = 'Episode %s is not available: %s' % (group_id, reason)
 237                 raise ExtractorError(msg, expected=True)
 238
 239             for item in self._extract_items(playlist):
 240                 kind = item.get('kind')
 241                 if kind != 'programme' and kind != 'radioProgramme':
 242                     continue
 243                 title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
 244                 description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
 245                 programme_id = item.get('identifier')
 246                 duration = int(item.get('duration'))
 247                 formats, subtitles = self._download_media_selector(programme_id)
 248
 249         if self._downloader.params.get('listsubtitles', False):
 250             self._list_available_subtitles(programme_id, subtitles)
 251             return
 252
 253         self._sort_formats(formats)
 254
 255         return {
 256             'id': programme_id,
 257             'title': title,
 258             'description': description,
 259             'duration': duration,
 260             'formats': formats,
 261             'subtitles': subtitles,
 262         }