Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/bbccouk.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import xml.etree.ElementTree
   5
   6 from .subtitles import SubtitlesInfoExtractor
   7 from ..utils import ExtractorError
   8 from ..compat import compat_HTTPError
   9
  10
  11 class BBCCoUkIE(SubtitlesInfoExtractor):
  12     IE_NAME = 'bbc.co.uk'
  13     IE_DESC = 'BBC iPlayer'
  14     _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:programmes|iplayer/episode)/(?P<id>[\da-z]{8})'
  15
  16     _TESTS = [
  17         {
  18             'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
  19             'info_dict': {
  20                 'id': 'b039d07m',
  21                 'ext': 'flv',
  22                 'title': 'Kaleidoscope: Leonard Cohen',
  23                 'description': 'md5:db4755d7a665ae72343779f7dacb402c',
  24                 'duration': 1740,
  25             },
  26             'params': {
  27                 # rtmp download
  28                 'skip_download': True,
  29             }
  30         },
  31         {
  32             'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
  33             'info_dict': {
  34                 'id': 'b00yng1d',
  35                 'ext': 'flv',
  36                 'title': 'The Man in Black: Series 3: The Printed Name',
  37                 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
  38                 'duration': 1800,
  39             },
  40             'params': {
  41                 # rtmp download
  42                 'skip_download': True,
  43             },
  44             'skip': 'Episode is no longer available on BBC iPlayer Radio',
  45         },
  46         {
  47             'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
  48             'info_dict': {
  49                 'id': 'b00yng1d',
  50                 'ext': 'flv',
  51                 'title': 'The Voice UK: Series 3: Blind Auditions 5',
  52                 '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.",
  53                 'duration': 5100,
  54             },
  55             'params': {
  56                 # rtmp download
  57                 'skip_download': True,
  58             },
  59             'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
  60         },
  61         {
  62             'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
  63             'info_dict': {
  64                 'id': 'b03k3pb7',
  65                 'ext': 'flv',
  66                 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
  67                 'description': '2. Invasion',
  68                 'duration': 3600,
  69             },
  70             'params': {
  71                 # rtmp download
  72                 'skip_download': True,
  73             },
  74             'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
  75         },
  76     ]
  77
  78     def _extract_asx_playlist(self, connection, programme_id):
  79         asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
  80         return [ref.get('href') for ref in asx.findall('./Entry/ref')]
  81
  82     def _extract_connection(self, connection, programme_id):
  83         formats = []
  84         protocol = connection.get('protocol')
  85         supplier = connection.get('supplier')
  86         if protocol == 'http':
  87             href = connection.get('href')
  88             # ASX playlist
  89             if supplier == 'asx':
  90                 for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
  91                     formats.append({
  92                         'url': ref,
  93                         'format_id': 'ref%s_%s' % (i, supplier),
  94                     })
  95             # Direct link
  96             else:
  97                 formats.append({
  98                     'url': href,
  99                     'format_id': supplier,
 100                 })
 101         elif protocol == 'rtmp':
 102             application = connection.get('application', 'ondemand')
 103             auth_string = connection.get('authString')
 104             identifier = connection.get('identifier')
 105             server = connection.get('server')
 106             formats.append({
 107                 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
 108                 'play_path': identifier,
 109                 'app': '%s?%s' % (application, auth_string),
 110                 'page_url': 'http://www.bbc.co.uk',
 111                 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
 112                 'rtmp_live': False,
 113                 'ext': 'flv',
 114                 'format_id': supplier,
 115             })
 116         return formats
 117
 118     def _extract_items(self, playlist):
 119         return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
 120
 121     def _extract_medias(self, media_selection):
 122         error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error')
 123         if error is not None:
 124             raise ExtractorError(
 125                 '%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True)
 126         return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
 127
 128     def _extract_connections(self, media):
 129         return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
 130
 131     def _extract_video(self, media, programme_id):
 132         formats = []
 133         vbr = int(media.get('bitrate'))
 134         vcodec = media.get('encoding')
 135         service = media.get('service')
 136         width = int(media.get('width'))
 137         height = int(media.get('height'))
 138         file_size = int(media.get('media_file_size'))
 139         for connection in self._extract_connections(media):
 140             conn_formats = self._extract_connection(connection, programme_id)
 141             for format in conn_formats:
 142                 format.update({
 143                     'format_id': '%s_%s' % (service, format['format_id']),
 144                     'width': width,
 145                     'height': height,
 146                     'vbr': vbr,
 147                     'vcodec': vcodec,
 148                     'filesize': file_size,
 149                 })
 150             formats.extend(conn_formats)
 151         return formats
 152
 153     def _extract_audio(self, media, programme_id):
 154         formats = []
 155         abr = int(media.get('bitrate'))
 156         acodec = media.get('encoding')
 157         service = media.get('service')
 158         for connection in self._extract_connections(media):
 159             conn_formats = self._extract_connection(connection, programme_id)
 160             for format in conn_formats:
 161                 format.update({
 162                     'format_id': '%s_%s' % (service, format['format_id']),
 163                     'abr': abr,
 164                     'acodec': acodec,
 165                 })
 166             formats.extend(conn_formats)
 167         return formats
 168
 169     def _extract_captions(self, media, programme_id):
 170         subtitles = {}
 171         for connection in self._extract_connections(media):
 172             captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
 173             lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
 174             ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}'))
 175             srt = ''
 176             for pos, p in enumerate(ps):
 177                 srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'),
 178                                                           p.text.strip() if p.text is not None else '')
 179             subtitles[lang] = srt
 180         return subtitles
 181
 182     def _download_media_selector(self, programme_id):
 183         try:
 184             media_selection = self._download_xml(
 185                 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s' % programme_id,
 186                 programme_id, 'Downloading media selection XML')
 187         except ExtractorError as ee:
 188             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 189                 media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().encode('utf-8'))
 190             else:
 191                 raise
 192
 193         formats = []
 194         subtitles = None
 195
 196         for media in self._extract_medias(media_selection):
 197             kind = media.get('kind')
 198             if kind == 'audio':
 199                 formats.extend(self._extract_audio(media, programme_id))
 200             elif kind == 'video':
 201                 formats.extend(self._extract_video(media, programme_id))
 202             elif kind == 'captions':
 203                 subtitles = self._extract_captions(media, programme_id)
 204
 205         return formats, subtitles
 206
 207     def _real_extract(self, url):
 208         group_id = self._match_id(url)
 209
 210         webpage = self._download_webpage(url, group_id, 'Downloading video page')
 211
 212         programme_id = self._search_regex(
 213             r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False)
 214         if programme_id:
 215             player = self._download_json(
 216                 'http://www.bbc.co.uk/iplayer/episode/%s.json' % group_id,
 217                 group_id)['jsConf']['player']
 218             title = player['title']
 219             description = player['subtitle']
 220             duration = player['duration']
 221             formats, subtitles = self._download_media_selector(programme_id)
 222         else:
 223             playlist = self._download_xml(
 224                 'http://www.bbc.co.uk/iplayer/playlist/%s' % group_id,
 225                 group_id, 'Downloading playlist XML')
 226
 227             no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
 228             if no_items is not None:
 229                 reason = no_items.get('reason')
 230                 if reason == 'preAvailability':
 231                     msg = 'Episode %s is not yet available' % group_id
 232                 elif reason == 'postAvailability':
 233                     msg = 'Episode %s is no longer available' % group_id
 234                 elif reason == 'noMedia':
 235                     msg = 'Episode %s is not currently available' % group_id
 236                 else:
 237                     msg = 'Episode %s is not available: %s' % (group_id, reason)
 238                 raise ExtractorError(msg, expected=True)
 239
 240             for item in self._extract_items(playlist):
 241                 kind = item.get('kind')
 242                 if kind != 'programme' and kind != 'radioProgramme':
 243                     continue
 244                 title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
 245                 description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
 246                 programme_id = item.get('identifier')
 247                 duration = int(item.get('duration'))
 248                 formats, subtitles = self._download_media_selector(programme_id)
 249
 250         if self._downloader.params.get('listsubtitles', False):
 251             self._list_available_subtitles(programme_id, subtitles)
 252             return
 253
 254         self._sort_formats(formats)
 255
 256         return {
 257             'id': programme_id,
 258             'title': title,
 259             'description': description,
 260             'duration': duration,
 261             'formats': formats,
 262             'subtitles': subtitles,
 263         }