yt_dlp/extractor/srgssr.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     ExtractorError,
   4     float_or_none,
   5     int_or_none,
   6     join_nonempty,
   7     parse_iso8601,
   8     qualities,
   9     try_get,
  10 )
  11
  12
  13 class SRGSSRIE(InfoExtractor):
  14     _VALID_URL = r'''(?x)
  15                     (?:
  16                         https?://tp\.srgssr\.ch/p(?:/[^/]+)+\?urn=urn|
  17                         srgssr
  18                     ):
  19                     (?P<bu>
  20                         srf|rts|rsi|rtr|swi
  21                     ):(?:[^:]+:)?
  22                     (?P<type>
  23                         video|audio
  24                     ):
  25                     (?P<id>
  26                         [0-9a-f\-]{36}|\d+
  27                     )
  28                     '''
  29     _GEO_BYPASS = False
  30     _GEO_COUNTRIES = ['CH']
  31
  32     _ERRORS = {
  33         'AGERATING12': 'To protect children under the age of 12, this video is only available between 8 p.m. and 6 a.m.',
  34         'AGERATING18': 'To protect children under the age of 18, this video is only available between 11 p.m. and 5 a.m.',
  35         # 'ENDDATE': 'For legal reasons, this video was only available for a specified period of time.',
  36         'GEOBLOCK': 'For legal reasons, this video is only available in Switzerland.',
  37         'LEGAL': 'The video cannot be transmitted for legal reasons.',
  38         'STARTDATE': 'This video is not yet available. Please try again later.',
  39     }
  40     _DEFAULT_LANGUAGE_CODES = {
  41         'srf': 'de',
  42         'rts': 'fr',
  43         'rsi': 'it',
  44         'rtr': 'rm',
  45         'swi': 'en',
  46     }
  47
  48     def _get_tokenized_src(self, url, video_id, format_id):
  49         token = self._download_json(
  50             'http://tp.srgssr.ch/akahd/token?acl=*',
  51             video_id, 'Downloading %s token' % format_id, fatal=False) or {}
  52         auth_params = try_get(token, lambda x: x['token']['authparams'])
  53         if auth_params:
  54             url += ('?' if '?' not in url else '&') + auth_params
  55         return url
  56
  57     def _get_media_data(self, bu, media_type, media_id):
  58         query = {'onlyChapters': True} if media_type == 'video' else {}
  59         full_media_data = self._download_json(
  60             'https://il.srgssr.ch/integrationlayer/2.0/%s/mediaComposition/%s/%s.json'
  61             % (bu, media_type, media_id),
  62             media_id, query=query)['chapterList']
  63         try:
  64             media_data = next(
  65                 x for x in full_media_data if x.get('id') == media_id)
  66         except StopIteration:
  67             raise ExtractorError('No media information found')
  68
  69         block_reason = media_data.get('blockReason')
  70         if block_reason and block_reason in self._ERRORS:
  71             message = self._ERRORS[block_reason]
  72             if block_reason == 'GEOBLOCK':
  73                 self.raise_geo_restricted(
  74                     msg=message, countries=self._GEO_COUNTRIES)
  75             raise ExtractorError(
  76                 '%s said: %s' % (self.IE_NAME, message), expected=True)
  77
  78         return media_data
  79
  80     def _real_extract(self, url):
  81         bu, media_type, media_id = self._match_valid_url(url).groups()
  82         media_data = self._get_media_data(bu, media_type, media_id)
  83         title = media_data['title']
  84
  85         formats = []
  86         subtitles = {}
  87         q = qualities(['SD', 'HD'])
  88         for source in (media_data.get('resourceList') or []):
  89             format_url = source.get('url')
  90             if not format_url:
  91                 continue
  92             protocol = source.get('protocol')
  93             quality = source.get('quality')
  94             format_id = join_nonempty(protocol, source.get('encoding'), quality)
  95
  96             if protocol in ('HDS', 'HLS'):
  97                 if source.get('tokenType') == 'AKAMAI':
  98                     format_url = self._get_tokenized_src(
  99                         format_url, media_id, format_id)
 100                     fmts, subs = self._extract_akamai_formats_and_subtitles(
 101                         format_url, media_id)
 102                     formats.extend(fmts)
 103                     subtitles = self._merge_subtitles(subtitles, subs)
 104                 elif protocol == 'HLS':
 105                     m3u8_fmts, m3u8_subs = self._extract_m3u8_formats_and_subtitles(
 106                         format_url, media_id, 'mp4', 'm3u8_native',
 107                         m3u8_id=format_id, fatal=False)
 108                     formats.extend(m3u8_fmts)
 109                     subtitles = self._merge_subtitles(subtitles, m3u8_subs)
 110             elif protocol in ('HTTP', 'HTTPS'):
 111                 formats.append({
 112                     'format_id': format_id,
 113                     'url': format_url,
 114                     'quality': q(quality),
 115                 })
 116
 117         # This is needed because for audio medias the podcast url is usually
 118         # always included, even if is only an audio segment and not the
 119         # whole episode.
 120         if int_or_none(media_data.get('position')) == 0:
 121             for p in ('S', 'H'):
 122                 podcast_url = media_data.get('podcast%sdUrl' % p)
 123                 if not podcast_url:
 124                     continue
 125                 quality = p + 'D'
 126                 formats.append({
 127                     'format_id': 'PODCAST-' + quality,
 128                     'url': podcast_url,
 129                     'quality': q(quality),
 130                 })
 131
 132         if media_type == 'video':
 133             for sub in (media_data.get('subtitleList') or []):
 134                 sub_url = sub.get('url')
 135                 if not sub_url:
 136                     continue
 137                 lang = sub.get('locale') or self._DEFAULT_LANGUAGE_CODES[bu]
 138                 subtitles.setdefault(lang, []).append({
 139                     'url': sub_url,
 140                 })
 141
 142         return {
 143             'id': media_id,
 144             'title': title,
 145             'description': media_data.get('description'),
 146             'timestamp': parse_iso8601(media_data.get('date')),
 147             'thumbnail': media_data.get('imageUrl'),
 148             'duration': float_or_none(media_data.get('duration'), 1000),
 149             'subtitles': subtitles,
 150             'formats': formats,
 151         }
 152
 153
 154 class SRGSSRPlayIE(InfoExtractor):
 155     IE_DESC = 'srf.ch, rts.ch, rsi.ch, rtr.ch and swissinfo.ch play sites'
 156     _VALID_URL = r'''(?x)
 157                     https?://
 158                         (?:(?:www|play)\.)?
 159                         (?P<bu>srf|rts|rsi|rtr|swissinfo)\.ch/play/(?:tv|radio)/
 160                         (?:
 161                             [^/]+/(?P<type>video|audio)/[^?]+|
 162                             popup(?P<type_2>video|audio)player
 163                         )
 164                         \?.*?\b(?:id=|urn=urn:[^:]+:video:)(?P<id>[0-9a-f\-]{36}|\d+)
 165                     '''
 166
 167     _TESTS = [{
 168         'url': 'http://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5',
 169         'md5': '6db2226ba97f62ad42ce09783680046c',
 170         'info_dict': {
 171             'id': '28e1a57d-5b76-4399-8ab3-9097f071e6c5',
 172             'ext': 'mp4',
 173             'upload_date': '20130701',
 174             'title': 'Snowden beantragt Asyl in Russland',
 175             'timestamp': 1372708215,
 176             'duration': 113.827,
 177             'thumbnail': r're:^https?://.*1383719781\.png$',
 178         },
 179         'expected_warnings': ['Unable to download f4m manifest'],
 180     }, {
 181         'url': 'http://www.rtr.ch/play/radio/actualitad/audio/saira-tujetsch-tuttina-cuntinuar-cun-sedrun-muster-turissem?id=63cb0778-27f8-49af-9284-8c7a8c6d15fc',
 182         'info_dict': {
 183             'id': '63cb0778-27f8-49af-9284-8c7a8c6d15fc',
 184             'ext': 'mp3',
 185             'upload_date': '20151013',
 186             'title': 'Saira: Tujetsch - tuttina cuntinuar cun Sedrun Mustér Turissem',
 187             'timestamp': 1444709160,
 188             'duration': 336.816,
 189         },
 190         'params': {
 191             # rtmp download
 192             'skip_download': True,
 193         },
 194     }, {
 195         'url': 'http://www.rts.ch/play/tv/-/video/le-19h30?id=6348260',
 196         'md5': '67a2a9ae4e8e62a68d0e9820cc9782df',
 197         'info_dict': {
 198             'id': '6348260',
 199             'display_id': '6348260',
 200             'ext': 'mp4',
 201             'duration': 1796.76,
 202             'title': 'Le 19h30',
 203             'upload_date': '20141201',
 204             'timestamp': 1417458600,
 205             'thumbnail': r're:^https?://.*\.image',
 206         },
 207         'params': {
 208             # m3u8 download
 209             'skip_download': True,
 210         }
 211     }, {
 212         'url': 'http://play.swissinfo.ch/play/tv/business/video/why-people-were-against-tax-reforms?id=42960270',
 213         'info_dict': {
 214             'id': '42960270',
 215             'ext': 'mp4',
 216             'title': 'Why people were against tax reforms',
 217             'description': 'md5:7ac442c558e9630e947427469c4b824d',
 218             'duration': 94.0,
 219             'upload_date': '20170215',
 220             'timestamp': 1487173560,
 221             'thumbnail': r're:https?://www\.swissinfo\.ch/srgscalableimage/42961964',
 222             'subtitles': 'count:9',
 223         },
 224         'params': {
 225             'skip_download': True,
 226         }
 227     }, {
 228         'url': 'https://www.srf.ch/play/tv/popupvideoplayer?id=c4dba0ca-e75b-43b2-a34f-f708a4932e01',
 229         'only_matching': True,
 230     }, {
 231         'url': 'https://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?urn=urn:srf:video:28e1a57d-5b76-4399-8ab3-9097f071e6c5',
 232         'only_matching': True,
 233     }, {
 234         'url': 'https://www.rts.ch/play/tv/19h30/video/le-19h30?urn=urn:rts:video:6348260',
 235         'only_matching': True,
 236     }, {
 237         # audio segment, has podcastSdUrl of the full episode
 238         'url': 'https://www.srf.ch/play/radio/popupaudioplayer?id=50b20dc8-f05b-4972-bf03-e438ff2833eb',
 239         'only_matching': True,
 240     }]
 241
 242     def _real_extract(self, url):
 243         mobj = self._match_valid_url(url)
 244         bu = mobj.group('bu')
 245         media_type = mobj.group('type') or mobj.group('type_2')
 246         media_id = mobj.group('id')
 247         return self.url_result('srgssr:%s:%s:%s' % (bu[:3], media_type, media_id), 'SRGSSR')