yt_dlp/extractor/srgssr.py

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