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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_urllib_parse_urlparse
   8 from ..utils import (
   9     ExtractorError,
  10     parse_iso8601,
  11     qualities,
  12 )
  13
  14
  15 class SRGSSRIE(InfoExtractor):
  16     _VALID_URL = r'(?:https?://tp\.srgssr\.ch/p(?:/[^/]+)+\?urn=urn|srgssr):(?P<bu>srf|rts|rsi|rtr|swi):(?:[^:]+:)?(?P<type>video|audio):(?P<id>[0-9a-f\-]{36}|\d+)'
  17     _GEO_BYPASS = False
  18     _GEO_COUNTRIES = ['CH']
  19
  20     _ERRORS = {
  21         'AGERATING12': 'To protect children under the age of 12, this video is only available between 8 p.m. and 6 a.m.',
  22         'AGERATING18': 'To protect children under the age of 18, this video is only available between 11 p.m. and 5 a.m.',
  23         # 'ENDDATE': 'For legal reasons, this video was only available for a specified period of time.',
  24         'GEOBLOCK': 'For legal reasons, this video is only available in Switzerland.',
  25         'LEGAL': 'The video cannot be transmitted for legal reasons.',
  26         'STARTDATE': 'This video is not yet available. Please try again later.',
  27     }
  28
  29     def _get_tokenized_src(self, url, video_id, format_id):
  30         sp = compat_urllib_parse_urlparse(url).path.split('/')
  31         token = self._download_json(
  32             'http://tp.srgssr.ch/akahd/token?acl=/%s/%s/*' % (sp[1], sp[2]),
  33             video_id, 'Downloading %s token' % format_id, fatal=False) or {}
  34         auth_params = token.get('token', {}).get('authparams')
  35         if auth_params:
  36             url += '?' + auth_params
  37         return url
  38
  39     def get_media_data(self, bu, media_type, media_id):
  40         media_data = self._download_json(
  41             'http://il.srgssr.ch/integrationlayer/1.0/ue/%s/%s/play/%s.json' % (bu, media_type, media_id),
  42             media_id)[media_type.capitalize()]
  43
  44         if media_data.get('block') and media_data['block'] in self._ERRORS:
  45             message = self._ERRORS[media_data['block']]
  46             if media_data['block'] == 'GEOBLOCK':
  47                 self.raise_geo_restricted(
  48                     msg=message, countries=self._GEO_COUNTRIES)
  49             raise ExtractorError(
  50                 '%s said: %s' % (self.IE_NAME, message), expected=True)
  51
  52         return media_data
  53
  54     def _real_extract(self, url):
  55         bu, media_type, media_id = re.match(self._VALID_URL, url).groups()
  56
  57         media_data = self.get_media_data(bu, media_type, media_id)
  58
  59         metadata = media_data['AssetMetadatas']['AssetMetadata'][0]
  60         title = metadata['title']
  61         description = metadata.get('description')
  62         created_date = media_data.get('createdDate') or metadata.get('createdDate')
  63         timestamp = parse_iso8601(created_date)
  64
  65         thumbnails = [{
  66             'id': image.get('id'),
  67             'url': image['url'],
  68         } for image in media_data.get('Image', {}).get('ImageRepresentations', {}).get('ImageRepresentation', [])]
  69
  70         preference = qualities(['LQ', 'MQ', 'SD', 'HQ', 'HD'])
  71         formats = []
  72         for source in media_data.get('Playlists', {}).get('Playlist', []) + media_data.get('Downloads', {}).get('Download', []):
  73             protocol = source.get('@protocol')
  74             for asset in source['url']:
  75                 asset_url = asset['text']
  76                 quality = asset['@quality']
  77                 format_id = '%s-%s' % (protocol, quality)
  78                 if protocol.startswith('HTTP-HDS') or protocol.startswith('HTTP-HLS'):
  79                     asset_url = self._get_tokenized_src(asset_url, media_id, format_id)
  80                     if protocol.startswith('HTTP-HDS'):
  81                         formats.extend(self._extract_f4m_formats(
  82                             asset_url + ('?' if '?' not in asset_url else '&') + 'hdcore=3.4.0',
  83                             media_id, f4m_id=format_id, fatal=False))
  84                     elif protocol.startswith('HTTP-HLS'):
  85                         formats.extend(self._extract_m3u8_formats(
  86                             asset_url, media_id, 'mp4', 'm3u8_native',
  87                             m3u8_id=format_id, fatal=False))
  88                 else:
  89                     formats.append({
  90                         'format_id': format_id,
  91                         'url': asset_url,
  92                         'preference': preference(quality),
  93                         'ext': 'flv' if protocol == 'RTMP' else None,
  94                     })
  95         self._sort_formats(formats)
  96
  97         return {
  98             'id': media_id,
  99             'title': title,
 100             'description': description,
 101             'timestamp': timestamp,
 102             'thumbnails': thumbnails,
 103             'formats': formats,
 104         }
 105
 106
 107 class SRGSSRPlayIE(InfoExtractor):
 108     IE_DESC = 'srf.ch, rts.ch, rsi.ch, rtr.ch and swissinfo.ch play sites'
 109     _VALID_URL = r'https?://(?:(?:www|play)\.)?(?P<bu>srf|rts|rsi|rtr|swissinfo)\.ch/play/(?:tv|radio)/[^/]+/(?P<type>video|audio)/[^?]+\?id=(?P<id>[0-9a-f\-]{36}|\d+)'
 110
 111     _TESTS = [{
 112         'url': 'http://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5',
 113         'md5': 'da6b5b3ac9fa4761a942331cef20fcb3',
 114         'info_dict': {
 115             'id': '28e1a57d-5b76-4399-8ab3-9097f071e6c5',
 116             'ext': 'mp4',
 117             'upload_date': '20130701',
 118             'title': 'Snowden beantragt Asyl in Russland',
 119             'timestamp': 1372713995,
 120         }
 121     }, {
 122         # No Speichern (Save) button
 123         'url': 'http://www.srf.ch/play/tv/top-gear/video/jaguar-xk120-shadow-und-tornado-dampflokomotive?id=677f5829-e473-4823-ac83-a1087fe97faa',
 124         'md5': '0a274ce38fda48c53c01890651985bc6',
 125         'info_dict': {
 126             'id': '677f5829-e473-4823-ac83-a1087fe97faa',
 127             'ext': 'flv',
 128             'upload_date': '20130710',
 129             'title': 'Jaguar XK120, Shadow und Tornado-Dampflokomotive',
 130             'description': 'md5:88604432b60d5a38787f152dec89cd56',
 131             'timestamp': 1373493600,
 132         },
 133     }, {
 134         'url': 'http://www.rtr.ch/play/radio/actualitad/audio/saira-tujetsch-tuttina-cuntinuar-cun-sedrun-muster-turissem?id=63cb0778-27f8-49af-9284-8c7a8c6d15fc',
 135         'info_dict': {
 136             'id': '63cb0778-27f8-49af-9284-8c7a8c6d15fc',
 137             'ext': 'mp3',
 138             'upload_date': '20151013',
 139             'title': 'Saira: Tujetsch - tuttina cuntinuar cun Sedrun Mustér Turissem',
 140             'timestamp': 1444750398,
 141         },
 142         'params': {
 143             # rtmp download
 144             'skip_download': True,
 145         },
 146     }, {
 147         'url': 'http://www.rts.ch/play/tv/-/video/le-19h30?id=6348260',
 148         'md5': '67a2a9ae4e8e62a68d0e9820cc9782df',
 149         'info_dict': {
 150             'id': '6348260',
 151             'display_id': '6348260',
 152             'ext': 'mp4',
 153             'duration': 1796,
 154             'title': 'Le 19h30',
 155             'description': '',
 156             'uploader': '19h30',
 157             'upload_date': '20141201',
 158             'timestamp': 1417458600,
 159             'thumbnail': r're:^https?://.*\.image',
 160             'view_count': int,
 161         },
 162         'params': {
 163             # m3u8 download
 164             'skip_download': True,
 165         }
 166     }]
 167
 168     def _real_extract(self, url):
 169         bu, media_type, media_id = re.match(self._VALID_URL, url).groups()
 170         # other info can be extracted from url + '&layout=json'
 171         return self.url_result('srgssr:%s:%s:%s' % (bu[:3], media_type, media_id), 'SRGSSR')