youtube_dl/extractor/servus.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     determine_ext,
   7     float_or_none,
   8     int_or_none,
   9     unified_timestamp,
  10     urlencode_postdata,
  11     url_or_none,
  12 )
  13
  14
  15 class ServusIE(InfoExtractor):
  16     _VALID_URL = r'''(?x)
  17                     https?://
  18                         (?:www\.)?
  19                         (?:
  20                             servus\.com/(?:(?:at|de)/p/[^/]+|tv/videos)|
  21                             (?:servustv|pm-wissen)\.com/videos
  22                         )
  23                         /(?P<id>[aA]{2}-\w+|\d+-\d+)
  24                     '''
  25     _TESTS = [{
  26         # new URL schema
  27         'url': 'https://www.servustv.com/videos/aa-1t6vbu5pw1w12/',
  28         'md5': '60474d4c21f3eb148838f215c37f02b9',
  29         'info_dict': {
  30             'id': 'AA-1T6VBU5PW1W12',
  31             'ext': 'mp4',
  32             'title': 'Die Grünen aus Sicht des Volkes',
  33             'alt_title': 'Talk im Hangar-7 Voxpops Gruene',
  34             'description': 'md5:1247204d85783afe3682644398ff2ec4',
  35             'thumbnail': r're:^https?://.*\.jpg',
  36             'duration': 62.442,
  37             'timestamp': 1605193976,
  38             'upload_date': '20201112',
  39             'series': 'Talk im Hangar-7',
  40             'season': 'Season 9',
  41             'season_number': 9,
  42             'episode': 'Episode 31 - September 14',
  43             'episode_number': 31,
  44         }
  45     }, {
  46         # old URL schema
  47         'url': 'https://www.servus.com/de/p/Die-Gr%C3%BCnen-aus-Sicht-des-Volkes/AA-1T6VBU5PW1W12/',
  48         'only_matching': True,
  49     }, {
  50         'url': 'https://www.servus.com/at/p/Wie-das-Leben-beginnt/1309984137314-381415152/',
  51         'only_matching': True,
  52     }, {
  53         'url': 'https://www.servus.com/tv/videos/aa-1t6vbu5pw1w12/',
  54         'only_matching': True,
  55     }, {
  56         'url': 'https://www.servus.com/tv/videos/1380889096408-1235196658/',
  57         'only_matching': True,
  58     }, {
  59         'url': 'https://www.pm-wissen.com/videos/aa-24mus4g2w2112/',
  60         'only_matching': True,
  61     }]
  62
  63     def _real_extract(self, url):
  64         video_id = self._match_id(url).upper()
  65
  66         token = self._download_json(
  67             'https://auth.redbullmediahouse.com/token', video_id,
  68             'Downloading token', data=urlencode_postdata({
  69                 'grant_type': 'client_credentials',
  70             }), headers={
  71                 'Authorization': 'Basic SVgtMjJYNEhBNFdEM1cxMTpEdDRVSkFLd2ZOMG5IMjB1NGFBWTBmUFpDNlpoQ1EzNA==',
  72             })
  73         access_token = token['access_token']
  74         token_type = token.get('token_type', 'Bearer')
  75
  76         video = self._download_json(
  77             'https://sparkle-api.liiift.io/api/v1/stv/channels/international/assets/%s' % video_id,
  78             video_id, 'Downloading video JSON', headers={
  79                 'Authorization': '%s %s' % (token_type, access_token),
  80             })
  81
  82         formats = []
  83         thumbnail = None
  84         for resource in video['resources']:
  85             if not isinstance(resource, dict):
  86                 continue
  87             format_url = url_or_none(resource.get('url'))
  88             if not format_url:
  89                 continue
  90             extension = resource.get('extension')
  91             type_ = resource.get('type')
  92             if extension == 'jpg' or type_ == 'reference_keyframe':
  93                 thumbnail = format_url
  94                 continue
  95             ext = determine_ext(format_url)
  96             if type_ == 'dash' or ext == 'mpd':
  97                 formats.extend(self._extract_mpd_formats(
  98                     format_url, video_id, mpd_id='dash', fatal=False))
  99             elif type_ == 'hls' or ext == 'm3u8':
 100                 formats.extend(self._extract_m3u8_formats(
 101                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
 102                     m3u8_id='hls', fatal=False))
 103             elif extension == 'mp4' or ext == 'mp4':
 104                 formats.append({
 105                     'url': format_url,
 106                     'format_id': type_,
 107                     'width': int_or_none(resource.get('width')),
 108                     'height': int_or_none(resource.get('height')),
 109                 })
 110         self._sort_formats(formats)
 111
 112         attrs = {}
 113         for attribute in video['attributes']:
 114             if not isinstance(attribute, dict):
 115                 continue
 116             key = attribute.get('fieldKey')
 117             value = attribute.get('fieldValue')
 118             if not key or not value:
 119                 continue
 120             attrs[key] = value
 121
 122         title = attrs.get('title_stv') or video_id
 123         alt_title = attrs.get('title')
 124         description = attrs.get('long_description') or attrs.get('short_description')
 125         series = attrs.get('label')
 126         season = attrs.get('season')
 127         episode = attrs.get('chapter')
 128         duration = float_or_none(attrs.get('duration'), scale=1000)
 129         season_number = int_or_none(self._search_regex(
 130             r'Season (\d+)', season or '', 'season number', default=None))
 131         episode_number = int_or_none(self._search_regex(
 132             r'Episode (\d+)', episode or '', 'episode number', default=None))
 133
 134         return {
 135             'id': video_id,
 136             'title': title,
 137             'alt_title': alt_title,
 138             'description': description,
 139             'thumbnail': thumbnail,
 140             'duration': duration,
 141             'timestamp': unified_timestamp(video.get('lastPublished')),
 142             'series': series,
 143             'season': season,
 144             'season_number': season_number,
 145             'episode': episode,
 146             'episode_number': episode_number,
 147             'formats': formats,
 148         }