_ Git - youtube-dl/blob - youtube_dl/extractor/azmedien.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from .kaltura import KalturaIE
   9
  10
  11 class AZMedienBaseIE(InfoExtractor):
  12     _PARTNER_ID = '1719221'
  13
  14     def _kaltura_video(self, partner_id, entry_id):
  15         return self.url_result(
  16             'kaltura:%s:%s' % (partner_id, entry_id), ie=KalturaIE.ie_key(),
  17             video_id=entry_id)
  18
  19
  20 class AZMedienIE(AZMedienBaseIE):
  21     IE_DESC = 'AZ Medien videos'
  22     _VALID_URL = r'''(?x)
  23                     https?://
  24                         (?:www\.)?
  25                         (?P<host>
  26                             telezueri\.ch|
  27                             telebaern\.tv|
  28                             telem1\.ch
  29                         )/
  30                         [^/]+/
  31                         (?P<id>
  32                             [^/]+-(?P<article_id>\d+)
  33                         )
  34                         (?:
  35                             \#video=
  36                             (?P<kaltura_id>
  37                                 [_0-9a-z]+
  38                             )
  39                         )?
  40                     '''
  41
  42     _TESTS = [{
  43         'url': 'https://www.telezueri.ch/sonntalk/bundesrats-vakanzen-eu-rahmenabkommen-133214569',
  44         'info_dict': {
  45             'id': '1_anruz3wy',
  46             'ext': 'mp4',
  47             'title': 'Bundesrats-Vakanzen / EU-Rahmenabkommen',
  48             'description': 'md5:dd9f96751ec9c35e409a698a328402f3',
  49             'uploader_id': 'TVOnline',
  50             'upload_date': '20180930',
  51             'timestamp': 1538328802,
  52         },
  53         'params': {
  54             'skip_download': True,
  55         },
  56     }, {
  57         'url': 'https://www.telebaern.tv/telebaern-news/montag-1-oktober-2018-ganze-sendung-133531189#video=0_7xjo9lf1',
  58         'only_matching': True
  59     }]
  60
  61     def _real_extract(self, url):
  62         video_id = self._match_id(url)
  63         mobj = re.match(self._VALID_URL, url)
  64         entry_id = mobj.group('kaltura_id')
  65
  66         if not entry_id:
  67             webpage = self._download_webpage(url, video_id)
  68             api_path = self._search_regex(
  69                 r'["\']apiPath["\']\s*:\s*["\']([^"^\']+)["\']',
  70                 webpage, 'api path')
  71             api_url = 'https://www.%s%s' % (mobj.group('host'), api_path)
  72             payload = {
  73                 'query': '''query VideoContext($articleId: ID!) {
  74                     article: node(id: $articleId) {
  75                       ... on Article {
  76                         mainAssetRelation {
  77                           asset {
  78                             ... on VideoAsset {
  79                               kalturaId
  80                             }
  81                           }
  82                         }
  83                       }
  84                     }
  85                   }''',
  86                 'variables': {'articleId': 'Article:%s' % mobj.group('article_id')},
  87             }
  88             json_data = self._download_json(
  89                 api_url, video_id, headers={
  90                     'Content-Type': 'application/json',
  91                 },
  92                 data=json.dumps(payload).encode())
  93             entry_id = json_data['data']['article']['mainAssetRelation']['asset']['kalturaId']
  94
  95         return self._kaltura_video(self._PARTNER_ID, entry_id)