youtube_dl/extractor/sportdeutschland.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     parse_iso8601,
   9     sanitized_Request,
  10 )
  11
  12
  13 class SportDeutschlandIE(InfoExtractor):
  14     _VALID_URL = r'https?://sportdeutschland\.tv/(?P<sport>[^/?#]+)/(?P<id>[^?#/]+)(?:$|[?#])'
  15     _TESTS = [{
  16         'url': 'https://sportdeutschland.tv/badminton/re-live-deutsche-meisterschaften-2020-halbfinals?playlistId=0',
  17         'info_dict': {
  18             'id': 're-live-deutsche-meisterschaften-2020-halbfinals',
  19             'ext': 'mp4',
  20             'title': 're:Re-live: Deutsche Meisterschaften 2020.*Halbfinals',
  21             'categories': ['Badminton-Deutschland'],
  22             'view_count': int,
  23             'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
  24             'timestamp': int,
  25             'upload_date': '20200201',
  26             'description': 're:.*',  # meaningless description for THIS video
  27         },
  28     }]
  29
  30     def _real_extract(self, url):
  31         mobj = re.match(self._VALID_URL, url)
  32         video_id = mobj.group('id')
  33         sport_id = mobj.group('sport')
  34
  35         api_url = 'https://proxy.vidibusdynamic.net/ssl/backend.sportdeutschland.tv/api/permalinks/%s/%s?access_token=true' % (
  36             sport_id, video_id)
  37         req = sanitized_Request(api_url, headers={
  38             'Accept': 'application/vnd.vidibus.v2.html+json',
  39             'Referer': url,
  40         })
  41         data = self._download_json(req, video_id)
  42
  43         asset = data['asset']
  44         categories = [data['section']['title']]
  45
  46         formats = []
  47         smil_url = asset['video']
  48         if '.smil' in smil_url:
  49             m3u8_url = smil_url.replace('.smil', '.m3u8')
  50             formats.extend(
  51                 self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4'))
  52
  53             smil_doc = self._download_xml(
  54                 smil_url, video_id, note='Downloading SMIL metadata')
  55             base_url_el = smil_doc.find('./head/meta')
  56             if base_url_el:
  57                 base_url = base_url_el.attrib['base']
  58             formats.extend([{
  59                 'format_id': 'rmtp',
  60                 'url': base_url if base_url_el else n.attrib['src'],
  61                 'play_path': n.attrib['src'],
  62                 'ext': 'flv',
  63                 'preference': -100,
  64                 'format_note': 'Seems to fail at example stream',
  65             } for n in smil_doc.findall('./body/video')])
  66         else:
  67             formats.append({'url': smil_url})
  68
  69         self._sort_formats(formats)
  70
  71         return {
  72             'id': video_id,
  73             'formats': formats,
  74             'title': asset['title'],
  75             'thumbnail': asset.get('image'),
  76             'description': asset.get('teaser'),
  77             'duration': asset.get('duration'),
  78             'categories': categories,
  79             'view_count': asset.get('views'),
  80             'rtmp_live': asset.get('live'),
  81             'timestamp': parse_iso8601(asset.get('date')),
  82         }