yt_dlp/extractor/performgroup.py

   1 from .common import InfoExtractor
   2 from ..utils import int_or_none
   3
   4
   5 class PerformGroupIE(InfoExtractor):
   6     _VALID_URL = r'https?://player\.performgroup\.com/eplayer(?:/eplayer\.html|\.js)#/?(?P<id>[0-9a-f]{26})\.(?P<auth_token>[0-9a-z]{26})'
   7     _TESTS = [{
   8         # http://www.faz.net/aktuell/sport/fussball/wm-2018-playoffs-schweiz-besiegt-nordirland-1-0-15286104.html
   9         'url': 'http://player.performgroup.com/eplayer/eplayer.html#d478c41c5d192f56b9aa859de8.1w4crrej5w14e1ed4s1ce4ykab',
  10         'md5': '259cb03d142e2e52471e8837ecacb29f',
  11         'info_dict': {
  12             'id': 'xgrwobuzumes1lwjxtcdpwgxd',
  13             'ext': 'mp4',
  14             'title': 'Liga MX: Keine Einsicht nach Horrorfoul',
  15             'description': 'md5:7cd3b459c82725b021e046ab10bf1c5b',
  16             'timestamp': 1511533477,
  17             'upload_date': '20171124',
  18         }
  19     }]
  20
  21     def _call_api(self, service, auth_token, content_id, referer_url):
  22         return self._download_json(
  23             'http://ep3.performfeeds.com/ep%s/%s/%s/' % (service, auth_token, content_id),
  24             content_id, headers={
  25                 'Referer': referer_url,
  26                 'Origin': 'http://player.performgroup.com',
  27             }, query={
  28                 '_fmt': 'json',
  29             })
  30
  31     def _real_extract(self, url):
  32         player_id, auth_token = self._match_valid_url(url).groups()
  33         bootstrap = self._call_api('bootstrap', auth_token, player_id, url)
  34         video = bootstrap['config']['dataSource']['sourceItems'][0]['videos'][0]
  35         video_id = video['uuid']
  36         vod = self._call_api('vod', auth_token, video_id, url)
  37         media = vod['videos']['video'][0]['media']
  38
  39         formats = []
  40         hls_url = media.get('hls', {}).get('url')
  41         if hls_url:
  42             formats.extend(self._extract_m3u8_formats(hls_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
  43
  44         hds_url = media.get('hds', {}).get('url')
  45         if hds_url:
  46             formats.extend(self._extract_f4m_formats(hds_url + '?hdcore', video_id, f4m_id='hds', fatal=False))
  47
  48         for c in media.get('content', []):
  49             c_url = c.get('url')
  50             if not c_url:
  51                 continue
  52             tbr = int_or_none(c.get('bitrate'), 1000)
  53             format_id = 'http'
  54             if tbr:
  55                 format_id += '-%d' % tbr
  56             formats.append({
  57                 'format_id': format_id,
  58                 'url': c_url,
  59                 'tbr': tbr,
  60                 'width': int_or_none(c.get('width')),
  61                 'height': int_or_none(c.get('height')),
  62                 'filesize': int_or_none(c.get('fileSize')),
  63                 'vcodec': c.get('type'),
  64                 'fps': int_or_none(c.get('videoFrameRate')),
  65                 'vbr': int_or_none(c.get('videoRate'), 1000),
  66                 'abr': int_or_none(c.get('audioRate'), 1000),
  67             })
  68
  69         return {
  70             'id': video_id,
  71             'title': video['title'],
  72             'description': video.get('description'),
  73             'thumbnail': video.get('poster'),
  74             'duration': int_or_none(video.get('duration')),
  75             'timestamp': int_or_none(video.get('publishedTime'), 1000),
  76             'formats': formats,
  77         }