yt_dlp/extractor/audimedia.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     int_or_none,
   4     parse_iso8601,
   5 )
   6
   7
   8 class AudiMediaIE(InfoExtractor):
   9     _VALID_URL = r'https?://(?:www\.)?audi-mediacenter\.com/(?:en|de)/audimediatv/(?:video/)?(?P<id>[^/?#]+)'
  10     _TESTS = [{
  11         'url': 'https://www.audi-mediacenter.com/en/audimediatv/60-seconds-of-audi-sport-104-2015-wec-bahrain-rookie-test-1467',
  12         'md5': '79a8b71c46d49042609795ab59779b66',
  13         'info_dict': {
  14             'id': '1565',
  15             'ext': 'mp4',
  16             'title': '60 Seconds of Audi Sport 104/2015 - WEC Bahrain, Rookie Test',
  17             'description': 'md5:60e5d30a78ced725f7b8d34370762941',
  18             'upload_date': '20151124',
  19             'timestamp': 1448354940,
  20             'duration': 74022,
  21             'view_count': int,
  22         },
  23     }, {
  24         'url': 'https://www.audi-mediacenter.com/en/audimediatv/video/60-seconds-of-audi-sport-104-2015-wec-bahrain-rookie-test-2991',
  25         'only_matching': True,
  26     }]
  27
  28     def _real_extract(self, url):
  29         display_id = self._match_id(url)
  30         webpage = self._download_webpage(url, display_id)
  31
  32         raw_payload = self._search_regex([
  33             r'class="amtv-embed"[^>]+id="([0-9a-z-]+)"',
  34             r'id="([0-9a-z-]+)"[^>]+class="amtv-embed"',
  35             r'class=\\"amtv-embed\\"[^>]+id=\\"([0-9a-z-]+)\\"',
  36             r'id=\\"([0-9a-z-]+)\\"[^>]+class=\\"amtv-embed\\"',
  37             r'id=(?:\\)?"(amtve-[a-z]-\d+-[a-z]{2})',
  38         ], webpage, 'raw payload')
  39         _, stage_mode, video_id, _ = raw_payload.split('-')
  40
  41         # TODO: handle s and e stage_mode (live streams and ended live streams)
  42         if stage_mode not in ('s', 'e'):
  43             video_data = self._download_json(
  44                 'https://www.audimedia.tv/api/video/v1/videos/' + video_id,
  45                 video_id, query={
  46                     'embed[]': ['video_versions', 'thumbnail_image'],
  47                 })['results']
  48             formats = []
  49
  50             stream_url_hls = video_data.get('stream_url_hls')
  51             if stream_url_hls:
  52                 formats.extend(self._extract_m3u8_formats(
  53                     stream_url_hls, video_id, 'mp4',
  54                     entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
  55
  56             stream_url_hds = video_data.get('stream_url_hds')
  57             if stream_url_hds:
  58                 formats.extend(self._extract_f4m_formats(
  59                     stream_url_hds + '?hdcore=3.4.0',
  60                     video_id, f4m_id='hds', fatal=False))
  61
  62             for video_version in video_data.get('video_versions', []):
  63                 video_version_url = video_version.get('download_url') or video_version.get('stream_url')
  64                 if not video_version_url:
  65                     continue
  66                 f = {
  67                     'url': video_version_url,
  68                     'width': int_or_none(video_version.get('width')),
  69                     'height': int_or_none(video_version.get('height')),
  70                     'abr': int_or_none(video_version.get('audio_bitrate')),
  71                     'vbr': int_or_none(video_version.get('video_bitrate')),
  72                 }
  73                 bitrate = self._search_regex(r'(\d+)k', video_version_url, 'bitrate', default=None)
  74                 if bitrate:
  75                     f.update({
  76                         'format_id': f'http-{bitrate}',
  77                     })
  78                 formats.append(f)
  79
  80             return {
  81                 'id': video_id,
  82                 'title': video_data['title'],
  83                 'description': video_data.get('subtitle'),
  84                 'thumbnail': video_data.get('thumbnail_image', {}).get('file'),
  85                 'timestamp': parse_iso8601(video_data.get('publication_date')),
  86                 'duration': int_or_none(video_data.get('duration')),
  87                 'view_count': int_or_none(video_data.get('view_count')),
  88                 'formats': formats,
  89             }