youtube_dl/extractor/audimedia.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     int_or_none,
   7     parse_iso8601,
   8     sanitized_Request,
   9 )
  10
  11
  12 class AudiMediaIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?audi-mediacenter\.com/(?:en|de)/audimediatv/(?P<id>[^/?#]+)'
  14     _TEST = {
  15         'url': 'https://www.audi-mediacenter.com/en/audimediatv/60-seconds-of-audi-sport-104-2015-wec-bahrain-rookie-test-1467',
  16         'md5': '79a8b71c46d49042609795ab59779b66',
  17         'info_dict': {
  18             'id': '1565',
  19             'ext': 'mp4',
  20             'title': '60 Seconds of Audi Sport 104/2015 - WEC Bahrain, Rookie Test',
  21             'description': 'md5:60e5d30a78ced725f7b8d34370762941',
  22             'upload_date': '20151124',
  23             'timestamp': 1448354940,
  24             'duration': 74022,
  25             'view_count': int,
  26         }
  27     }
  28     # extracted from https://audimedia.tv/assets/embed/embedded-player.js (dataSourceAuthToken)
  29     _AUTH_TOKEN = 'e25b42847dba18c6c8816d5d8ce94c326e06823ebf0859ed164b3ba169be97f2'
  30
  31     def _real_extract(self, url):
  32         display_id = self._match_id(url)
  33         webpage = self._download_webpage(url, display_id)
  34
  35         raw_payload = self._search_regex([
  36             r'class="amtv-embed"[^>]+id="([^"]+)"',
  37             r'class=\\"amtv-embed\\"[^>]+id=\\"([^"]+)\\"',
  38         ], webpage, 'raw payload')
  39         _, stage_mode, video_id, lang = 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             request = sanitized_Request(
  44                 'https://audimedia.tv/api/video/v1/videos/%s?embed[]=video_versions&embed[]=thumbnail_image&where[content_language_iso]=%s' % (video_id, lang),
  45                 headers={'X-Auth-Token': self._AUTH_TOKEN})
  46             json_data = self._download_json(request, video_id)['results']
  47             formats = []
  48
  49             stream_url_hls = json_data.get('stream_url_hls')
  50             if stream_url_hls:
  51                 formats.extend(self._extract_m3u8_formats(
  52                     stream_url_hls, video_id, 'mp4',
  53                     entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
  54
  55             stream_url_hds = json_data.get('stream_url_hds')
  56             if stream_url_hds:
  57                 formats.extend(self._extract_f4m_formats(
  58                     stream_url_hds + '?hdcore=3.4.0',
  59                     video_id, f4m_id='hds', fatal=False))
  60
  61             for video_version in json_data.get('video_versions'):
  62                 video_version_url = video_version.get('download_url') or video_version.get('stream_url')
  63                 if not video_version_url:
  64                     continue
  65                 f = {
  66                     'url': video_version_url,
  67                     'width': int_or_none(video_version.get('width')),
  68                     'height': int_or_none(video_version.get('height')),
  69                     'abr': int_or_none(video_version.get('audio_bitrate')),
  70                     'vbr': int_or_none(video_version.get('video_bitrate')),
  71                 }
  72                 bitrate = self._search_regex(r'(\d+)k', video_version_url, 'bitrate', default=None)
  73                 if bitrate:
  74                     f.update({
  75                         'format_id': 'http-%s' % bitrate,
  76                     })
  77                 formats.append(f)
  78             self._sort_formats(formats)
  79
  80             return {
  81                 'id': video_id,
  82                 'title': json_data['title'],
  83                 'description': json_data.get('subtitle'),
  84                 'thumbnail': json_data.get('thumbnail_image', {}).get('file'),
  85                 'timestamp': parse_iso8601(json_data.get('publication_date')),
  86                 'duration': int_or_none(json_data.get('duration')),
  87                 'view_count': int_or_none(json_data.get('view_count')),
  88                 'formats': formats,
  89             }