yt_dlp/extractor/webcaster.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     determine_ext,
   6     join_nonempty,
   7     xpath_text,
   8 )
   9
  10
  11 class WebcasterIE(InfoExtractor):
  12     _VALID_URL = r'https?://bl\.webcaster\.pro/(?:quote|media)/start/free_(?P<id>[^/]+)'
  13     _TESTS = [{
  14         # http://video.khl.ru/quotes/393859
  15         'url': 'http://bl.webcaster.pro/quote/start/free_c8cefd240aa593681c8d068cff59f407_hd/q393859/eb173f99dd5f558674dae55f4ba6806d/1480289104?sr%3D105%26fa%3D1%26type_id%3D18',
  16         'md5': '0c162f67443f30916ff1c89425dcd4cd',
  17         'info_dict': {
  18             'id': 'c8cefd240aa593681c8d068cff59f407_hd',
  19             'ext': 'mp4',
  20             'title': 'Сибирь - Нефтехимик. Лучшие моменты первого периода',
  21             'thumbnail': r're:^https?://.*\.jpg$',
  22         },
  23     }, {
  24         'url': 'http://bl.webcaster.pro/media/start/free_6246c7a4453ac4c42b4398f840d13100_hd/2_2991109016/e8d0d82587ef435480118f9f9c41db41/4635726126',
  25         'only_matching': True,
  26     }]
  27
  28     def _real_extract(self, url):
  29         video_id = self._match_id(url)
  30
  31         video = self._download_xml(url, video_id)
  32
  33         title = xpath_text(video, './/event_name', 'event name', fatal=True)
  34
  35         formats = []
  36         for format_id in (None, 'noise'):
  37             track_tag = join_nonempty('track', format_id, delim='_')
  38             for track in video.findall(f'.//iphone/{track_tag}'):
  39                 track_url = track.text
  40                 if not track_url:
  41                     continue
  42                 if determine_ext(track_url) == 'm3u8':
  43                     m3u8_formats = self._extract_m3u8_formats(
  44                         track_url, video_id, 'mp4',
  45                         entry_protocol='m3u8_native',
  46                         m3u8_id=join_nonempty('hls', format_id, delim='-'), fatal=False)
  47                     for f in m3u8_formats:
  48                         f.update({
  49                             'source_preference': 0 if format_id == 'noise' else 1,
  50                             'format_note': track.get('title'),
  51                         })
  52                     formats.extend(m3u8_formats)
  53
  54         thumbnail = xpath_text(video, './/image', 'thumbnail')
  55
  56         return {
  57             'id': video_id,
  58             'title': title,
  59             'thumbnail': thumbnail,
  60             'formats': formats,
  61         }
  62
  63
  64 class WebcasterFeedIE(InfoExtractor):
  65     _VALID_URL = r'https?://bl\.webcaster\.pro/feed/start/free_(?P<id>[^/]+)'
  66     _EMBED_REGEX = [r'<(?:object|a[^>]+class=["\']webcaster-player["\'])[^>]+data(?:-config)?=(["\']).*?config=(?P<url>https?://bl\.webcaster\.pro/feed/start/free_.*?)(?:[?&]|\1)']
  67     _TEST = {
  68         'url': 'http://bl.webcaster.pro/feed/start/free_c8cefd240aa593681c8d068cff59f407_hd/q393859/eb173f99dd5f558674dae55f4ba6806d/1480289104',
  69         'only_matching': True,
  70     }
  71
  72     def _extract_from_webpage(self, url, webpage):
  73         yield from super()._extract_from_webpage(url, webpage)
  74
  75         for secure in (True, False):
  76             video_url = self._og_search_video_url(webpage, secure=secure, default=None)
  77             if video_url:
  78                 mobj = re.search(
  79                     r'config=(?P<url>https?://bl\.webcaster\.pro/feed/start/free_[^?&=]+)',
  80                     video_url)
  81                 if mobj:
  82                     yield self.url_result(mobj.group('url'), self)
  83
  84     def _real_extract(self, url):
  85         video_id = self._match_id(url)
  86
  87         feed = self._download_xml(url, video_id)
  88
  89         video_url = xpath_text(
  90             feed, ('video_hd', 'video'), 'video url', fatal=True)
  91
  92         return self.url_result(video_url, WebcasterIE.ie_key())