yt_dlp/extractor/sendtonews.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     float_or_none,
   6     parse_iso8601,
   7     update_url_query,
   8     int_or_none,
   9     determine_protocol,
  10     unescapeHTML,
  11 )
  12
  13
  14 class SendtoNewsIE(InfoExtractor):
  15     _VALID_URL = r'https?://embed\.sendtonews\.com/player2/embedplayer\.php\?.*\bSC=(?P<id>[0-9A-Za-z-]+)'
  16
  17     _TEST = {
  18         # From http://cleveland.cbslocal.com/2016/05/16/indians-score-season-high-15-runs-in-blowout-win-over-reds-rapid-reaction/
  19         'url': 'http://embed.sendtonews.com/player2/embedplayer.php?SC=GxfCe0Zo7D-175909-5588&type=single&autoplay=on&sound=YES',
  20         'info_dict': {
  21             'id': 'GxfCe0Zo7D-175909-5588'
  22         },
  23         'playlist_count': 8,
  24         # test the first video only to prevent lengthy tests
  25         'playlist': [{
  26             'info_dict': {
  27                 'id': '240385',
  28                 'ext': 'mp4',
  29                 'title': 'Indians introduce Encarnacion',
  30                 'description': 'Indians president of baseball operations Chris Antonetti and Edwin Encarnacion discuss the slugger\'s three-year contract with Cleveland',
  31                 'duration': 137.898,
  32                 'thumbnail': r're:https?://.*\.jpg$',
  33                 'upload_date': '20170105',
  34                 'timestamp': 1483649762,
  35             },
  36         }],
  37         'params': {
  38             # m3u8 download
  39             'skip_download': True,
  40         },
  41     }
  42
  43     _URL_TEMPLATE = '//embed.sendtonews.com/player2/embedplayer.php?SC=%s'
  44
  45     @classmethod
  46     def _extract_embed_urls(cls, url, webpage):
  47         mobj = re.search(r'''(?x)<script[^>]+src=([\'"])
  48             (?:https?:)?//embed\.sendtonews\.com/player/responsiveembed\.php\?
  49                 .*\bSC=(?P<SC>[0-9a-zA-Z-]+).*
  50             \1>''', webpage)
  51         if mobj:
  52             sc = mobj.group('SC')
  53             yield cls._URL_TEMPLATE % sc
  54
  55     def _real_extract(self, url):
  56         playlist_id = self._match_id(url)
  57
  58         data_url = update_url_query(
  59             url.replace('embedplayer.php', 'data_read.php'),
  60             {'cmd': 'loadInitial'})
  61         playlist_data = self._download_json(data_url, playlist_id)
  62
  63         entries = []
  64         for video in playlist_data['playlistData'][0]:
  65             info_dict = self._parse_jwplayer_data(
  66                 video['jwconfiguration'],
  67                 require_title=False, m3u8_id='hls', rtmp_params={'no_resume': True})
  68
  69             for f in info_dict['formats']:
  70                 if f.get('tbr'):
  71                     continue
  72                 tbr = int_or_none(self._search_regex(
  73                     r'/(\d+)k/', f['url'], 'bitrate', default=None))
  74                 if not tbr:
  75                     continue
  76                 f.update({
  77                     'format_id': '%s-%d' % (determine_protocol(f), tbr),
  78                     'tbr': tbr,
  79                 })
  80             # 'tbr' was explicitly set to be preferred over 'height' originally,
  81             # So this is being kept unless someone can confirm this is unnecessary
  82             self._sort_formats(info_dict['formats'], ('tbr', 'res'))
  83
  84             thumbnails = []
  85             if video.get('thumbnailUrl'):
  86                 thumbnails.append({
  87                     'id': 'normal',
  88                     'url': video['thumbnailUrl'],
  89                 })
  90             if video.get('smThumbnailUrl'):
  91                 thumbnails.append({
  92                     'id': 'small',
  93                     'url': video['smThumbnailUrl'],
  94                 })
  95             info_dict.update({
  96                 'title': video['S_headLine'].strip(),
  97                 'description': unescapeHTML(video.get('S_fullStory')),
  98                 'thumbnails': thumbnails,
  99                 'duration': float_or_none(video.get('SM_length')),
 100                 'timestamp': parse_iso8601(video.get('S_sysDate'), delimiter=' '),
 101             })
 102             entries.append(info_dict)
 103
 104         return self.playlist_result(entries, playlist_id)