yt_dlp/extractor/sendtonews.py

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