youtube_dl/extractor/sendtonews.py

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