Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/sendtonews.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .jwplatform import JWPlatformBaseIE
   7 from ..compat import compat_parse_qs
   8 from ..utils import (
   9     ExtractorError,
  10     parse_duration,
  11 )
  12
  13
  14 class SendtoNewsIE(JWPlatformBaseIE):
  15     _VALID_URL = r'https?://embed\.sendtonews\.com/player/embed\.php\?(?P<query>[^#]+)'
  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/player/embed.php?SK=GxfCe0Zo7D&MK=175909&PK=5588&autoplay=on&sound=yes',
  20         'info_dict': {
  21             'id': 'GxfCe0Zo7D-175909-5588',
  22             'ext': 'mp4',
  23             'title': 'Recap: CLE 15, CIN 6',
  24             'description': '5/16/16: Indians\' bats explode for 15 runs in a win',
  25             'duration': 49,
  26         },
  27         'params': {
  28             # m3u8 download
  29             'skip_download': True,
  30         },
  31     }
  32
  33     _URL_TEMPLATE = '//embed.sendtonews.com/player/embed.php?SK=%s&MK=%s&PK=%s'
  34
  35     @classmethod
  36     def _extract_url(cls, webpage):
  37         mobj = re.search(r'''(?x)<script[^>]+src=([\'"])
  38             (?:https?:)?//embed\.sendtonews\.com/player/responsiveembed\.php\?
  39                 .*\bSC=(?P<SC>[0-9a-zA-Z-]+).*
  40             \1>''', webpage)
  41         if mobj:
  42             sk, mk, pk = mobj.group('SC').split('-')
  43             return cls._URL_TEMPLATE % (sk, mk, pk)
  44
  45     def _real_extract(self, url):
  46         mobj = re.match(self._VALID_URL, url)
  47         params = compat_parse_qs(mobj.group('query'))
  48
  49         if 'SK' not in params or 'MK' not in params or 'PK' not in params:
  50             raise ExtractorError('Invalid URL', expected=True)
  51
  52         video_id = '-'.join([params['SK'][0], params['MK'][0], params['PK'][0]])
  53
  54         webpage = self._download_webpage(url, video_id)
  55
  56         jwplayer_data_str = self._search_regex(
  57             r'jwplayer\("[^"]+"\)\.setup\((.+?)\);', webpage, 'JWPlayer data')
  58         js_vars = {
  59             'w': 1024,
  60             'h': 768,
  61             'modeVar': 'html5',
  62         }
  63         for name, val in js_vars.items():
  64             js_val = '%d' % val if isinstance(val, int) else '"%s"' % val
  65             jwplayer_data_str = jwplayer_data_str.replace(':%s,' % name, ':%s,' % js_val)
  66
  67         info_dict = self._parse_jwplayer_data(
  68             self._parse_json(jwplayer_data_str, video_id),
  69             video_id, require_title=False, rtmp_params={'no_resume': True})
  70
  71         title = self._html_search_regex(
  72             r'<div[^>]+class="embedTitle">([^<]+)</div>', webpage, 'title')
  73         description = self._html_search_regex(
  74             r'<div[^>]+class="embedSubTitle">([^<]+)</div>', webpage,
  75             'description', fatal=False)
  76         duration = parse_duration(self._html_search_regex(
  77             r'<div[^>]+class="embedDetails">([0-9:]+)', webpage,
  78             'duration', fatal=False))
  79
  80         info_dict.update({
  81             'title': title,
  82             'description': description,
  83             'duration': duration,
  84         })
  85
  86         return info_dict