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

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from .cbs import CBSBaseIE
   6 from ..utils import (
   7     parse_duration,
   8 )
   9
  10
  11 class CBSNewsIE(CBSBaseIE):
  12     IE_DESC = 'CBS News'
  13     _VALID_URL = r'https?://(?:www\.)?cbsnews\.com/(?:news|videos)/(?P<id>[\da-z_-]+)'
  14
  15     _TESTS = [
  16         {
  17             'url': 'http://www.cbsnews.com/news/tesla-and-spacex-elon-musks-industrial-empire/',
  18             'info_dict': {
  19                 'id': 'tesla-and-spacex-elon-musks-industrial-empire',
  20                 'ext': 'flv',
  21                 'title': 'Tesla and SpaceX: Elon Musk\'s industrial empire',
  22                 'thumbnail': 'http://beta.img.cbsnews.com/i/2014/03/30/60147937-2f53-4565-ad64-1bdd6eb64679/60-0330-pelley-640x360.jpg',
  23                 'duration': 791,
  24             },
  25             'params': {
  26                 # rtmp download
  27                 'skip_download': True,
  28             },
  29         },
  30         {
  31             'url': 'http://www.cbsnews.com/videos/fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack/',
  32             'info_dict': {
  33                 'id': 'SNJBOYzXiWBOvaLsdzwH8fmtP1SCd91Y',
  34                 'ext': 'mp4',
  35                 'title': 'Fort Hood shooting: Army downplays mental illness as cause of attack',
  36                 'description': 'md5:4a6983e480542d8b333a947bfc64ddc7',
  37                 'upload_date': '19700101',
  38                 'uploader': 'CBSI-NEW',
  39                 'thumbnail': 're:^https?://.*\.jpg$',
  40                 'duration': 205,
  41                 'subtitles': {
  42                     'en': [{
  43                         'ext': 'ttml',
  44                     }],
  45                 },
  46             },
  47             'params': {
  48                 # m3u8 download
  49                 'skip_download': True,
  50             },
  51         },
  52     ]
  53
  54     def _real_extract(self, url):
  55         video_id = self._match_id(url)
  56
  57         webpage = self._download_webpage(url, video_id)
  58
  59         video_info = self._parse_json(self._html_search_regex(
  60             r'(?:<ul class="media-list items" id="media-related-items"><li data-video-info|<div id="cbsNewsVideoPlayer" data-video-player-options)=\'({.+?})\'',
  61             webpage, 'video JSON info'), video_id)
  62
  63         item = video_info['item'] if 'item' in video_info else video_info
  64         guid = item['mpxRefId']
  65         return self._extract_video_info('byGuid=%s' % guid, guid)
  66
  67
  68 class CBSNewsLiveVideoIE(InfoExtractor):
  69     IE_DESC = 'CBS News Live Videos'
  70     _VALID_URL = r'https?://(?:www\.)?cbsnews\.com/live/video/(?P<id>[\da-z_-]+)'
  71
  72     _TEST = {
  73         'url': 'http://www.cbsnews.com/live/video/clinton-sanders-prepare-to-face-off-in-nh/',
  74         'info_dict': {
  75             'id': 'clinton-sanders-prepare-to-face-off-in-nh',
  76             'ext': 'flv',
  77             'title': 'Clinton, Sanders Prepare To Face Off In NH',
  78             'duration': 334,
  79         },
  80     }
  81
  82     def _real_extract(self, url):
  83         video_id = self._match_id(url)
  84
  85         webpage = self._download_webpage(url, video_id)
  86
  87         video_info = self._parse_json(self._html_search_regex(
  88             r'data-story-obj=\'({.+?})\'', webpage, 'video JSON info'), video_id)['story']
  89
  90         hdcore_sign = 'hdcore=3.3.1'
  91         f4m_formats = self._extract_f4m_formats(video_info['url'] + '&' + hdcore_sign, video_id)
  92         if f4m_formats:
  93             for entry in f4m_formats:
  94                 # URLs without the extra param induce an 404 error
  95                 entry.update({'extra_param_to_segment_url': hdcore_sign})
  96         self._sort_formats(f4m_formats)
  97
  98         return {
  99             'id': video_id,
 100             'title': video_info['headline'],
 101             'thumbnail': video_info.get('thumbnail_url_hd') or video_info.get('thumbnail_url_sd'),
 102             'duration': parse_duration(video_info.get('segmentDur')),
 103             'formats': f4m_formats,
 104         }