_ Git - youtube-dl/blob - youtube_dl/extractor/cbsnews.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from .theplatform import ThePlatformIE
   6 from ..utils import parse_duration
   7
   8
   9 class CBSNewsIE(ThePlatformIE):
  10     IE_DESC = 'CBS News'
  11     _VALID_URL = r'http://(?:www\.)?cbsnews\.com/(?:news|videos)/(?P<id>[\da-z_-]+)'
  12
  13     _TESTS = [
  14         {
  15             'url': 'http://www.cbsnews.com/news/tesla-and-spacex-elon-musks-industrial-empire/',
  16             'info_dict': {
  17                 'id': 'tesla-and-spacex-elon-musks-industrial-empire',
  18                 'ext': 'flv',
  19                 'title': 'Tesla and SpaceX: Elon Musk\'s industrial empire',
  20                 'thumbnail': 'http://beta.img.cbsnews.com/i/2014/03/30/60147937-2f53-4565-ad64-1bdd6eb64679/60-0330-pelley-640x360.jpg',
  21                 'duration': 791,
  22             },
  23             'params': {
  24                 # rtmp download
  25                 'skip_download': True,
  26             },
  27         },
  28         {
  29             'url': 'http://www.cbsnews.com/videos/fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack/',
  30             'info_dict': {
  31                 'id': 'fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack',
  32                 'ext': 'mp4',
  33                 'title': 'Fort Hood shooting: Army downplays mental illness as cause of attack',
  34                 'thumbnail': 're:^https?://.*\.jpg$',
  35                 'duration': 205,
  36                 'subtitles': {
  37                     'en': [{
  38                         'ext': 'ttml',
  39                     }],
  40                 },
  41             },
  42             'params': {
  43                 # m3u8 download
  44                 'skip_download': True,
  45             },
  46         },
  47     ]
  48
  49     def _real_extract(self, url):
  50         video_id = self._match_id(url)
  51
  52         webpage = self._download_webpage(url, video_id)
  53
  54         video_info = self._parse_json(self._html_search_regex(
  55             r'(?:<ul class="media-list items" id="media-related-items"><li data-video-info|<div id="cbsNewsVideoPlayer" data-video-player-options)=\'({.+?})\'',
  56             webpage, 'video JSON info'), video_id)
  57
  58         item = video_info['item'] if 'item' in video_info else video_info
  59         title = item.get('articleTitle') or item.get('hed')
  60         duration = item.get('duration')
  61         thumbnail = item.get('mediaImage') or item.get('thumbnail')
  62
  63         subtitles = {}
  64         if 'mpxRefId' in video_info:
  65             subtitles['en'] = [{
  66                 'ext': 'ttml',
  67                 'url': 'http://www.cbsnews.com/videos/captions/%s.adb_xml' % video_info['mpxRefId'],
  68             }]
  69
  70         formats = []
  71         for format_id in ['RtmpMobileLow', 'RtmpMobileHigh', 'Hls', 'RtmpDesktop']:
  72             pid = item.get('media' + format_id)
  73             if not pid:
  74                 continue
  75             release_url = 'http://link.theplatform.com/s/dJ5BDC/%s?format=SMIL&mbr=true' % pid
  76             tp_formats, tp_subtitles = self._extract_theplatform_smil(release_url, video_id, 'Downloading %s SMIL data' % pid)
  77             formats.extend(tp_formats)
  78             subtitles = self._merge_subtitles(subtitles, tp_subtitles)
  79         self._sort_formats(formats)
  80
  81         return {
  82             'id': video_id,
  83             'title': title,
  84             'thumbnail': thumbnail,
  85             'duration': duration,
  86             'formats': formats,
  87             'subtitles': subtitles,
  88         }
  89
  90
  91 class CBSNewsLiveVideoIE(InfoExtractor):
  92     IE_DESC = 'CBS News Live Videos'
  93     _VALID_URL = r'http://(?:www\.)?cbsnews\.com/live/video/(?P<id>[\da-z_-]+)'
  94
  95     _TEST = {
  96         'url': 'http://www.cbsnews.com/live/video/clinton-sanders-prepare-to-face-off-in-nh/',
  97         'info_dict': {
  98             'id': 'clinton-sanders-prepare-to-face-off-in-nh',
  99             'ext': 'flv',
 100             'title': 'Clinton, Sanders Prepare To Face Off In NH',
 101             'duration': 334,
 102         },
 103     }
 104
 105     def _real_extract(self, url):
 106         video_id = self._match_id(url)
 107
 108         webpage = self._download_webpage(url, video_id)
 109
 110         video_info = self._parse_json(self._html_search_regex(
 111             r'data-story-obj=\'({.+?})\'', webpage, 'video JSON info'), video_id)['story']
 112
 113         hdcore_sign = 'hdcore=3.3.1'
 114         f4m_formats = self._extract_f4m_formats(video_info['url'] + '&' + hdcore_sign, video_id)
 115         if f4m_formats:
 116             for entry in f4m_formats:
 117                 # URLs without the extra param induce an 404 error
 118                 entry.update({'extra_param_to_segment_url': hdcore_sign})
 119
 120         return {
 121             'id': video_id,
 122             'title': video_info['headline'],
 123             'thumbnail': video_info.get('thumbnail_url_hd') or video_info.get('thumbnail_url_sd'),
 124             'duration': parse_duration(video_info.get('segmentDur')),
 125             'formats': f4m_formats,
 126         }