_ 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 _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
  50         closed_caption_e = smil.find(self._xpath_ns('.//param[@name=\'ClosedCaptionURL\']', namespace))
  51         return {
  52             'en': [{
  53                 'ext': 'ttml',
  54                 'url': closed_caption_e.attrib['value'],
  55             }]
  56         } if closed_caption_e is not None and closed_caption_e.attrib.get('value') else []
  57
  58     def _real_extract(self, url):
  59         video_id = self._match_id(url)
  60
  61         webpage = self._download_webpage(url, video_id)
  62
  63         video_info = self._parse_json(self._html_search_regex(
  64             r'(?:<ul class="media-list items" id="media-related-items"><li data-video-info|<div id="cbsNewsVideoPlayer" data-video-player-options)=\'({.+?})\'',
  65             webpage, 'video JSON info'), video_id)
  66
  67         item = video_info['item'] if 'item' in video_info else video_info
  68         title = item.get('articleTitle') or item.get('hed')
  69         duration = item.get('duration')
  70         thumbnail = item.get('mediaImage') or item.get('thumbnail')
  71
  72         subtitles = {}
  73         formats = []
  74         for format_id in ['RtmpMobileLow', 'RtmpMobileHigh', 'Hls', 'RtmpDesktop']:
  75             pid = item.get('media' + format_id)
  76             if not pid:
  77                 continue
  78             release_url = 'http://link.theplatform.com/s/dJ5BDC/%s?format=SMIL&mbr=true' % pid
  79             tp_formats, tp_subtitles = self._extract_theplatform_smil(release_url, video_id, 'Downloading %s SMIL data' % pid)
  80             formats.extend(tp_formats)
  81             subtitles = self._merge_subtitles(subtitles, tp_subtitles)
  82         self._sort_formats(formats)
  83
  84         return {
  85             'id': video_id,
  86             'title': title,
  87             'thumbnail': thumbnail,
  88             'duration': duration,
  89             'formats': formats,
  90             'subtitles': subtitles,
  91         }
  92
  93
  94 class CBSNewsLiveVideoIE(InfoExtractor):
  95     IE_DESC = 'CBS News Live Videos'
  96     _VALID_URL = r'http://(?:www\.)?cbsnews\.com/live/video/(?P<id>[\da-z_-]+)'
  97
  98     _TEST = {
  99         'url': 'http://www.cbsnews.com/live/video/clinton-sanders-prepare-to-face-off-in-nh/',
 100         'info_dict': {
 101             'id': 'clinton-sanders-prepare-to-face-off-in-nh',
 102             'ext': 'flv',
 103             'title': 'Clinton, Sanders Prepare To Face Off In NH',
 104             'duration': 334,
 105         },
 106     }
 107
 108     def _real_extract(self, url):
 109         video_id = self._match_id(url)
 110
 111         webpage = self._download_webpage(url, video_id)
 112
 113         video_info = self._parse_json(self._html_search_regex(
 114             r'data-story-obj=\'({.+?})\'', webpage, 'video JSON info'), video_id)['story']
 115
 116         hdcore_sign = 'hdcore=3.3.1'
 117         f4m_formats = self._extract_f4m_formats(video_info['url'] + '&' + hdcore_sign, video_id)
 118         if f4m_formats:
 119             for entry in f4m_formats:
 120                 # URLs without the extra param induce an 404 error
 121                 entry.update({'extra_param_to_segment_url': hdcore_sign})
 122
 123         return {
 124             'id': video_id,
 125             'title': video_info['headline'],
 126             'thumbnail': video_info.get('thumbnail_url_hd') or video_info.get('thumbnail_url_sd'),
 127             'duration': parse_duration(video_info.get('segmentDur')),
 128             'formats': f4m_formats,
 129         }