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

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import json
   6
   7 from .theplatform import ThePlatformIE
   8 from ..utils import remove_start
   9
  10
  11 class CBSNewsIE(ThePlatformIE):
  12     IE_DESC = 'CBS News'
  13     _VALID_URL = r'http://(?:www\.)?cbsnews\.com/(?:[^/]+/)+(?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': 'fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack',
  34                 'ext': 'mp4',
  35                 'title': 'Fort Hood shooting: Army downplays mental illness as cause of attack',
  36                 'thumbnail': 're:^https?://.*\.jpg$',
  37                 'duration': 205,
  38                 'subtitles': {
  39                     'en': [{
  40                         'ext': 'ttml',
  41                     }],
  42                 },
  43             },
  44             'params': {
  45                 # m3u8 download
  46                 'skip_download': True,
  47             },
  48         },
  49     ]
  50
  51     def _real_extract(self, url):
  52         mobj = re.match(self._VALID_URL, url)
  53         video_id = mobj.group('id')
  54
  55         webpage = self._download_webpage(url, video_id)
  56
  57         video_info = json.loads(self._html_search_regex(
  58             r'(?:<ul class="media-list items" id="media-related-items"><li data-video-info|<div id="cbsNewsVideoPlayer" data-video-player-options)=\'({.+?})\'',
  59             webpage, 'video JSON info'))
  60
  61         item = video_info['item'] if 'item' in video_info else video_info
  62         title = item.get('articleTitle') or item.get('hed')
  63         duration = item.get('duration')
  64         thumbnail = item.get('mediaImage') or item.get('thumbnail')
  65
  66         subtitles = {}
  67         if 'mpxRefId' in video_info:
  68             subtitles['en'] = [{
  69                 'ext': 'ttml',
  70                 'url': 'http://www.cbsnews.com/videos/captions/%s.adb_xml' % video_info['mpxRefId'],
  71             }]
  72
  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         }