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

   1 import json
   2 import re
   3
   4 from .common import InfoExtractor
   5
   6
   7 class CollegeHumorIE(InfoExtractor):
   8     _VALID_URL = r'^(?:https?://)?(?:www\.)?collegehumor\.com/(video|embed|e)/(?P<videoid>[0-9]+)/?(?P<shorttitle>.*)$'
   9
  10     _TESTS = [{
  11         u'url': u'http://www.collegehumor.com/video/6902724/comic-con-cosplay-catastrophe',
  12         u'file': u'6902724.mp4',
  13         u'md5': u'dcc0f5c1c8be98dc33889a191f4c26bd',
  14         u'info_dict': {
  15             u'title': u'Comic-Con Cosplay Catastrophe',
  16             u'description': u'Fans get creative this year at San Diego.  Too',
  17         },
  18     },
  19     {
  20         u'url': u'http://www.collegehumor.com/video/3505939/font-conference',
  21         u'file': u'3505939.mp4',
  22         u'md5': u'72fa701d8ef38664a4dbb9e2ab721816',
  23         u'info_dict': {
  24             u'title': u'Font Conference',
  25             u'description': u'This video wasn\'t long enough, so we made it double-spaced.',
  26         },
  27     }]
  28
  29     def _real_extract(self, url):
  30         mobj = re.match(self._VALID_URL, url)
  31         video_id = mobj.group('videoid')
  32
  33         jsonUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id + '.json'
  34         data = json.loads(self._download_webpage(
  35             jsonUrl, video_id, u'Downloading info JSON'))
  36         vdata = data['video']
  37
  38         PREFS = {'high_quality': 2, 'low_quality': 0}
  39         formats = []
  40         for format_key in ('mp4', 'webm'):
  41             for qname, qurl in vdata[format_key].items():
  42                 formats.append({
  43                     'format_id': format_key + '_' + qname,
  44                     'url': qurl,
  45                     'format': format_key,
  46                     'preference': PREFS.get(qname),
  47                 })
  48
  49         self._sort_formats(formats)
  50
  51         return {
  52             'id': video_id,
  53             'title': vdata['title'],
  54             'description': vdata.get('description'),
  55             'thumbnail': vdata.get('thumbnail'),
  56             'formats': formats,
  57         }