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

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     compat_urllib_parse_urlparse,
   6     determine_ext,
   7
   8     ExtractorError,
   9 )
  10
  11
  12 class CollegeHumorIE(InfoExtractor):
  13     _VALID_URL = r'^(?:https?://)?(?:www\.)?collegehumor\.com/(video|embed|e)/(?P<videoid>[0-9]+)/?(?P<shorttitle>.*)$'
  14
  15     _TESTS = [{
  16         u'url': u'http://www.collegehumor.com/video/6902724/comic-con-cosplay-catastrophe',
  17         u'file': u'6902724.mp4',
  18         u'md5': u'1264c12ad95dca142a9f0bf7968105a0',
  19         u'info_dict': {
  20             u'title': u'Comic-Con Cosplay Catastrophe',
  21             u'description': u'Fans get creative this year at San Diego.  Too creative.  And yes, that\'s really Joss Whedon.',
  22         },
  23     },
  24     {
  25         u'url': u'http://www.collegehumor.com/video/3505939/font-conference',
  26         u'file': u'3505939.mp4',
  27         u'md5': u'c51ca16b82bb456a4397987791a835f5',
  28         u'info_dict': {
  29             u'title': u'Font Conference',
  30             u'description': u'This video wasn\'t long enough, so we made it double-spaced.',
  31         },
  32     }]
  33
  34     def _real_extract(self, url):
  35         mobj = re.match(self._VALID_URL, url)
  36         if mobj is None:
  37             raise ExtractorError(u'Invalid URL: %s' % url)
  38         video_id = mobj.group('videoid')
  39
  40         info = {
  41             'id': video_id,
  42             'uploader': None,
  43             'upload_date': None,
  44         }
  45
  46         self.report_extraction(video_id)
  47         xmlUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id
  48         mdoc = self._download_xml(xmlUrl, video_id,
  49                                          u'Downloading info XML',
  50                                          u'Unable to download video info XML')
  51
  52         try:
  53             videoNode = mdoc.findall('./video')[0]
  54             youtubeIdNode = videoNode.find('./youtubeID')
  55             if youtubeIdNode is not None:
  56                 return self.url_result(youtubeIdNode.text, 'Youtube')
  57             info['description'] = videoNode.findall('./description')[0].text
  58             info['title'] = videoNode.findall('./caption')[0].text
  59             info['thumbnail'] = videoNode.findall('./thumbnail')[0].text
  60             next_url = videoNode.findall('./file')[0].text
  61         except IndexError:
  62             raise ExtractorError(u'Invalid metadata XML file')
  63
  64         if next_url.endswith(u'manifest.f4m'):
  65             manifest_url = next_url + '?hdcore=2.10.3'
  66             adoc = self._download_xml(manifest_url, video_id,
  67                                          u'Downloading XML manifest',
  68                                          u'Unable to download video info XML')
  69
  70             try:
  71                 video_id = adoc.findall('./{http://ns.adobe.com/f4m/1.0}id')[0].text
  72             except IndexError:
  73                 raise ExtractorError(u'Invalid manifest file')
  74             url_pr = compat_urllib_parse_urlparse(info['thumbnail'])
  75             info['url'] = url_pr.scheme + '://' + url_pr.netloc + video_id[:-2].replace('.csmil','').replace(',','')
  76             info['ext'] = 'mp4'
  77         else:
  78             # Old-style direct links
  79             info['url'] = next_url
  80             info['ext'] = determine_ext(info['url'])
  81
  82         return info