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

   1 import re
   2 import xml.etree.ElementTree
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     compat_urllib_parse_urlparse,
   7
   8     ExtractorError,
   9 )
  10
  11
  12 class CollegeHumorIE(InfoExtractor):
  13     _VALID_URL = r'^(?:https?://)?(?:www\.)?collegehumor\.com/video/(?P<videoid>[0-9]+)/(?P<shorttitle>.*)$'
  14
  15     _TEST = {
  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     def _real_extract(self, url):
  26         mobj = re.match(self._VALID_URL, url)
  27         if mobj is None:
  28             raise ExtractorError(u'Invalid URL: %s' % url)
  29         video_id = mobj.group('videoid')
  30
  31         info = {
  32             'id': video_id,
  33             'uploader': None,
  34             'upload_date': None,
  35         }
  36
  37         self.report_extraction(video_id)
  38         xmlUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id
  39         metaXml = self._download_webpage(xmlUrl, video_id,
  40                                          u'Downloading info XML',
  41                                          u'Unable to download video info XML')
  42
  43         mdoc = xml.etree.ElementTree.fromstring(metaXml)
  44         try:
  45             videoNode = mdoc.findall('./video')[0]
  46             info['description'] = videoNode.findall('./description')[0].text
  47             info['title'] = videoNode.findall('./caption')[0].text
  48             info['thumbnail'] = videoNode.findall('./thumbnail')[0].text
  49             manifest_url = videoNode.findall('./file')[0].text
  50         except IndexError:
  51             raise ExtractorError(u'Invalid metadata XML file')
  52
  53         manifest_url += '?hdcore=2.10.3'
  54         manifestXml = self._download_webpage(manifest_url, video_id,
  55                                              u'Downloading XML manifest',
  56                                              u'Unable to download video info XML')
  57
  58         adoc = xml.etree.ElementTree.fromstring(manifestXml)
  59         try:
  60             media_node = adoc.findall('./{http://ns.adobe.com/f4m/1.0}media')[0]
  61             node_id = media_node.attrib['url']
  62             video_id = adoc.findall('./{http://ns.adobe.com/f4m/1.0}id')[0].text
  63         except IndexError as err:
  64             raise ExtractorError(u'Invalid manifest file')
  65
  66         url_pr = compat_urllib_parse_urlparse(info['thumbnail'])
  67
  68         info['url'] = url_pr.scheme + '://' + url_pr.netloc + video_id[:-2].replace('.csmil','').replace(',','')
  69         info['ext'] = 'mp4'
  70         return [info]