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

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     compat_urllib_request,
   6     compat_urllib_parse,
   7
   8     ExtractorError,
   9     unescapeHTML,
  10 )
  11
  12 class DailymotionIE(InfoExtractor):
  13     """Information Extractor for Dailymotion"""
  14
  15     _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/video/([^/]+)'
  16     IE_NAME = u'dailymotion'
  17
  18     def _real_extract(self, url):
  19         # Extract id and simplified title from URL
  20         mobj = re.match(self._VALID_URL, url)
  21
  22         video_id = mobj.group(1).split('_')[0].split('?')[0]
  23
  24         video_extension = 'mp4'
  25
  26         # Retrieve video webpage to extract further information
  27         request = compat_urllib_request.Request(url)
  28         request.add_header('Cookie', 'family_filter=off')
  29         webpage = self._download_webpage(request, video_id)
  30
  31         # Extract URL, uploader and title from webpage
  32         self.report_extraction(video_id)
  33         mobj = re.search(r'\s*var flashvars = (.*)', webpage)
  34         if mobj is None:
  35             raise ExtractorError(u'Unable to extract media URL')
  36         flashvars = compat_urllib_parse.unquote(mobj.group(1))
  37
  38         for key in ['hd1080URL', 'hd720URL', 'hqURL', 'sdURL', 'ldURL', 'video_url']:
  39             if key in flashvars:
  40                 max_quality = key
  41                 self.to_screen(u'Using %s' % key)
  42                 break
  43         else:
  44             raise ExtractorError(u'Unable to extract video URL')
  45
  46         mobj = re.search(r'"' + max_quality + r'":"(.+?)"', flashvars)
  47         if mobj is None:
  48             raise ExtractorError(u'Unable to extract video URL')
  49
  50         video_url = compat_urllib_parse.unquote(mobj.group(1)).replace('\\/', '/')
  51
  52         # TODO: support choosing qualities
  53
  54         mobj = re.search(r'<meta property="og:title" content="(?P<title>[^"]*)" />', webpage)
  55         if mobj is None:
  56             raise ExtractorError(u'Unable to extract title')
  57         video_title = unescapeHTML(mobj.group('title'))
  58
  59         video_uploader = None
  60         video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
  61                                              # Looking for official user
  62                                              r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
  63                                             webpage, 'video uploader')
  64
  65         video_upload_date = None
  66         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
  67         if mobj is not None:
  68             video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
  69
  70         return [{
  71             'id':       video_id,
  72             'url':      video_url,
  73             'uploader': video_uploader,
  74             'upload_date':  video_upload_date,
  75             'title':    video_title,
  76             'ext':      video_extension,
  77         }]