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

   1 import re
   2 import json
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     compat_urllib_request,
   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     _TEST = {
  18         u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
  19         u'file': u'x33vw9.mp4',
  20         u'md5': u'392c4b85a60a90dc4792da41ce3144eb',
  21         u'info_dict': {
  22             u"uploader": u"Alex and Van .",
  23             u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\""
  24         }
  25     }
  26
  27     def _real_extract(self, url):
  28         # Extract id and simplified title from URL
  29         mobj = re.match(self._VALID_URL, url)
  30
  31         video_id = mobj.group(1).split('_')[0].split('?')[0]
  32
  33         video_extension = 'mp4'
  34
  35         # Retrieve video webpage to extract further information
  36         request = compat_urllib_request.Request(url)
  37         request.add_header('Cookie', 'family_filter=off')
  38         webpage = self._download_webpage(request, video_id)
  39
  40         # Extract URL, uploader and title from webpage
  41         self.report_extraction(video_id)
  42
  43
  44         mobj = re.search(r'<meta property="og:title" content="(?P<title>[^"]*)" />', webpage)
  45         if mobj is None:
  46             raise ExtractorError(u'Unable to extract title')
  47         video_title = unescapeHTML(mobj.group('title'))
  48
  49         video_uploader = None
  50         video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
  51                                              # Looking for official user
  52                                              r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
  53                                             webpage, 'video uploader')
  54
  55         video_upload_date = None
  56         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
  57         if mobj is not None:
  58             video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
  59
  60         embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
  61         embed_page = self._download_webpage(embed_url, video_id,
  62                                             u'Downloading embed page')
  63         info = self._search_regex(r'var info = ({.*?}),', embed_page, 'video info')
  64         info = json.loads(info)
  65
  66         # TODO: support choosing qualities
  67
  68         for key in ['stream_h264_hd1080_url','stream_h264_hd_url',
  69                     'stream_h264_hq_url','stream_h264_url',
  70                     'stream_h264_ld_url']:
  71             if info.get(key):#key in info and info[key]:
  72                 max_quality = key
  73                 self.to_screen(u'Using %s' % key)
  74                 break
  75         else:
  76             raise ExtractorError(u'Unable to extract video URL')
  77         video_url = info[max_quality]
  78
  79         return [{
  80             'id':       video_id,
  81             'url':      video_url,
  82             'uploader': video_uploader,
  83             'upload_date':  video_upload_date,
  84             'title':    video_title,
  85             'ext':      video_extension,
  86         }]