_ 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 )
  10
  11 class DailymotionIE(InfoExtractor):
  12     """Information Extractor for Dailymotion"""
  13
  14     _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/video/([^/]+)'
  15     IE_NAME = u'dailymotion'
  16     _TEST = {
  17         u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
  18         u'file': u'x33vw9.mp4',
  19         u'md5': u'392c4b85a60a90dc4792da41ce3144eb',
  20         u'info_dict': {
  21             u"uploader": u"Alex and Van .",
  22             u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\""
  23         }
  24     }
  25
  26     def _real_extract(self, url):
  27         # Extract id and simplified title from URL
  28         mobj = re.match(self._VALID_URL, url)
  29
  30         video_id = mobj.group(1).split('_')[0].split('?')[0]
  31
  32         video_extension = 'mp4'
  33
  34         # Retrieve video webpage to extract further information
  35         request = compat_urllib_request.Request(url)
  36         request.add_header('Cookie', 'family_filter=off')
  37         webpage = self._download_webpage(request, video_id)
  38
  39         # Extract URL, uploader and title from webpage
  40         self.report_extraction(video_id)
  41
  42         video_title = self._html_search_regex(r'<meta property="og:title" content="(.*?)" />',
  43                                               webpage, 'title')
  44
  45         video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
  46                                              # Looking for official user
  47                                              r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
  48                                             webpage, 'video uploader')
  49
  50         video_upload_date = None
  51         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
  52         if mobj is not None:
  53             video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
  54
  55         embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
  56         embed_page = self._download_webpage(embed_url, video_id,
  57                                             u'Downloading embed page')
  58         info = self._search_regex(r'var info = ({.*?}),', embed_page, 'video info')
  59         info = json.loads(info)
  60
  61         # TODO: support choosing qualities
  62
  63         for key in ['stream_h264_hd1080_url','stream_h264_hd_url',
  64                     'stream_h264_hq_url','stream_h264_url',
  65                     'stream_h264_ld_url']:
  66             if info.get(key):#key in info and info[key]:
  67                 max_quality = key
  68                 self.to_screen(u'Using %s' % key)
  69                 break
  70         else:
  71             raise ExtractorError(u'Unable to extract video URL')
  72         video_url = info[max_quality]
  73
  74         return [{
  75             'id':       video_id,
  76             'url':      video_url,
  77             'uploader': video_uploader,
  78             'upload_date':  video_upload_date,
  79             'title':    video_title,
  80             'ext':      video_extension,
  81             'thumbnail': info['thumbnail_url']
  82         }]