Merge remote-tracking branch 'joksnet/ytsearch_decode_request'
authorPhilipp Hagemeister <phihag@phihag.de>
Sun, 24 Feb 2013 22:14:34 +0000 (23:14 +0100)
committerPhilipp Hagemeister <phihag@phihag.de>
Sun, 24 Feb 2013 22:14:34 +0000 (23:14 +0100)
1  2 
youtube_dl/InfoExtractors.py

index 378f8a634e9a75525b9d93f1cb5f82ec8bf0b0ea,89e7cd74cc511f8571cc5686986105f77f7d52ff..3b1e14ba97d2b2a8eacd7ed4ec9c0b7716039b7d
@@@ -1472,7 -1472,7 +1472,7 @@@ class YoutubeSearchIE(InfoExtractor)
              result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
              request = compat_urllib_request.Request(result_url)
              try:
-                 data = compat_urllib_request.urlopen(request).read()
+                 data = compat_urllib_request.urlopen(request).read().decode('utf-8')
              except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
                  self._downloader.trouble(u'ERROR: unable to download API page: %s' % compat_str(err))
                  return
@@@ -3627,22 -3627,18 +3627,22 @@@ class SteamIE(InfoExtractor)
          mweb = re.finditer(urlRE, webpage)
          namesRE = r'<span class="title">(?P<videoName>.+?)</span>'
          titles = re.finditer(namesRE, webpage)
 +        thumbsRE = r'<img class="movie_thumb" src="(?P<thumbnail>.+?)">'
 +        thumbs = re.finditer(thumbsRE, webpage)
          videos = []
 -        for vid,vtitle in zip(mweb,titles):
 +        for vid,vtitle,thumb in zip(mweb,titles,thumbs):
              video_id = vid.group('videoID')
              title = vtitle.group('videoName')
              video_url = vid.group('videoURL')
 +            video_thumb = thumb.group('thumbnail')
              if not video_url:
                  self._downloader.trouble(u'ERROR: Cannot find video url for %s' % video_id)
              info = {
                  'id':video_id,
                  'url':video_url,
                  'ext': 'flv',
 -                'title': unescapeHTML(title)
 +                'title': unescapeHTML(title),
 +                'thumbnail': video_thumb
                    }
              videos.append(info)
          return videos
@@@ -4015,30 -4011,31 +4015,30 @@@ class TEDIE(InfoExtractor)
                       ([.\s]*?)data-playlist_item_id="(\d+)"
                       ([.\s]*?)data-mediaslug="(?P<mediaSlug>.+?)"
                       '''
 -        video_name_RE=r'<p\ class="talk-title"><a href="/talks/(.+).html">(?P<fullname>.+?)</a></p>'
 +        video_name_RE=r'<p\ class="talk-title"><a href="(?P<talk_url>/talks/(.+).html)">(?P<fullname>.+?)</a></p>'
          webpage=self._download_webpage(url, playlist_id, 'Downloading playlist webpage')
          m_videos=re.finditer(video_RE,webpage,re.VERBOSE)
          m_names=re.finditer(video_name_RE,webpage)
          info=[]
          for m_video, m_name in zip(m_videos,m_names):
 -            video_dic={
 -                       'id': m_video.group('video_id'),
 -                       'url': self._talk_video_link(m_video.group('mediaSlug')),
 -                       'ext': 'mp4',
 -                       'title': m_name.group('fullname')
 -                       }
 -            info.append(video_dic)
 +            video_id=m_video.group('video_id')
 +            talk_url='http://www.ted.com%s' % m_name.group('talk_url')
 +            info.append(self._talk_info(talk_url,video_id))
          return info
 +
      def _talk_info(self, url, video_id=0):
          """Return the video for the talk in the url"""
          m=re.match(self._VALID_URL, url,re.VERBOSE)
          videoName=m.group('name')
          webpage=self._download_webpage(url, video_id, 'Downloading \"%s\" page' % videoName)
          # If the url includes the language we get the title translated
 -        title_RE=r'<h1><span id="altHeadline" >(?P<title>[\s\w:/\.\?=\+-\\\']*)</span></h1>'
 +        title_RE=r'<h1><span id="altHeadline" >(?P<title>.*)</span></h1>'
          title=re.search(title_RE, webpage).group('title')
          info_RE=r'''<script\ type="text/javascript">var\ talkDetails\ =(.*?)
                          "id":(?P<videoID>[\d]+).*?
                          "mediaSlug":"(?P<mediaSlug>[\w\d]+?)"'''
 +        thumb_RE=r'</span>[\s.]*</div>[\s.]*<img src="(?P<thumbnail>.*?)"'
 +        thumb_match=re.search(thumb_RE,webpage)
          info_match=re.search(info_RE,webpage,re.VERBOSE)
          video_id=info_match.group('videoID')
          mediaSlug=info_match.group('mediaSlug')
                  'id': video_id,
                  'url': video_url,
                  'ext': 'mp4',
 -                'title': title
 +                'title': title,
 +                'thumbnail': thumb_match.group('thumbnail')
                  }
          return info