Fix typo

[youtube-dl] / youtube_dl / InfoExtractors.py

diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index 23bd21af51a3d64ccf987c5ba5f8fb89d8d0902a..d661d517dea96e09819ec52f58cf0205f40b0d72 100755 (executable)
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -1366,6 +1366,9 @@ class GenericIE(InfoExtractor):
         if mobj is None:
             # Broaden the search a little bit
             mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
+        if mobj is None:
+            # Broaden the search a little bit: JWPlayer JS loader
+            mobj = re.search(r'[^A-Za-z0-9]?file:\s*["\'](http[^\'"&]*)', webpage)
         if mobj is None:
             self._downloader.trouble(u'ERROR: Invalid URL: %s' % url)
             return
@@ -1469,7 +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
@@ -1918,9 +1921,8 @@ class BlipTVUserIE(InfoExtractor):
 
         while True:
             self.report_download_page(username, pagenum)
-
-            request = compat_urllib_request.Request( page_base + "&page=" + str(pagenum) )
-
+            url = page_base + "&page=" + str(pagenum)
+            request = compat_urllib_request.Request( url )
             try:
                 page = compat_urllib_request.urlopen(request).read().decode('utf-8')
             except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
@@ -3624,18 +3626,22 @@ 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
@@ -4008,31 +4014,30 @@ 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')
@@ -4041,7 +4046,8 @@ class TEDIE(InfoExtractor):
                 'id': video_id,
                 'url': video_url,
                 'ext': 'mp4',
-                'title': title
+                'title': title,
+                'thumbnail': thumb_match.group('thumbnail')
                 }
         return info