Fix encoding in youtube subtitle download (Closes #669)

[youtube-dl] / youtube_dl / InfoExtractors.py

diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index 50a5a5cfb922c010be12cff4a018ab08ee1cd981..b99a6c505239aeac965e8727c8c106b39eeeaadb 100755 (executable)
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -264,13 +264,18 @@ class YoutubeIE(InfoExtractor):
             srt_lang = list(srt_lang_list.keys())[0]
         if not srt_lang in srt_lang_list:
             return (u'WARNING: no closed captions found in the specified language', None)
-        request = compat_urllib_request.Request('http://www.youtube.com/api/timedtext?lang=%s&name=%s&v=%s' % (srt_lang, srt_lang_list[srt_lang], video_id))
+        params = compat_urllib_parse.urlencode({
+            'lang': srt_lang,
+            'name': srt_lang_list[srt_lang].encode('utf-8'),
+            'v': video_id,
+        })
+        url = 'http://www.youtube.com/api/timedtext?' + params
         try:
-            srt_xml = compat_urllib_request.urlopen(request).read().decode('utf-8')
+            srt_xml = compat_urllib_request.urlopen(url).read().decode('utf-8')
         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
             return (u'WARNING: unable to download video subtitles: %s' % compat_str(err), None)
         if not srt_xml:
-            return (u'WARNING: unable to download video subtitles', None)
+            return (u'WARNING: Did not fetch video subtitles', None)
         return (None, self._closed_captions_xml_to_srt(srt_xml))
 
     def _print_formats(self, formats):
@@ -2050,8 +2055,10 @@ class FacebookIE(InfoExtractor):
         if not m:
             raise ExtractorError(u'Cannot parse data')
         data = dict(json.loads(m.group(1)))
-        video_url = compat_urllib_parse.unquote(data['hd_src'])
-        video_duration = int(data['video_duration'])
+        params_raw = compat_urllib_parse.unquote(data['params'])
+        params = json.loads(params_raw)
+        video_url = params['hd_src']
+        video_duration = int(params['video_duration'])
 
         m = re.search('<h2 class="uiHeaderTitle">([^<]+)</h2>', webpage)
         if not m:
@@ -2064,7 +2071,7 @@ class FacebookIE(InfoExtractor):
             'url': video_url,
             'ext': 'mp4',
             'duration': video_duration,
-            'thumbnail': data['thumbnail_src'],
+            'thumbnail': params['thumbnail_src'],
         }
         return [info]
 
@@ -2869,8 +2876,7 @@ class StanfordOpenClassroomIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         if mobj is None:
-            self._downloader.trouble(u'ERROR: invalid URL: %s' % url)
-            return
+            raise ExtractorError(u'Invalid URL: %s' % url)
 
         if mobj.group('course') and mobj.group('video'): # A specific video
             course = mobj.group('course')
@@ -2907,12 +2913,9 @@ class StanfordOpenClassroomIE(InfoExtractor):
                 'upload_date': None,
             }
 
-            self.report_download_webpage(info['id'])
-            try:
-                coursepage = compat_urllib_request.urlopen(url).read()
-            except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-                self._downloader.trouble(u'ERROR: unable to download course info page: ' + compat_str(err))
-                return
+            coursepage = self._download_webpage(url, info['id'],
+                                        note='Downloading course info page',
+                                        errnote='Unable to download course info page')
 
             m = re.search('<h1>([^<]+)</h1>', coursepage)
             if m:
@@ -2936,7 +2939,6 @@ class StanfordOpenClassroomIE(InfoExtractor):
                 assert entry['type'] == 'reference'
                 results += self.extract(entry['url'])
             return results
-
         else: # Root page
             info = {
                 'id': 'Stanford OpenClassroom',
@@ -3861,7 +3863,7 @@ class YouJizzIE(InfoExtractor):
 
 class EightTracksIE(InfoExtractor):
     IE_NAME = '8tracks'
-    _VALID_URL = r'https?://8tracks.com/(?P<user>[^/]+)/(?P<id>[^/]+)'
+    _VALID_URL = r'https?://8tracks.com/(?P<user>[^/]+)/(?P<id>[^/#]+)(?:#.*)?$'
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -3892,7 +3894,9 @@ class EightTracksIE(InfoExtractor):
             info = {
                 'id': track_data['id'],
                 'url': track_data['track_file_stream_url'],
-                'title': track_data['name'],
+                'title': track_data['performer'] + u' - ' + track_data['name'],
+                'raw_title': track_data['name'],
+                'uploader_id': data['user']['login'],
                 'ext': 'm4a',
             }
             res.append(info)