From 60e47a2699bea32f12a92e59d02d1ba0e9cbdec9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
 <jaime.marquinez.ferrandiz@gmail.com>
Date: Wed, 31 Dec 2014 15:44:15 +0100
Subject: [PATCH] [youtube] Use '_download_xml' for getting the available
 subtitles

---
 youtube_dl/extractor/youtube.py | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index 224f1b041..7f5aeb25b 100644
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -612,24 +612,23 @@ def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
 
     def _get_available_subtitles(self, video_id, webpage):
         try:
-            sub_list = self._download_webpage(
+            subs_doc = self._download_xml(
                 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
                 video_id, note=False)
         except ExtractorError as err:
             self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
             return {}
-        lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
 
         sub_lang_list = {}
-        for l in lang_list:
-            lang = l[1]
+        for track in subs_doc.findall('track'):
+            lang = track.attrib['lang_code']
             if lang in sub_lang_list:
                 continue
             params = compat_urllib_parse.urlencode({
                 'lang': lang,
                 'v': video_id,
                 'fmt': self._downloader.params.get('subtitlesformat', 'srt'),
-                'name': unescapeHTML(l[0]).encode('utf-8'),
+                'name': track.attrib['name'].encode('utf-8'),
             })
             url = 'https://www.youtube.com/api/timedtext?' + params
             sub_lang_list[lang] = url