X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;ds=sidebyside;f=youtube_dl%2Fextractor%2Ffktv.py;h=5f6e65daed2d5dc2c18c09a97450ae2a9c88e2df;hb=9f0ee2a3883ec6f6fdccba90085cb925aaa2f617;hp=93c4fd641b34a77383eb87abc8b9f6f469252704;hpb=2e40a12225fbe64b84b3975b3063a676df0f4522;p=youtube-dl

diff --git a/youtube_dl/extractor/fktv.py b/youtube_dl/extractor/fktv.py
index 93c4fd641..5f6e65dae 100644
--- a/youtube_dl/extractor/fktv.py
+++ b/youtube_dl/extractor/fktv.py
@@ -1,11 +1,10 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     clean_html,
     determine_ext,
+    js_to_json,
 )
 
 
@@ -20,23 +19,33 @@ class FKTVIE(InfoExtractor):
             'id': '1',
             'ext': 'mp4',
             'title': 'Folge 1 vom 10. April 2007',
+            'thumbnail': 're:^https?://.*\.jpg$',
         },
     }
 
     def _real_extract(self, url):
         episode = self._match_id(url)
 
-        webpage = self._download_webpage('http://fernsehkritik.tv/folge-%s/play' % episode, episode)
-        title = clean_html(self._html_search_regex('<h3>([^<]+?)</h3>', webpage, 'title'))
-        matches = re.search(r'(?s)<video[^>]*poster="([^"]+)"[^>]*>(.*?)</video>', webpage)
-        if matches:
-            poster, sources = matches.groups()
-            urls = re.findall(r'(?s)<source[^>]*src="([^"]+)"[^>]*>', sources)
-            if sources:
-                formats = [{'url': url, 'format_id': determine_ext(url)} for url in urls]
-                return {
-                    'id': episode,
-                    'title': title,
-                    'formats': formats,
-                    'thumbnail': poster,
-                }
+        webpage = self._download_webpage(
+            'http://fernsehkritik.tv/folge-%s/play' % episode, episode)
+        title = clean_html(self._html_search_regex(
+            '<h3>([^<]+)</h3>', webpage, 'title'))
+        thumbnail = self._search_regex(r'POSTER\s*=\s*"([^"]+)', webpage, 'thumbnail', fatal=False)
+        sources = self._parse_json(self._search_regex(r'(?s)MEDIA\s*=\s*(\[.+?\]);', webpage, 'media'), episode, js_to_json)
+
+        formats = []
+        for source in sources:
+            furl = source.get('src')
+            if furl:
+                formats.append({
+                    'url': furl,
+                    'format_id': determine_ext(furl),
+                })
+        self._sort_formats(formats)
+
+        return {
+            'id': episode,
+            'title': title,
+            'formats': formats,
+            'thumbnail': thumbnail,
+        }