Merge pull request #7045 from remitamine/ign

[youtube-dl] / youtube_dl / extractor / fktv.py

diff --git a/youtube_dl/extractor/fktv.py b/youtube_dl/extractor/fktv.py
index 289cbb8c87a374d82f41f2d9840f57f6c4de63dd..5f6e65daed2d5dc2c18c09a97450ae2a9c88e2df 100644 (file)
--- a/youtube_dl/extractor/fktv.py
+++ b/youtube_dl/extractor/fktv.py
@@ -1,12 +1,10 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     clean_html,
     determine_ext,
-    ExtractorError,
+    js_to_json,
 )
 
 
@@ -28,21 +26,26 @@ class FKTVIE(InfoExtractor):
     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 is None:
-            raise ExtractorError('Unable to extract the video')
-
-        poster, sources = matches.groups()
-        if poster is None:
-            self.report_warning('unable to extract thumbnail')
+        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)
 
-        urls = re.findall(r'<source[^>]+src="([^"]+)"', sources)
-        formats = [{'url': url, 'format_id': determine_ext(url)} for url in urls]
         return {
             'id': episode,
             'title': title,
             'formats': formats,
-            'thumbnail': poster,
+            'thumbnail': thumbnail,
         }