[fktv] Correct thumbnail extraction and add the test

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

diff --git a/youtube_dl/extractor/fktv.py b/youtube_dl/extractor/fktv.py
index 74c6cf8663dd4e815885324240933e9f35e47927..289cbb8c87a374d82f41f2d9840f57f6c4de63dd 100644 (file)
--- a/youtube_dl/extractor/fktv.py
+++ b/youtube_dl/extractor/fktv.py
@@ -21,6 +21,7 @@ class FKTVIE(InfoExtractor):
             'id': '1',
             'ext': 'mp4',
             'title': 'Folge 1 vom 10. April 2007',
+            'thumbnail': 're:^https?://.*\.jpg$',
         },
     }
 
@@ -28,13 +29,16 @@ class FKTVIE(InfoExtractor):
         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)
+        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()
-        urls = re.findall(r'(?s)<source[^>]*src="([^"]+)"[^>]*>', sources)
+        if poster is None:
+            self.report_warning('unable to extract thumbnail')
+
+        urls = re.findall(r'<source[^>]+src="([^"]+)"', sources)
         formats = [{'url': url, 'format_id': determine_ext(url)} for url in urls]
         return {
             'id': episode,