[fktv] Correct thumbnail extraction and add the test

This commit is contained in:
Yen Chi Hsuan 2015-09-25 17:58:44 +08:00
parent 140359fc2c
commit 5773803961

View file

@ -21,6 +21,7 @@ class FKTVIE(InfoExtractor):
'id': '1', 'id': '1',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Folge 1 vom 10. April 2007', 'title': 'Folge 1 vom 10. April 2007',
'thumbnail': 're:^https?://.*\.jpg$',
}, },
} }
@ -29,11 +30,14 @@ def _real_extract(self, url):
webpage = self._download_webpage('http://fernsehkritik.tv/folge-%s/play' % episode, episode) webpage = self._download_webpage('http://fernsehkritik.tv/folge-%s/play' % episode, episode)
title = clean_html(self._html_search_regex('<h3>([^<]+)</h3>', webpage, 'title')) title = clean_html(self._html_search_regex('<h3>([^<]+)</h3>', webpage, 'title'))
matches = re.search(r'(?s)<video[^>]+poster="([^"]+)"[^>]*>(.*)</video>', webpage) matches = re.search(r'(?s)<video[^>]+(?:poster="([^"]+)")?[^>]*>(.*)</video>', webpage)
if matches is None: if matches is None:
raise ExtractorError('Unable to extract the video') raise ExtractorError('Unable to extract the video')
poster, sources = matches.groups() poster, sources = matches.groups()
if poster is None:
self.report_warning('unable to extract thumbnail')
urls = re.findall(r'<source[^>]+src="([^"]+)"', sources) urls = re.findall(r'<source[^>]+src="([^"]+)"', sources)
formats = [{'url': url, 'format_id': determine_ext(url)} for url in urls] formats = [{'url': url, 'format_id': determine_ext(url)} for url in urls]
return { return {