[ard:mediathek] Fix title and description extraction (closes #18349)

pull/8/head
Alexander Seiler 6 years ago committed by Sergey M
parent ebb0449049
commit 8c5879715f

@ -173,13 +173,18 @@ class ARDMediathekIE(InfoExtractor):
title = self._html_search_regex( title = self._html_search_regex(
[r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>', [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
r'<meta name="dcterms\.title" content="(.*?)"/>', r'<meta name="dcterms\.title" content="(.*?)"/>',
r'<h4 class="headline">(.*?)</h4>'], r'<h4 class="headline">(.*?)</h4>',
r'<title[^>]*>(.*?)</title>'],
webpage, 'title') webpage, 'title')
description = self._html_search_meta( description = self._html_search_meta(
'dcterms.abstract', webpage, 'description', default=None) 'dcterms.abstract', webpage, 'description', default=None)
if description is None: if description is None:
description = self._html_search_meta( description = self._html_search_meta(
'description', webpage, 'meta description') 'description', webpage, 'meta description', default=None)
if description is None:
description = self._html_search_regex(
r'<p\s+class="teasertext">(.+?)</p>',
webpage, 'teaser text', default=None)
# Thumbnail is sometimes not present. # Thumbnail is sometimes not present.
# It is in the mobile version, but that seems to use a different URL # It is in the mobile version, but that seems to use a different URL

Loading…
Cancel
Save