commit 939be9adfe810ada7dbd5e9032bcfec19fafa14b
parent 2688664762f406b1ba2913af25ee3a2d2ba58038
Author: John Hawkinson <jhawk@mit.edu>
Date: Wed, 22 Nov 2017 11:47:02 -0500
[JWPlatform] Support iframes
Support content.jwplatform... src attributes inside <iframe> tags in
addition to <script> tags. Just a regexp change.
Add a test (currently Generic_72).
Diffstat:
2 files changed, 13 insertions(+), 1 deletion(-)
diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
@@ -1137,6 +1137,18 @@ class GenericIE(InfoExtractor):
}
},
{
+ # JWPlatform iframe
+ 'url': 'https://www.mediaite.com/tv/dem-senator-claims-gary-cohn-faked-a-bad-connection-during-trump-call-to-get-him-off-the-phone/',
+ 'md5': 'ca00a040364b5b439230e7ebfd02c4e9',
+ 'info_dict': {
+ 'id': 'O0c5JcKT',
+ 'ext': 'mp4',
+ 'upload_date': '20171122',
+ 'timestamp': 1511366290,
+ 'title': 'Dem Senator Claims Gary Cohn Faked a Bad Connection During Trump Call to Get Him Off the Phone',
+ },
+ },
+ {
# Video.js embed, multiple formats
'url': 'http://ortcam.com/solidworks-урок-6-настройка-чертежа_33f9b7351.html',
'info_dict': {
diff --git a/youtube_dl/extractor/jwplatform.py b/youtube_dl/extractor/jwplatform.py
@@ -24,7 +24,7 @@ class JWPlatformIE(InfoExtractor):
@staticmethod
def _extract_url(webpage):
mobj = re.search(
- r'<script[^>]+?src=["\'](?P<url>(?:https?:)?//content.jwplatform.com/players/[a-zA-Z0-9]{8})',
+ r'<(script|iframe)[^>]+?src=["\'](?P<url>(?:https?:)?//content.jwplatform.com/players/[a-zA-Z0-9]{8})',
webpage)
if mobj:
return mobj.group('url')