From 99ef96f84c2c9fa1267d6edb836c8f1945295424 Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Thu, 21 Apr 2016 18:07:03 +0800 Subject: [PATCH] [gdcvault] Fix for videos with hard-coded hostnames Fixes #9248 --- youtube_dl/extractor/gdcvault.py | 28 ++++++++++++++++++++++------ 1 file changed, 22 insertions(+), 6 deletions(-) diff --git a/youtube_dl/extractor/gdcvault.py b/youtube_dl/extractor/gdcvault.py index 25e93c9a43..3ebcaf7339 100644 --- a/youtube_dl/extractor/gdcvault.py +++ b/youtube_dl/extractor/gdcvault.py @@ -51,18 +51,34 @@ class GDCVaultIE(InfoExtractor): { 'url': 'http://gdcvault.com/play/1020791/', 'only_matching': True, - } + }, + { + 'url': 'http://gdcvault.com/play/1023460/Tenacious-Design-and-The-Interface', + 'md5': 'a8efb6c31ed06ca8739294960b2dbabd', + 'info_dict': { + 'id': '1023460', + 'ext': 'mp4', + 'display_id': 'Tenacious-Design-and-The-Interface', + 'title': 'Tenacious Design and The Interface of \'Destiny\'', + }, + }, ] def _parse_mp4(self, xml_description): video_formats = [] - mp4_video = xml_description.find('./metadata/mp4video') - if mp4_video is None: - return None + video_root = None + + mp4_video = xml_description.find('./metadata/mp4video') + if mp4_video is not None: + mobj = re.match(r'(?Phttps?://.*?/).*', mp4_video.text) + video_root = mobj.group('root') + if video_root is None: + # Hard-coded in http://evt.dispeak.com/ubm/gdc/sf16/custom/player2.js + video_root = 'http://s3-2u.digitallyspeaking.com/' - mobj = re.match(r'(?Phttps?://.*?/).*', mp4_video.text) - video_root = mobj.group('root') formats = xml_description.findall('./metadata/MBRVideos/MBRVideo') + if not formats: + return None for format in formats: mobj = re.match(r'mp4\:(?P.*)', format.find('streamName').text) url = video_root + mobj.group('path')