From 6b3fbd3425180a00d99e91e6f2b9b5896c567ea4 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Thu, 11 Feb 2016 22:02:37 +0600 Subject: [PATCH] [pbs] Fix multi part videos extraction --- youtube_dl/extractor/pbs.py | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/youtube_dl/extractor/pbs.py b/youtube_dl/extractor/pbs.py index c159c00718..cca012953d 100644 --- a/youtube_dl/extractor/pbs.py +++ b/youtube_dl/extractor/pbs.py @@ -366,10 +366,14 @@ class PBSIE(InfoExtractor): webpage, 'upload date', default=None)) # tabbed frontline videos - tabbed_videos = re.findall( - r']+class="videotab[^"]*"[^>]+vid="(\d+)"', webpage) - if tabbed_videos: - return tabbed_videos, presumptive_id, upload_date + MULTI_PART_REGEXES = ( + r']+class="videotab[^"]*"[^>]+vid="(\d+)"', + r']+href=["\']#video-\d+["\'][^>]+data-coveid=["\'](\d+)', + ) + for p in MULTI_PART_REGEXES: + tabbed_videos = re.findall(p, webpage) + if tabbed_videos: + return tabbed_videos, presumptive_id, upload_date MEDIA_ID_REGEXES = [ r"div\s*:\s*'videoembed'\s*,\s*mediaid\s*:\s*'(\d+)'", # frontline video embed