From 8aa0bd5d10627ece3c1815c01d02fb8bf22847a7 Mon Sep 17 00:00:00 2001
From: bashonly <bashonly@bashonly.com>
Date: Sun, 29 Jan 2023 00:59:37 -0600
Subject: [PATCH] [extractor/generic] Avoid catastrophic backtracking in KVS
regex
Authored by: bashonly
---
yt_dlp/extractor/generic.py | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/yt_dlp/extractor/generic.py b/yt_dlp/extractor/generic.py
index 04677b23f1..9e4df4cead 100644
--- a/yt_dlp/extractor/generic.py
+++ b/yt_dlp/extractor/generic.py
@@ -2637,11 +2637,11 @@ class GenericIE(InfoExtractor):
# Look for generic KVS player (before json-ld bc of some urls that break otherwise)
found = self._search_regex((
- r'<script\b[^>]+?\bsrc\s*=\s*(["\'])https?://(?:\S+?/)+kt_player\.js\?v=(?P<ver>\d+(?:\.\d+)+)\1[^>]*>',
- r'kt_player\s*\(\s*(["\'])(?:(?!\1)[\w\W])+\1\s*,\s*(["\'])https?://(?:\S+?/)+kt_player\.swf\?v=(?P<ver>\d+(?:\.\d+)+)\2\s*,',
+ r'<script\b[^>]+?\bsrc\s*=\s*(["\'])https?://(?:(?!\1)[^?#])+/kt_player\.js\?v=(?P<ver>\d+(?:\.\d+)+)\1[^>]*>',
+ r'kt_player\s*\(\s*(["\'])(?:(?!\1)[\w\W])+\1\s*,\s*(["\'])https?://(?:(?!\2)[^?#])+/kt_player\.swf\?v=(?P<ver>\d+(?:\.\d+)+)\2\s*,',
), webpage, 'KVS player', group='ver', default=False)
if found:
- self.report_detected('KWS Player')
+ self.report_detected('KVS Player')
if found.split('.')[0] not in ('4', '5', '6'):
self.report_warning(f'Untested major version ({found}) in player engine - download may fail.')
return [self._extract_kvs(url, webpage, video_id)]