From 70b99872c670aeb11f2cfa11ed8ff82a79f129c2 Mon Sep 17 00:00:00 2001
From: Kovid Goyal <kovid@kovidgoyal.net>
Date: Fri, 20 Feb 2009 13:07:32 -0800
Subject: [PATCH] EPUB Output:Be less aggressive when stripping invalid HTML
 constructs inserted by MS Word. Fixes regression in the Time recipe.

---
 src/calibre/ebooks/html.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/src/calibre/ebooks/html.py b/src/calibre/ebooks/html.py
index 1c15973d3b..1f1e6b94b1 100644
--- a/src/calibre/ebooks/html.py
+++ b/src/calibre/ebooks/html.py
@@ -331,7 +331,8 @@ class PreProcessor(object):
                   # Convert all entities, since lxml doesn't handle them well
                   (re.compile(r'&(\S+?);'), convert_entities),
                   # Remove the <![if/endif tags inserted by everybody's darling, MS Word
-                  (re.compile(r'(?i)<{0,1}!\[(end){0,1}if[^>]*>'), lambda match: ''),
+                  (re.compile(r'</{0,1}!\[(end){0,1}if\]{0,1}>', re.IGNORECASE), 
+                   lambda match: ''),
                   ]
                      
     # Fix pdftohtml markup