From ffb5b6f8b33ba246b4aeacd7b46e0d834c266f51 Mon Sep 17 00:00:00 2001
From: Kovid Goyal <kovid@kovidgoyal.net>
Date: Wed, 24 Nov 2010 23:19:06 -0700
Subject: [PATCH] MOBI Input: Workaround MOBI files that have invalid
 <p><blockquote> constructs (<blockquote> cannot be nested inside <p> in
 XHTML)

---
 src/calibre/ebooks/mobi/reader.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/src/calibre/ebooks/mobi/reader.py b/src/calibre/ebooks/mobi/reader.py
index c4845f9443..309023ede9 100644
--- a/src/calibre/ebooks/mobi/reader.py
+++ b/src/calibre/ebooks/mobi/reader.py
@@ -480,6 +480,9 @@ class MobiReader(object):
         # - lxml and beautifulsoup expect/assume a specific order based on xhtml spec
         self.processed_html = re.sub(r'(?i)(?P<styletags>(<(h\d+|i|b|u|em|small|big|strong|tt)>\s*){1,})(?P<para><p[^>]*>)', '\g<para>'+'\g<styletags>', self.processed_html)
         self.processed_html = re.sub(r'(?i)(?P<para></p[^>]*>)\s*(?P<styletags>(</(h\d+|i|b|u|em|small|big|strong|tt)>\s*){1,})', '\g<styletags>'+'\g<para>', self.processed_html)
+        self.processed_html = re.sub(r'(?i)(?P<blockquote>(</blockquote[^>]*>\s*){1,})(?P<para></p[^>]*>)', '\g<para>'+'\g<blockquote>', self.processed_html)
+        self.processed_html = re.sub(r'(?i)(?P<para><p[^>]*>)\s*(?P<blockquote>(<blockquote[^>]*>\s*){1,})', '\g<blockquote>'+'\g<para>', self.processed_html)
+
 
     def remove_random_bytes(self, html):
         return re.sub('\x14|\x15|\x19|\x1c|\x1d|\xef|\x12|\x13|\xec|\x08',