From dd12f725d5c9fb668ce91bce8997ee883c71a731 Mon Sep 17 00:00:00 2001
From: John Schember <john@nachtimwald.com>
Date: Sun, 2 Aug 2009 14:14:14 -0400
Subject: [PATCH] Fix bug #3050: Prepare string for xml

---
 src/calibre/ebooks/fb2/fb2ml.py | 15 +++++----------
 src/calibre/ebooks/rb/rbml.py   |  5 +++--
 2 files changed, 8 insertions(+), 12 deletions(-)

diff --git a/src/calibre/ebooks/fb2/fb2ml.py b/src/calibre/ebooks/fb2/fb2ml.py
index 11a718886f..2de2c599cf 100644
--- a/src/calibre/ebooks/fb2/fb2ml.py
+++ b/src/calibre/ebooks/fb2/fb2ml.py
@@ -13,10 +13,11 @@ from base64 import b64encode
 
 from lxml import etree
 
+from calibre import prepare_string_for_xml
+from calibre.constants import __appname__, __version__
 from calibre.ebooks.oeb.base import XHTML, XHTML_NS, barename, namespace
 from calibre.ebooks.oeb.stylizer import Stylizer
 from calibre.ebooks.oeb.base import OEB_IMAGES
-from calibre.constants import __appname__, __version__
 
 TAG_MAP = {
     'b' : 'strong',
@@ -56,7 +57,6 @@ class FB2MLizer(object):
         output += self.fb2_body_footer()
         output += self.fb2mlize_images()
         output += self.fb2_footer()
-        output = self.clean_text(output)
         return u'<?xml version="1.0" encoding="UTF-8"?>\n%s' % etree.tostring(etree.fromstring(output), encoding=unicode, pretty_print=True)
 
     def fb2_header(self):
@@ -90,11 +90,6 @@ class FB2MLizer(object):
                 images += '<binary id="%s" content-type="%s">%s\n</binary>' % (os.path.basename(item.href),  item.media_type, data)
         return images
 
-    def clean_text(self, text):
-        text = text.replace('&', '')
-
-        return text
-
     def dump_text(self, elem, stylizer, tag_stack=[]):
         if not isinstance(elem.tag, basestring) \
            or namespace(elem.tag) != XHTML_NS:
@@ -130,7 +125,7 @@ class FB2MLizer(object):
                 tag_stack.append(style_tag)
 
         if hasattr(elem, 'text') and elem.text != None and elem.text.strip() != '':
-            fb2_text += elem.text
+            fb2_text += prepare_string_for_xml(elem.text)
         
         for item in elem:
             fb2_text += self.dump_text(item, stylizer, tag_stack)
@@ -142,9 +137,9 @@ class FB2MLizer(object):
 
         if hasattr(elem, 'tail') and elem.tail != None and elem.tail.strip() != '':
             if 'p' not in tag_stack:
-                fb2_text += '<p>%s</p>' % elem.tail
+                fb2_text += '<p>%s</p>' % prepare_string_for_xml(elem.tail)
             else:
-                fb2_text += elem.tail
+                fb2_text += prepare_string_for_xml(elem.tail)
             
         return fb2_text
 
diff --git a/src/calibre/ebooks/rb/rbml.py b/src/calibre/ebooks/rb/rbml.py
index bc9248f8b0..82f8a17281 100644
--- a/src/calibre/ebooks/rb/rbml.py
+++ b/src/calibre/ebooks/rb/rbml.py
@@ -11,6 +11,7 @@ Transform OEB content into RB compatible markup.
 import os
 import re
 
+from calibre import prepare_string_for_xml
 from calibre.ebooks.oeb.base import XHTML, XHTML_NS, barename, namespace
 from calibre.ebooks.oeb.stylizer import Stylizer
 
@@ -149,7 +150,7 @@ class RBMLizer(object):
 
         # Proccess tags that contain text.
         if hasattr(elem, 'text') and elem.text != None and elem.text.strip() != '':
-            text += elem.text
+            text += prepare_string_for_xml(elem.text)
 
         for item in elem:
             text += self.dump_text(item, stylizer, tag_stack)
@@ -161,7 +162,7 @@ class RBMLizer(object):
         text += self.close_tags(close_tag_list)
 
         if hasattr(elem, 'tail') and elem.tail != None and elem.tail.strip() != '':
-                text += elem.tail
+                text += prepare_string_for_xml(elem.tail)
 
         return text