From 125f80888e70711c2f779a64465b22e697cfc8ad Mon Sep 17 00:00:00 2001
From: Christophe Matthieu <chm@openerp.com>
Date: Wed, 19 Dec 2012 11:37:14 +0100
Subject: [PATCH 1/4] [FIX] tools mail: return the initial variable if is false
 or if is not an instance of basestring. remove xml and doctype tags.

bzr revid: chm@openerp.com-20121219103714-u52hfmqr5u508xa2
---
 openerp/tools/mail.py | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/openerp/tools/mail.py b/openerp/tools/mail.py
index a54f99ba1ad..9df7499d0c8 100644
--- a/openerp/tools/mail.py
+++ b/openerp/tools/mail.py
@@ -130,11 +130,15 @@ def html_email_clean(html):
         dest += source[idx:]
         return dest
 
-    if not html:
+    if not html or not isinstance(html, basestring):
         return html
 
     html = ustr(html)
 
+    # 0. remove doctype and xml tags
+    doctype = re.compile(r"(<[?]xml.*?[?]>|<[!]DOCTYPE.*?>)[\n\r\s]*", re.IGNORECASE | re.DOTALL)
+    html = doctype.sub(r"", html)
+
     # 1. <br[ /]> -> \n, because otherwise the tree is obfuscated
     br_tags = re.compile(r'([<]\s*[bB][rR]\s*\/?[>])')
     html = _replace_matching_regex(br_tags, html, '__BR_TAG__')

From a31af13a3ed672ef80b99b9ae622064a7291d350 Mon Sep 17 00:00:00 2001
From: Christophe Matthieu <chm@openerp.com>
Date: Wed, 19 Dec 2012 12:18:53 +0100
Subject: [PATCH 2/4] [IMP] tools mail: add test message with xml and doctype
 tags don't crash

bzr revid: chm@openerp.com-20121219111853-zj16hdg7ujakksqt
---
 openerp/tests/test_mail.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/openerp/tests/test_mail.py b/openerp/tests/test_mail.py
index a8c806a13d0..fbf2caa2b62 100644
--- a/openerp/tests/test_mail.py
+++ b/openerp/tests/test_mail.py
@@ -177,6 +177,9 @@ class TestCleaner(unittest2.TestCase):
         new_html = html_email_clean(False)
         self.assertEqual(new_html, False, 'html_email_cleaner did change a False in an other value.')
 
+        # Test6: Message with xml and doctype tags don't crash
+        html_email_clean(u'<?xml version="1.0" encoding="iso-8859-1"?>\n<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"\n         "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">\n<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">\n <head>\n  <title>404 - Not Found</title>\n </head>\n <body>\n  <h1>404 - Not Found</h1>\n </body>\n</html>\n')
+
 class TestHtmlTools(unittest2.TestCase):
     """ Test some of our generic utility functions about html """
 

From 111e080e48682c1e8fb98c061b4acf07e2852b57 Mon Sep 17 00:00:00 2001
From: Christophe Matthieu <chm@openerp.com>
Date: Wed, 19 Dec 2012 13:40:38 +0100
Subject: [PATCH 3/4] [IMP] tools mail: don't remove xml and doctype but remove
 encoding attribute

bzr revid: chm@openerp.com-20121219124038-05bf07960h05x0o8
---
 openerp/tools/mail.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/openerp/tools/mail.py b/openerp/tools/mail.py
index 9df7499d0c8..e6b471a26ba 100644
--- a/openerp/tools/mail.py
+++ b/openerp/tools/mail.py
@@ -135,8 +135,8 @@ def html_email_clean(html):
 
     html = ustr(html)
 
-    # 0. remove doctype and xml tags
-    doctype = re.compile(r"(<[?]xml.*?[?]>|<[!]DOCTYPE.*?>)[\n\r\s]*", re.IGNORECASE | re.DOTALL)
+    # 0. remove encoding attribute inside tags
+    doctype = re.compile(r'(<[^>]*\s)(encoding=(["\'][^"\']*?["\']|[^\s\n\r>]+)(\s[^>]*|/)?>)', re.IGNORECASE | re.DOTALL)
     html = doctype.sub(r"", html)
 
     # 1. <br[ /]> -> \n, because otherwise the tree is obfuscated

From bf115734369b185cd3d979245de65276fd885bcb Mon Sep 17 00:00:00 2001
From: Christophe Matthieu <chm@openerp.com>
Date: Wed, 19 Dec 2012 15:39:12 +0100
Subject: [PATCH 4/4] [IMP] tools mail: encoding : add test for removing
 encoding attribute

bzr revid: chm@openerp.com-20121219143912-os160iv0q0jeyhld
---
 openerp/tests/test_mail.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/openerp/tests/test_mail.py b/openerp/tests/test_mail.py
index fbf2caa2b62..b3c11a13bfe 100644
--- a/openerp/tests/test_mail.py
+++ b/openerp/tests/test_mail.py
@@ -178,7 +178,8 @@ class TestCleaner(unittest2.TestCase):
         self.assertEqual(new_html, False, 'html_email_cleaner did change a False in an other value.')
 
         # Test6: Message with xml and doctype tags don't crash
-        html_email_clean(u'<?xml version="1.0" encoding="iso-8859-1"?>\n<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"\n         "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">\n<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">\n <head>\n  <title>404 - Not Found</title>\n </head>\n <body>\n  <h1>404 - Not Found</h1>\n </body>\n</html>\n')
+        new_html = html_email_clean(u'<?xml version="1.0" encoding="iso-8859-1"?>\n<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"\n         "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">\n<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">\n <head>\n  <title>404 - Not Found</title>\n </head>\n <body>\n  <h1>404 - Not Found</h1>\n </body>\n</html>\n')
+        self.assertNotIn('encoding', new_html, 'html_email_cleaner did not remove correctly encoding attributes')
 
 class TestHtmlTools(unittest2.TestCase):
     """ Test some of our generic utility functions about html """