From 5b2d9ddf69cecfb9ad4e687fab3f34ecc5a9ea4f Mon Sep 17 00:00:00 2001
From: "R. David Murray" <rdmurray@bitdance.com>
Date: Sun, 9 Jan 2011 02:35:24 +0000
Subject: #5871: protect against header injection attacks.

This makes Header.encode throw a HeaderParseError if it winds up
formatting a header such that a continuation line has no leading
whitespace and looks like a header.  Since Header accepts values
containing newlines and preserves them (and this is by design), without
this fix any program that took user input (say, a subject in a web form)
and passed it to the email package as a header was vulnerable to header
injection attacks.  (As far as we know this has never been exploited.)

Thanks to Jakub Wilk for reporting this vulnerability.
---
 Lib/email/header.py          | 10 +++++++++-
 Lib/email/test/test_email.py | 12 ++++++++++++
 Misc/NEWS                    |  7 +++++++
 3 files changed, 28 insertions(+), 1 deletion(-)

diff --git a/Lib/email/header.py b/Lib/email/header.py
index f90883f..35cdb2b 100644
--- a/Lib/email/header.py
+++ b/Lib/email/header.py
@@ -47,6 +47,10 @@ ecre = re.compile(r'''
 # For use with .match()
 fcre = re.compile(r'[\041-\176]+:$')
 
+# Find a header embeded in a putative header value.  Used to check for
+# header injection attack.
+_embeded_header = re.compile(r'\n[^ \t]+:')
+
 
 
 # Helpers
@@ -320,7 +324,11 @@ class Header:
             if len(lines) > 1:
                 formatter.newline()
             formatter.add_transition()
-        return formatter._str(linesep)
+        value = formatter._str(linesep)
+        if _embeded_header.search(value):
+            raise HeaderParseError("header value appears to contain "
+                "an embedded header: {!r}".format(value))
+        return value
 
     def _normalize(self):
         # Step 1: Normalize the chunks so that all runs of identical charsets
diff --git a/Lib/email/test/test_email.py b/Lib/email/test/test_email.py
index a1798ce..e4083ad 100644
--- a/Lib/email/test/test_email.py
+++ b/Lib/email/test/test_email.py
@@ -561,6 +561,18 @@ class TestMessageAPI(TestEmailBase):
             "attachment; filename*=utf-8''Fu%C3%9Fballer%20%5Bfilename%5D.ppt",
             msg['Content-Disposition'])
 
+    # Issue 5871: reject an attempt to embed a header inside a header value
+    # (header injection attack).
+    def test_embeded_header_via_Header_rejected(self):
+        msg = Message()
+        msg['Dummy'] = Header('dummy\nX-Injected-Header: test')
+        self.assertRaises(errors.HeaderParseError, msg.as_string)
+
+    def test_embeded_header_via_string_rejected(self):
+        msg = Message()
+        msg['Dummy'] = 'dummy\nX-Injected-Header: test'
+        self.assertRaises(errors.HeaderParseError, msg.as_string)
+
 
 # Test the email.encoders module
 class TestEncoders(unittest.TestCase):
diff --git a/Misc/NEWS b/Misc/NEWS
index c1d4260..2f8b6db 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -40,6 +40,13 @@ Core and Builtins
 Library
 -------
 
+- Issue #5871: email.header.Header.encode now raises an error if any
+  continuation line in the formatted value has no leading white space
+  and looks like a header.  Since Generator uses Header to format all
+  headers, this check is made for all headers in any serialized message
+  at serialization time.  This provides protection against header
+  injection attacks.
+
 - Issue #10859: Make ``contextlib.GeneratorContextManager`` officially
   private by renaming it to ``_GeneratorContextManager``.
 
-- 
cgit v0.12