From: R. David Murray <rdmurray@bitdance.com>
Date: Wed, 4 Aug 2010 00:05:50 +0000 (+0000)
Subject: Merged revisions 83690 via svnmerge from
X-Git-Tag: v2.7.1rc1~469
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=75a292e5be429977656acaf846345be9dc0a6b8e;p=python

Merged revisions 83690 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/branches/py3k

........
  r83690 | r.david.murray | 2010-08-03 18:14:10 -0400 (Tue, 03 Aug 2010) | 10 lines

  #3196: if needed pad a short base64 encoded word before trying to decode.

  The RFCs encourage following Postel's law: be liberal in what you accept.
  So if someone forgot to pad the base64 encoded word payload to an
  even four bytes, we add the padding before handing it to base64mime.decode.
  Previously, missing padding resulted in a HeaderParseError.

  Patch by Jason Williams.
........
---

diff --git a/Lib/email/header.py b/Lib/email/header.py
index 6e03922aa0..9e91dc2077 100644
--- a/Lib/email/header.py
+++ b/Lib/email/header.py
@@ -92,6 +92,9 @@ def decode_header(header):
                 if encoding == 'q':
                     dec = email.quoprimime.header_decode(encoded)
                 elif encoding == 'b':
+                    paderr = len(encoded) % 4   # Postel's law: add missing padding
+                    if paderr:
+                        encoded += '==='[:4 - paderr]
                     try:
                         dec = email.base64mime.decode(encoded)
                     except binascii.Error:
diff --git a/Lib/email/test/test_email.py b/Lib/email/test/test_email.py
index 067937a975..e1754e9f75 100644
--- a/Lib/email/test/test_email.py
+++ b/Lib/email/test/test_email.py
@@ -1611,6 +1611,15 @@ class TestRFC2047(unittest.TestCase):
                               ('rg', None), ('\xe5', 'iso-8859-1'),
                               ('sbord', None)])
 
+    def test_rfc2047_B_bad_padding(self):
+        s = '=?iso-8859-1?B?%s?='
+        data = [                                # only test complete bytes
+            ('dm==', 'v'), ('dm=', 'v'), ('dm', 'v'),
+            ('dmk=', 'vi'), ('dmk', 'vi')
+          ]
+        for q, a in data:
+            dh = decode_header(s % q)
+            self.assertEqual(dh, [(a, 'iso-8859-1')])
 
 
 # Test the MIMEMessage class
@@ -3064,7 +3073,7 @@ A very long line that must get split to something other than at the
 
     def test_broken_base64_header(self):
         raises = self.assertRaises
-        s = 'Subject: =?EUC-KR?B?CSixpLDtKSC/7Liuvsax4iC6uLmwMcijIKHaILzSwd/H0SC8+LCjwLsgv7W/+Mj3IQ?='
+        s = 'Subject: =?EUC-KR?B?CSixpLDtKSC/7Liuvsax4iC6uLmwMcijIKHaILzSwd/H0SC8+LCjwLsgv7W/+Mj3I ?='
         raises(Errors.HeaderParseError, decode_header, s)
 
 
diff --git a/Lib/email/test/test_email_renamed.py b/Lib/email/test/test_email_renamed.py
index 976d892f4a..481c40f135 100644
--- a/Lib/email/test/test_email_renamed.py
+++ b/Lib/email/test/test_email_renamed.py
@@ -2958,7 +2958,7 @@ A very long line that must get split to something other than at the
 
     def test_broken_base64_header(self):
         raises = self.assertRaises
-        s = 'Subject: =?EUC-KR?B?CSixpLDtKSC/7Liuvsax4iC6uLmwMcijIKHaILzSwd/H0SC8+LCjwLsgv7W/+Mj3IQ?='
+        s = 'Subject: =?EUC-KR?B?CSixpLDtKSC/7Liuvsax4iC6uLmwMcijIKHaILzSwd/H0SC8+LCjwLsgv7W/+Mj3I ?='
         raises(errors.HeaderParseError, decode_header, s)
 
 
diff --git a/Misc/ACKS b/Misc/ACKS
index 197a5022ca..089939ee1e 100644
--- a/Misc/ACKS
+++ b/Misc/ACKS
@@ -831,6 +831,7 @@ Felix Wiemann
 Gerry Wiener
 Frank Wierzbicki
 Bryce "Zooko" Wilcox-O'Hearn
+Jason Williams
 John Williams
 Sue Williams
 Gerald S. Williams
diff --git a/Misc/NEWS b/Misc/NEWS
index d8b079bdb5..545ef74f21 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -24,6 +24,9 @@ Core and Builtins
 Library
 -------
 
+- Issue #3196: email header decoding is now forgiving if an RFC2047
+  encoded word encoded in base64 is lacking padding.
+
 - Issue #9444: Argparse now uses the first element of prefix_chars as
   the option character for the added 'h/help' option if prefix_chars
   does not contain a '-', instead of raising an error.