More RFC 2231 improvements for the email 4.0 package. As Mark Sapiro rightly

points out there are really two types of continued headers defined in this
RFC (i.e. "encoded" parameters with the form "name*0*=" and unencoded
parameters with the form "name*0="), but we were were handling them both the
same way and that isn't correct.

This patch should be much more RFC compliant in that only encoded params are
%-decoded and the charset/language information is only extract if there are
any encoded params in the segments.  If there are no encoded params then the
RFC says that there will be no charset/language parts.

Note however that this will change the return value for Message.get_param() in
some cases.  For example, whereas before if you had all unencoded param
continuations you would have still gotten a 3-tuple back from this method
(with charset and language == None), you will now get just a string.  I don't
believe this is a backward incompatible change though because the
documentation for this method already indicates that either return value is
possible and that you must do an isinstance(val, tuple) check to discriminate
between the two.  (Yeah that API kind of sucks but we can't change /that/
without breaking code.)

Test cases, some documentation updates, and a NEWS item accompany this patch.

3 files changed, 277 insertions, 41 deletions

diff --git a/Lib/email/test/test_email.py b/Lib/email/test/test_email.py
index a197a36..db0c2be 100644
--- a/Lib/email/test/test_email.py
+++ b/Lib/email/test/test_email.py
@@ -3005,14 +3005,29 @@ Content-Type: text/html; NAME*0=file____C__DOCUMENTS_20AND_20SETTINGS_FABIEN_LOC
 
 '''
         msg = email.message_from_string(m)
-        self.assertEqual(msg.get_param('NAME'),
-                         (None, None, 'file____C__DOCUMENTS_20AND_20SETTINGS_FABIEN_LOCAL_20SETTINGS_TEMP_nsmail.htm'))
+        param = msg.get_param('NAME')
+        self.failIf(isinstance(param, tuple))
+        self.assertEqual(
+            param,
+            'file____C__DOCUMENTS_20AND_20SETTINGS_FABIEN_LOCAL_20SETTINGS_TEMP_nsmail.htm')
 
     def test_rfc2231_no_language_or_charset_in_filename(self):
         m = '''\
 Content-Disposition: inline;
-\tfilename*0="This%20is%20even%20more%20";
-\tfilename*1="%2A%2A%2Afun%2A%2A%2A%20";
+\tfilename*0*="''This%20is%20even%20more%20";
+\tfilename*1*="%2A%2A%2Afun%2A%2A%2A%20";
+\tfilename*2="is it not.pdf"
+
+'''
+        msg = email.message_from_string(m)
+        self.assertEqual(msg.get_filename(),
+                         'This is even more ***fun*** is it not.pdf')
+
+    def test_rfc2231_no_language_or_charset_in_filename_encoded(self):
+        m = '''\
+Content-Disposition: inline;
+\tfilename*0*="''This%20is%20even%20more%20";
+\tfilename*1*="%2A%2A%2Afun%2A%2A%2A%20";
 \tfilename*2="is it not.pdf"
 
 '''
@@ -3020,11 +3035,37 @@ Content-Disposition: inline;
         self.assertEqual(msg.get_filename(),
                          'This is even more ***fun*** is it not.pdf')
 
+    def test_rfc2231_partly_encoded(self):
+        m = '''\
+Content-Disposition: inline;
+\tfilename*0="''This%20is%20even%20more%20";
+\tfilename*1*="%2A%2A%2Afun%2A%2A%2A%20";
+\tfilename*2="is it not.pdf"
+
+'''
+        msg = email.message_from_string(m)
+        self.assertEqual(
+            msg.get_filename(),
+            'This%20is%20even%20more%20***fun*** is it not.pdf')
+
+    def test_rfc2231_partly_nonencoded(self):
+        m = '''\
+Content-Disposition: inline;
+\tfilename*0="This%20is%20even%20more%20";
+\tfilename*1="%2A%2A%2Afun%2A%2A%2A%20";
+\tfilename*2="is it not.pdf"
+
+'''
+        msg = email.message_from_string(m)
+        self.assertEqual(
+            msg.get_filename(),
+            'This%20is%20even%20more%20%2A%2A%2Afun%2A%2A%2A%20is it not.pdf')
+
     def test_rfc2231_no_language_or_charset_in_boundary(self):
         m = '''\
 Content-Type: multipart/alternative;
-\tboundary*0="This%20is%20even%20more%20";
-\tboundary*1="%2A%2A%2Afun%2A%2A%2A%20";
+\tboundary*0*="''This%20is%20even%20more%20";
+\tboundary*1*="%2A%2A%2Afun%2A%2A%2A%20";
 \tboundary*2="is it not.pdf"
 
 '''
@@ -3036,8 +3077,8 @@ Content-Type: multipart/alternative;
         # This is a nonsensical charset value, but tests the code anyway
         m = '''\
 Content-Type: text/plain;
-\tcharset*0="This%20is%20even%20more%20";
-\tcharset*1="%2A%2A%2Afun%2A%2A%2A%20";
+\tcharset*0*="This%20is%20even%20more%20";
+\tcharset*1*="%2A%2A%2Afun%2A%2A%2A%20";
 \tcharset*2="is it not.pdf"
 
 '''
@@ -3048,12 +3089,98 @@ Content-Type: text/plain;
     def test_rfc2231_unknown_encoding(self):
         m = """\
 Content-Transfer-Encoding: 8bit
-Content-Disposition: inline; filename*0=X-UNKNOWN''myfile.txt
+Content-Disposition: inline; filename*=X-UNKNOWN''myfile.txt
 
 """
         msg = email.message_from_string(m)
         self.assertEqual(msg.get_filename(), 'myfile.txt')
 
+    def test_rfc2231_single_tick_in_filename_extended(self):
+        eq = self.assertEqual
+        m = """\
+Content-Type: application/x-foo;
+\tname*0*=\"Frank's\"; name*1*=\" Document\"
+
+"""
+        msg = email.message_from_string(m)
+        charset, language, s = msg.get_param('name')
+        eq(charset, None)
+        eq(language, None)
+        eq(s, "Frank's Document")
+
+    def test_rfc2231_single_tick_in_filename(self):
+        m = """\
+Content-Type: application/x-foo; name*0=\"Frank's\"; name*1=\" Document\"
+
+"""
+        msg = email.message_from_string(m)
+        param = msg.get_param('name')
+        self.failIf(isinstance(param, tuple))
+        self.assertEqual(param, "Frank's Document")
+
+    def test_rfc2231_tick_attack_extended(self):
+        eq = self.assertEqual
+        m = """\
+Content-Type: application/x-foo;
+\tname*0*=\"us-ascii'en-us'Frank's\"; name*1*=\" Document\"
+
+"""
+        msg = email.message_from_string(m)
+        charset, language, s = msg.get_param('name')
+        eq(charset, 'us-ascii')
+        eq(language, 'en-us')
+        eq(s, "Frank's Document")
+
+    def test_rfc2231_tick_attack(self):
+        m = """\
+Content-Type: application/x-foo;
+\tname*0=\"us-ascii'en-us'Frank's\"; name*1=\" Document\"
+
+"""
+        msg = email.message_from_string(m)
+        param = msg.get_param('name')
+        self.failIf(isinstance(param, tuple))
+        self.assertEqual(param, "us-ascii'en-us'Frank's Document")
+
+    def test_rfc2231_no_extended_values(self):
+        eq = self.assertEqual
+        m = """\
+Content-Type: application/x-foo; name=\"Frank's Document\"
+
+"""
+        msg = email.message_from_string(m)
+        eq(msg.get_param('name'), "Frank's Document")
+
+    def test_rfc2231_encoded_then_unencoded_segments(self):
+        eq = self.assertEqual
+        m = """\
+Content-Type: application/x-foo;
+\tname*0*=\"us-ascii'en-us'My\";
+\tname*1=\" Document\";
+\tname*2*=\" For You\"
+
+"""
+        msg = email.message_from_string(m)
+        charset, language, s = msg.get_param('name')
+        eq(charset, 'us-ascii')
+        eq(language, 'en-us')
+        eq(s, 'My Document For You')
+
+    def test_rfc2231_unencoded_then_encoded_segments(self):
+        eq = self.assertEqual
+        m = """\
+Content-Type: application/x-foo;
+\tname*0=\"us-ascii'en-us'My\";
+\tname*1*=\" Document\";
+\tname*2*=\" For You\"
+
+"""
+        msg = email.message_from_string(m)
+        charset, language, s = msg.get_param('name')
+        eq(charset, 'us-ascii')
+        eq(language, 'en-us')
+        eq(s, 'My Document For You')
+
 
 
 def _testclasses():
diff --git a/Lib/email/test/test_email_renamed.py b/Lib/email/test/test_email_renamed.py
index 4cfca66..680a725 100644
--- a/Lib/email/test/test_email_renamed.py
+++ b/Lib/email/test/test_email_renamed.py
@@ -3011,14 +3011,29 @@ Content-Type: text/html; NAME*0=file____C__DOCUMENTS_20AND_20SETTINGS_FABIEN_LOC
 
 '''
         msg = email.message_from_string(m)
-        self.assertEqual(msg.get_param('NAME'),
-                         (None, None, 'file____C__DOCUMENTS_20AND_20SETTINGS_FABIEN_LOCAL_20SETTINGS_TEMP_nsmail.htm'))
+        param = msg.get_param('NAME')
+        self.failIf(isinstance(param, tuple))
+        self.assertEqual(
+            param,
+            'file____C__DOCUMENTS_20AND_20SETTINGS_FABIEN_LOCAL_20SETTINGS_TEMP_nsmail.htm')
 
     def test_rfc2231_no_language_or_charset_in_filename(self):
         m = '''\
 Content-Disposition: inline;
-\tfilename*0="This%20is%20even%20more%20";
-\tfilename*1="%2A%2A%2Afun%2A%2A%2A%20";
+\tfilename*0*="''This%20is%20even%20more%20";
+\tfilename*1*="%2A%2A%2Afun%2A%2A%2A%20";
+\tfilename*2="is it not.pdf"
+
+'''
+        msg = email.message_from_string(m)
+        self.assertEqual(msg.get_filename(),
+                         'This is even more ***fun*** is it not.pdf')
+
+    def test_rfc2231_no_language_or_charset_in_filename_encoded(self):
+        m = '''\
+Content-Disposition: inline;
+\tfilename*0*="''This%20is%20even%20more%20";
+\tfilename*1*="%2A%2A%2Afun%2A%2A%2A%20";
 \tfilename*2="is it not.pdf"
 
 '''
@@ -3026,11 +3041,37 @@ Content-Disposition: inline;
         self.assertEqual(msg.get_filename(),
                          'This is even more ***fun*** is it not.pdf')
 
+    def test_rfc2231_partly_encoded(self):
+        m = '''\
+Content-Disposition: inline;
+\tfilename*0="''This%20is%20even%20more%20";
+\tfilename*1*="%2A%2A%2Afun%2A%2A%2A%20";
+\tfilename*2="is it not.pdf"
+
+'''
+        msg = email.message_from_string(m)
+        self.assertEqual(
+            msg.get_filename(),
+            'This%20is%20even%20more%20***fun*** is it not.pdf')
+
+    def test_rfc2231_partly_nonencoded(self):
+        m = '''\
+Content-Disposition: inline;
+\tfilename*0="This%20is%20even%20more%20";
+\tfilename*1="%2A%2A%2Afun%2A%2A%2A%20";
+\tfilename*2="is it not.pdf"
+
+'''
+        msg = email.message_from_string(m)
+        self.assertEqual(
+            msg.get_filename(),
+            'This%20is%20even%20more%20%2A%2A%2Afun%2A%2A%2A%20is it not.pdf')
+
     def test_rfc2231_no_language_or_charset_in_boundary(self):
         m = '''\
 Content-Type: multipart/alternative;
-\tboundary*0="This%20is%20even%20more%20";
-\tboundary*1="%2A%2A%2Afun%2A%2A%2A%20";
+\tboundary*0*="''This%20is%20even%20more%20";
+\tboundary*1*="%2A%2A%2Afun%2A%2A%2A%20";
 \tboundary*2="is it not.pdf"
 
 '''
@@ -3042,8 +3083,8 @@ Content-Type: multipart/alternative;
         # This is a nonsensical charset value, but tests the code anyway
         m = '''\
 Content-Type: text/plain;
-\tcharset*0="This%20is%20even%20more%20";
-\tcharset*1="%2A%2A%2Afun%2A%2A%2A%20";
+\tcharset*0*="This%20is%20even%20more%20";
+\tcharset*1*="%2A%2A%2Afun%2A%2A%2A%20";
 \tcharset*2="is it not.pdf"
 
 '''
@@ -3054,16 +3095,17 @@ Content-Type: text/plain;
     def test_rfc2231_unknown_encoding(self):
         m = """\
 Content-Transfer-Encoding: 8bit
-Content-Disposition: inline; filename*0=X-UNKNOWN''myfile.txt
+Content-Disposition: inline; filename*=X-UNKNOWN''myfile.txt
 
 """
         msg = email.message_from_string(m)
         self.assertEqual(msg.get_filename(), 'myfile.txt')
 
-    def test_rfc2231_single_tick_in_filename(self):
+    def test_rfc2231_single_tick_in_filename_extended(self):
         eq = self.assertEqual
         m = """\
-Content-Type: application/x-foo; name*0=\"Frank's\"; name*1=\" Document\"
+Content-Type: application/x-foo;
+\tname*0*=\"Frank's\"; name*1*=\" Document\"
 
 """
         msg = email.message_from_string(m)
@@ -3072,11 +3114,21 @@ Content-Type: application/x-foo; name*0=\"Frank's\"; name*1=\" Document\"
         eq(language, None)
         eq(s, "Frank's Document")
 
-    def test_rfc2231_tick_attack(self):
+    def test_rfc2231_single_tick_in_filename(self):
+        m = """\
+Content-Type: application/x-foo; name*0=\"Frank's\"; name*1=\" Document\"
+
+"""
+        msg = email.message_from_string(m)
+        param = msg.get_param('name')
+        self.failIf(isinstance(param, tuple))
+        self.assertEqual(param, "Frank's Document")
+
+    def test_rfc2231_tick_attack_extended(self):
         eq = self.assertEqual
         m = """\
 Content-Type: application/x-foo;
-\tname*0=\"us-ascii'en-us'Frank's\"; name*1=\" Document\"
+\tname*0*=\"us-ascii'en-us'Frank's\"; name*1*=\" Document\"
 
 """
         msg = email.message_from_string(m)
@@ -3085,6 +3137,17 @@ Content-Type: application/x-foo;
         eq(language, 'en-us')
         eq(s, "Frank's Document")
 
+    def test_rfc2231_tick_attack(self):
+        m = """\
+Content-Type: application/x-foo;
+\tname*0=\"us-ascii'en-us'Frank's\"; name*1=\" Document\"
+
+"""
+        msg = email.message_from_string(m)
+        param = msg.get_param('name')
+        self.failIf(isinstance(param, tuple))
+        self.assertEqual(param, "us-ascii'en-us'Frank's Document")
+
     def test_rfc2231_no_extended_values(self):
         eq = self.assertEqual
         m = """\
@@ -3094,6 +3157,36 @@ Content-Type: application/x-foo; name=\"Frank's Document\"
         msg = email.message_from_string(m)
         eq(msg.get_param('name'), "Frank's Document")
 
+    def test_rfc2231_encoded_then_unencoded_segments(self):
+        eq = self.assertEqual
+        m = """\
+Content-Type: application/x-foo;
+\tname*0*=\"us-ascii'en-us'My\";
+\tname*1=\" Document\";
+\tname*2*=\" For You\"
+
+"""
+        msg = email.message_from_string(m)
+        charset, language, s = msg.get_param('name')
+        eq(charset, 'us-ascii')
+        eq(language, 'en-us')
+        eq(s, 'My Document For You')
+
+    def test_rfc2231_unencoded_then_encoded_segments(self):
+        eq = self.assertEqual
+        m = """\
+Content-Type: application/x-foo;
+\tname*0=\"us-ascii'en-us'My\";
+\tname*1*=\" Document\";
+\tname*2*=\" For You\"
+
+"""
+        msg = email.message_from_string(m)
+        charset, language, s = msg.get_param('name')
+        eq(charset, 'us-ascii')
+        eq(language, 'en-us')
+        eq(s, 'My Document For You')
+
 
 
 def _testclasses():
diff --git a/Lib/email/utils.py b/Lib/email/utils.py
index ea59c27..26ebb0e 100644
--- a/Lib/email/utils.py
+++ b/Lib/email/utils.py
@@ -25,6 +25,7 @@ import time
 import base64
 import random
 import socket
+import urllib
 import warnings
 from cStringIO import StringIO
 
@@ -231,16 +232,14 @@ def unquote(str):
 # RFC2231-related functions - parameter encoding and decoding
 def decode_rfc2231(s):
     """Decode string according to RFC 2231"""
-    import urllib
     parts = s.split(TICK, 2)
     if len(parts) <= 2:
-        return None, None, urllib.unquote(s)
+        return None, None, s
     if len(parts) > 3:
         charset, language = parts[:2]
         s = TICK.join(parts[2:])
-    else:
-        charset, language, s = parts
-    return charset, language, urllib.unquote(s)
+        return charset, language, s
+    return parts
 
 
 def encode_rfc2231(s, charset=None, language=None):
@@ -264,37 +263,54 @@ rfc2231_continuation = re.compile(r'^(?P<name>\w+)\*((?P<num>[0-9]+)\*?)?$')
 def decode_params(params):
     """Decode parameters list according to RFC 2231.
 
-    params is a sequence of 2-tuples containing (content type, string value).
+    params is a sequence of 2-tuples containing (param name, string value).
     """
+    # Copy params so we don't mess with the original
+    params = params[:]
     new_params = []
-    # maps parameter's name to a list of continuations
+    # Map parameter's name to a list of continuations.  The values are a
+    # 3-tuple of the continuation number, the string value, and a flag
+    # specifying whether a particular segment is %-encoded.
     rfc2231_params = {}
-    # params is a sequence of 2-tuples containing (content_type, string value)
-    name, value = params[0]
+    name, value = params.pop(0)
     new_params.append((name, value))
-    # Cycle through each of the rest of the parameters.
-    for name, value in params[1:]:
+    while params:
+        name, value = params.pop(0)
+        if name.endswith('*'):
+            encoded = True
+        else:
+            encoded = False
         value = unquote(value)
         mo = rfc2231_continuation.match(name)
         if mo:
             name, num = mo.group('name', 'num')
             if num is not None:
                 num = int(num)
-            rfc2231_param1 = rfc2231_params.setdefault(name, [])
-            rfc2231_param1.append((num, value))
+            rfc2231_params.setdefault(name, []).append((num, value, encoded))
         else:
             new_params.append((name, '"%s"' % quote(value)))
     if rfc2231_params:
         for name, continuations in rfc2231_params.items():
             value = []
+            extended = False
             # Sort by number
             continuations.sort()
-            # And now append all values in num order
-            for num, continuation in continuations:
-                value.append(continuation)
-            charset, language, value = decode_rfc2231(EMPTYSTRING.join(value))
-            new_params.append(
-                (name, (charset, language, '"%s"' % quote(value))))
+            # And now append all values in numerical order, converting
+            # %-encodings for the encoded segments.  If any of the
+            # continuation names ends in a *, then the entire string, after
+            # decoding segments and concatenating, must have the charset and
+            # language specifiers at the beginning of the string.
+            for num, s, encoded in continuations:
+                if encoded:
+                    s = urllib.unquote(s)
+                    extended = True
+                value.append(s)
+            value = quote(EMPTYSTRING.join(value))
+            if extended:
+                charset, language, value = decode_rfc2231(value)
+                new_params.append((name, (charset, language, '"%s"' % value)))
+            else:
+                new_params.append((name, '"%s"' % value))
     return new_params
 
 def collapse_rfc2231_value(value, errors='replace',