[3.10] gh-121284: Fix email address header folding with parsed encoded-word (GH-122754) (GH-131411)

encukou · medmunds · bitdancer · web-flow · commit cfaee20d7eaa · 2025-04-03T18:27:20.000+02:00
Email generators using email.policy.default may convert an RFC 2047 encoded-word to unencoded form during header refolding. In a structured header, this could allow 'specials' chars outside a quoted-string, leading to invalid address headers and enabling spoofing. This change ensures a parsed encoded-word that contains specials is kept as an encoded-word while the header is refolded. [Better fix from @bitdancer.] (cherry picked from commit 295b53d) Co-authored-by: Mike Edmunds <medmunds@gmail.com> Co-authored-by: R David Murray <rdmurray@bitdance.com>
diff --git a/Lib/email/_header_value_parser.py b/Lib/email/_header_value_parser.py
@@ -1037,7 +1037,7 @@ def get_fws(value):
     fws = WhiteSpaceTerminal(value[:len(value)-len(newvalue)], 'fws')
     return fws, newvalue
 
-def get_encoded_word(value):
+def get_encoded_word(value, terminal_type='vtext'):
     """ encoded-word = "=?" charset "?" encoding "?" encoded-text "?="
 
     """
@@ -1076,7 +1076,7 @@ def get_encoded_word(value):
             ew.append(token)
             continue
         chars, *remainder = _wsp_splitter(text, 1)
-        vtext = ValueTerminal(chars, 'vtext')
+        vtext = ValueTerminal(chars, terminal_type)
         _validate_xtext(vtext)
         ew.append(vtext)
         text = ''.join(remainder)
@@ -1118,7 +1118,7 @@ def get_unstructured(value):
         valid_ew = True
         if value.startswith('=?'):
             try:
-                token, value = get_encoded_word(value)
+                token, value = get_encoded_word(value, 'utext')
             except _InvalidEwError:
                 valid_ew = False
             except errors.HeaderParseError:
@@ -1147,7 +1147,7 @@ def get_unstructured(value):
         # the parser to go in an infinite loop.
         if valid_ew and rfc2047_matcher.search(tok):
             tok, *remainder = value.partition('=?')
-        vtext = ValueTerminal(tok, 'vtext')
+        vtext = ValueTerminal(tok, 'utext')
         _validate_xtext(vtext)
         unstructured.append(vtext)
         value = ''.join(remainder)
@@ -2781,7 +2781,7 @@ def _refold_parse_tree(parse_tree, *, policy):
             continue
         tstr = str(part)
         if not want_encoding:
-            if part.token_type == 'ptext':
+            if part.token_type in ('ptext', 'vtext'):
                 # Encode if tstr contains special characters.
                 want_encoding = not SPECIALSNL.isdisjoint(tstr)
             else:
diff --git a/Lib/test/test_email/test__header_value_parser.py b/Lib/test/test_email/test__header_value_parser.py
@@ -2946,6 +2946,31 @@ def test_address_list_with_unicode_names_in_quotes(self):
             '=?utf-8?q?H=C3=BCbsch?= Kaktus <beautiful@example.com>,\n'
                 ' =?utf-8?q?bei=C3=9Ft_bei=C3=9Ft?= <biter@example.com>\n')
 
+    def test_address_list_with_specials_in_encoded_word(self):
+        # An encoded-word parsed from a structured header must remain
+        # encoded when it contains specials. Regression for gh-121284.
+        policy = self.policy.clone(max_line_length=40)
+        cases = [
+            # (to, folded)
+            ('=?utf-8?q?A_v=C3=A9ry_long_name_with=2C_comma?= <to@example.com>',
+             'A =?utf-8?q?v=C3=A9ry_long_name_with?=\n'
+             ' =?utf-8?q?=2C?= comma <to@example.com>\n'),
+            ('=?utf-8?q?This_long_name_does_not_need_encoded=2Dword?= <to@example.com>',
+             'This long name does not need\n'
+             ' encoded-word <to@example.com>\n'),
+            ('"A véry long name with, comma" <to@example.com>',
+             # (This isn't the best fold point, but it's not invalid.)
+             'A =?utf-8?q?v=C3=A9ry_long_name_with?=\n'
+             ' =?utf-8?q?=2C?= comma <to@example.com>\n'),
+            ('"A véry long name containing a, comma" <to@example.com>',
+             'A =?utf-8?q?v=C3=A9ry?= long name\n'
+             ' containing =?utf-8?q?a=2C?= comma\n'
+             ' <to@example.com>\n'),
+        ]
+        for (to, folded) in cases:
+            with self.subTest(to=to):
+                self._test(parser.get_address_list(to)[0], folded, policy=policy)
+
     # XXX Need tests with comments on various sides of a unicode token,
     # and with unicode tokens in the comments.  Spaces inside the quotes
     # currently don't do the right thing.
diff --git a/Misc/NEWS.d/next/Security/2024-08-06-12-27-34.gh-issue-121284.8rwPxe.rst b/Misc/NEWS.d/next/Security/2024-08-06-12-27-34.gh-issue-121284.8rwPxe.rst
@@ -0,0 +1,7 @@
+Fix bug in the folding of rfc2047 encoded-words when flattening an email message
+using a modern email policy. Previously when an encoded-word was too long
+for a line, it would be decoded, split across lines, and re-encoded. But commas
+and other special characters in the original text could be left unencoded and
+unquoted. This could theoretically be used to spoof header lines using
+a carefully constructed encoded-word if the resulting rendered email was
+transmitted or re-parsed.
