Merge: #17369: Improve handling of broken RFC2231 values in get_filename.

Lib/email/utils.py

@@ -347,6 +347,10 @@ def collapse_rfc2231_value(value, errors='replace',
     # object.  We do not want bytes() normal utf-8 decoder, we want a straight
     # interpretation of the string as character bytes.
     charset, language, text = value
+    if charset is None:
+        # Issue 17369: if charset/lang is None, decode_rfc2231 couldn't parse
+        # the value, so use the fallback_charset.
+        charset = fallback_charset
     rawbytes = bytes(text, 'raw-unicode-escape')
     try:
         return str(rawbytes, charset, errors)

Lib/test/test_email/test_email.py

@@ -5052,6 +5052,26 @@ Content-Type: application/x-foo; name*0=\"Frank's\"; name*1=\" Document\"
         self.assertNotIsInstance(param, tuple)
         self.assertEqual(param, "Frank's Document")
 
+    def test_rfc2231_missing_tick(self):
+        m = '''\
+Content-Disposition: inline;
+\tfilename*0*="'This%20is%20broken";
+'''
+        msg = email.message_from_string(m)
+        self.assertEqual(
+            msg.get_filename(),
+            "'This is broken")
+
+    def test_rfc2231_missing_tick_with_encoded_non_ascii(self):
+        m = '''\
+Content-Disposition: inline;
+\tfilename*0*="'This%20is%E2broken";
+'''
+        msg = email.message_from_string(m)
+        self.assertEqual(
+            msg.get_filename(),
+            "'This is\ufffdbroken")
+
     # test_headerregistry.TestContentTypeHeader.rfc2231_single_quote_in_value_with_charset_and_lang
     def test_rfc2231_tick_attack_extended(self):
         eq = self.assertEqual

Misc/NEWS

@@ -24,6 +24,10 @@ Core and Builtins
 Library
 -------
 
+- Issue #17369: get_filename was raising an exception if the filename
+  parameter's RFC2231 encoding was broken in certain ways.  This was
+  a regression relative to python2.
+
 - Issue #20013: Some imap servers disconnect if the current mailbox is
   deleted, and imaplib did not handle that case gracefully.  Now it
   handles the 'bye' correctly.