Sitelet https://github.com/python/cpython/commit/9148b77e0af91cdacaa7fe3dfac09635c3fe9a74
Skip to content

Commit 9148b77

Browse files
miss-islingtont-8ch
miss-islington
and
t-8ch
authored
[3.12] gh-100884: email/_header_value_parser: don't encode list separators (GH-100885) (GH-115592)
ListSeparator should not be encoded. This could happen when a long line pushes its separator to the next line, which would have been encoded. (cherry picked from commit 09fab93) Co-authored-by: Thomas Weißschuh <thomas@t-8ch.de>
1 parent 068e219 commit 9148b77

3 files changed

Lines changed: 9 additions & 1 deletion

File tree

‎Lib/email/_header_value_parser.py‎

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -949,6 +949,7 @@ class _InvalidEwError(errors.HeaderParseError):
949949
# up other parse trees. Maybe should have tests for that, too.
950950
DOT = ValueTerminal('.', 'dot')
951951
ListSeparator = ValueTerminal(',', 'list-separator')
952+
ListSeparator.as_ew_allowed = False
952953
RouteComponentMarker = ValueTerminal('@', 'route-component-marker')
953954

954955
#
@@ -2022,7 +2023,7 @@ def get_address_list(value):
20222023
address_list.defects.append(errors.InvalidHeaderDefect(
20232024
"invalid address in address-list"))
20242025
if value: # Must be a , at this point.
2025-
address_list.append(ValueTerminal(',', 'list-separator'))
2026+
address_list.append(ListSeparator)
20262027
value = value[1:]
20272028
return address_list, value
20282029

‎Lib/test/test_email/test__header_value_parser.py‎

Lines changed: 5 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -2985,6 +2985,11 @@ def test_address_list_with_unicode_names_in_quotes(self):
29852985
'=?utf-8?q?H=C3=BCbsch?= Kaktus <beautiful@example.com>,\n'
29862986
' =?utf-8?q?bei=C3=9Ft_bei=C3=9Ft?= <biter@example.com>\n')
29872987

2988+
def test_address_list_with_list_separator_after_fold(self):
2989+
to = '0123456789' * 8 + '@foo, ä <foo@bar>'
2990+
self._test(parser.get_address_list(to)[0],
2991+
'0123456789' * 8 + '@foo,\n =?utf-8?q?=C3=A4?= <foo@bar>\n')
2992+
29882993
# XXX Need tests with comments on various sides of a unicode token,
29892994
# and with unicode tokens in the comments. Spaces inside the quotes
29902995
# currently don't do the right thing.

‎Misc/NEWS.d/next/Library/2023-01-09-14-08-02.gh-issue-100884.DcmdLl.rst‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,2 @@
1+
email: fix misfolding of comma in address-lists over multiple lines in
2+
combination with unicode encoding.

0 commit comments

Comments
 (0)