Collin Funk <[email protected]> writes:

> Using it with that removed, i.e.,
> '\xC3|\xED\xBA\xAD|\u0089|\xED\xA6\xBF\xED\xBF\xBF\n', doesn't seem to
> work though, so I'll have to look into that.

Oops, there wasn't actually an issue. I just didn't see the newline I
copied over for some reason.

Pushed the attached patch.

Collin

>From d1d311e7a244b908833fdae2f84fb5a613c055c7 Mon Sep 17 00:00:00 2001
Message-ID: <d1d311e7a244b908833fdae2f84fb5a613c055c7.1764395375.git.collin.fu...@gmail.com>
From: Collin Funk <[email protected]>
Date: Fri, 28 Nov 2025 21:28:21 -0800
Subject: [PATCH] tests: tac: test --separator with invalid UTF-8

* tests/tac/tac-locale.sh: Test that a string containing invalid UTF-8
characters can be used for --separator.
---
 tests/tac/tac-locale.sh | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tests/tac/tac-locale.sh b/tests/tac/tac-locale.sh
index 2bb6e404c..f6383c2da 100755
--- a/tests/tac/tac-locale.sh
+++ b/tests/tac/tac-locale.sh
@@ -38,6 +38,8 @@ export LC_ALL=$LOCALE_FR_UTF8
 if test "$(locale charmap 2>/dev/null)" = UTF-8; then
   check_separator '\u0434'  # д
   check_separator '\u0434\u0436'  # дж
+  # invalid UTF8|unpaired surrogate|C1 control|noncharacter
+  check_separator '\xC3|\xED\xBA\xAD|\u0089|\xED\xA6\xBF\xED\xBF\xBF'
 fi
 
 Exit $fail
-- 
2.52.0

Reply via email to