On 30/06/2025 20:47, Bruno Haible via GNU coreutils Bug Reports wrote:
After this week's changes in 'od', the CI reports a test failure on OpenBSD
and Solaris:


FAIL: tests/od/od
Ah right, strerror(ERANGE) varies by platform.
I'll push the attached later to address this.

Marking this as done.

thanks,
Padraig
From ee22d706e31430a1f729ca55c56e18ed566ba0c5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?P=C3=A1draig=20Brady?= <p...@draigbrady.com>
Date: Mon, 30 Jun 2025 23:28:32 +0100
Subject: [PATCH] tests: avoid false failure due to varied ERANGE errors

* tests/od/od.pl: Use the system error string for ERANGE,
rather than hardcoding a particular one.
Fixes https://bugs.gnu.org/78934
---
 tests/od/od.pl | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/tests/od/od.pl b/tests/od/od.pl
index e4e328ad6..40cc4289a 100755
--- a/tests/od/od.pl
+++ b/tests/od/od.pl
@@ -25,6 +25,9 @@ use strict;
 
 my $prog = 'od';
 
+use Errno qw(ERANGE);
+my $ERANGE = do { local $! = ERANGE; "$!" };
+
 # Use a file in /proc whose size is not likely to
 # change between the wc and od invocations.
 my $proc_file = '/proc/version';
@@ -76,11 +79,11 @@ my @Tests =
 
      # Overflowing traditional offsets should be diagnosed.
      ['overflow-off-1', '-', '7' x 255, {IN_PIPE=>""}, {EXIT=>1},
-      {ERR=>"od: ".('7' x 255).": Numerical result out of range\n"}],
+      {ERR=>"od: ".('7' x 255).": $ERANGE\n"}],
      ['overflow-off-2', '-', ('9' x 254).'.', {IN_PIPE=>""}, {EXIT=>1},
-      {ERR=>"od: ".('9' x 254).".: Numerical result out of range\n"}],
+      {ERR=>"od: ".('9' x 254).".: $ERANGE\n"}],
      ['overflow-off-3', '-', '0x'.('f' x 253), {IN_PIPE=>""}, {EXIT=>1},
-      {ERR=>"od: 0x".('f' x 253).": Numerical result out of range\n"}],
+      {ERR=>"od: 0x".('f' x 253).": $ERANGE\n"}],
 
      # Ensure that a large width does not cause trouble.
      # From coreutils-7.0 through coreutils-8.21, these would print
-- 
2.49.0

Reply via email to