On Fri, Aug 25, 2023 at 10:10 AM Andres Freund <and...@anarazel.de> wrote:
> Let's see what the buildfarm says - it's not inconceivable that it'll show
> some issues.
Apparently Solaris doesn't like "cp -a", per animal "margay". I think
"cp -RPp" should be enough everywhere?
https://docs.oracle.com/cd/E88353_01/html/E37839/cp-1.html
https://pubs.opengroup.org/onlinepubs/9699919799.2013edition/utilities/cp.html
From fd6c558e6bd43eef40d633ace763d8a2088c2509 Mon Sep 17 00:00:00 2001
From: Thomas Munro <thomas.mu...@gmail.com>
Date: Fri, 25 Aug 2023 17:30:48 +1200
Subject: [PATCH] Avoid non-POSIX cp flags.
Commit 252dcb32 introduced cp -a, but apparently Solaris doesn't like
it. Use cp -RPp instead.
diff --git a/src/test/perl/PostgreSQL/Test/Cluster.pm b/src/test/perl/PostgreSQL/Test/Cluster.pm
index 426f94ff09..227c34ab4d 100644
--- a/src/test/perl/PostgreSQL/Test/Cluster.pm
+++ b/src/test/perl/PostgreSQL/Test/Cluster.pm
@@ -549,7 +549,7 @@ sub init
}
else
{
- @copycmd = qw(cp -a);
+ @copycmd = qw(cp -RPp);
$expected_exitcode = 0;
}
diff --git a/src/test/regress/pg_regress.c b/src/test/regress/pg_regress.c
index 06674141a3..ec67588cf5 100644
--- a/src/test/regress/pg_regress.c
+++ b/src/test/regress/pg_regress.c
@@ -2355,7 +2355,7 @@ regression_main(int argc, char *argv[],
else
{
#ifndef WIN32
- const char *copycmd = "cp -a \"%s\" \"%s/data\"";
+ const char *copycmd = "cp -RPp \"%s\" \"%s/data\"";
int expected_exitcode = 0;
#else
const char *copycmd = "robocopy /E /NJS /NJH /NFL /NDL /NP \"%s\" \"%s/data\"";
--
2.39.2
