From: | Andrew Dunstan <andrew(at)dunslane(dot)net> |
---|---|
To: | Michael Paquier <michael(at)paquier(dot)xyz> |
Cc: | pgsql-committers(at)lists(dot)postgresql(dot)org |
Subject: | Re: pgsql: Remove command checks in tests of pg_basebackup and pg_receivewa |
Date: | 2022-05-09 14:10:27 |
Message-ID: | d7897033-f823-f670-ec57-d68eeb9ad7e3@dunslane.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-committers |
On 2022-05-09 Mo 01:28, Michael Paquier wrote:
> Hi Andrew,
>
> On Sat, May 07, 2022 at 09:13:48AM +0900, Michael Paquier wrote:
>> Let's do that for now then as we need a middle ground for HEAD. I'll
>> come up with something at the beginning of next week.
> And here you go as of the attached to show the idea. The CI is able
> to execute and detect the default commands for lz4, gzip and zstd,
> while one of my boxes without any of those commands would ignore them.
> What do you think?
The system() call should include redirection, since we only case about
the return code.
I would just write this
+ if ($rc != 0)
+ {
+ # Execution of the default command failed, so reset its
+ # environment value.
+ $ENV{$envname} = '';
+ return;
+ }
+
+ # Set the environment to the default as it exists.
+ $ENV{$envname} = $envdefault;
+ return;
as
$ENV{$envname} = $envdefault if $rc == 0;
There should be no need to set it to the empty string in case if
failure, we already know it's undefined.
Otherwise looks good.
cheers
andrew
--
Andrew Dunstan
EDB: https://www.enterprisedb.com
From | Date | Subject | |
---|---|---|---|
Next Message | Noah Misch | 2022-05-09 15:37:46 | pgsql: In REFRESH MATERIALIZED VIEW, set user ID before running user co |
Previous Message | Peter Eisentraut | 2022-05-09 10:33:48 | pgsql: Translation updates |