From: | Michael Paquier <michael(at)paquier(dot)xyz> |
---|---|
To: | Tristan Partin <tristan(at)neon(dot)tech> |
Cc: | pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Fix use of openssl.path() if openssl isn't found |
Date: | 2023-11-08 05:53:35 |
Message-ID: | ZUsiX0a8hIORUskh@paquier.xyz |
Views: | Whole Thread | Raw Message | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Tue, Nov 07, 2023 at 04:06:56PM -0600, Tristan Partin wrote:
> Found this issue during my Fedora 39 upgrade. Tested that uninstalling
> openssl still allows the various ssl tests to run and succeed.
Good catch. You are right that this is inconsistent with what we
expect in the test.
> +openssl_path = ''
> +if openssl.found()
> + openssl_path = openssl.path()
> +endif
> +
> tests += {
> 'name': 'ssl',
> 'sd': meson.current_source_dir(),
> @@ -7,7 +12,7 @@ tests += {
> 'tap': {
> 'env': {
> 'with_ssl': ssl_library,
> - 'OPENSSL': openssl.path(),
> + 'OPENSSL': openssl_path,
> },
> 'tests': [
> 't/001_ssltests.pl',
Okay, that's a nit and it leads to the same result, but why not using
the same one-liner style like all the other meson.build files that
rely on optional commands? See pg_verifybackup, pg_dump, etc. That
would be more consistent.
--
Michael
From | Date | Subject | |
---|---|---|---|
Next Message | Michael Paquier | 2023-11-08 05:59:07 | Re: Show WAL write and fsync stats in pg_stat_io |
Previous Message | Dilip Kumar | 2023-11-08 05:46:03 | Re: SLRU optimization - configurable buffer pool and partitioning the SLRU lock |