From: | Michael Paquier <michael(at)paquier(dot)xyz> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | Daniel Gustafsson <daniel(at)yesql(dot)se>, pgsql-committers(at)lists(dot)postgresql(dot)org, Andrew Dunstan <andrew(at)dunslane(dot)net> |
Subject: | Re: pgsql: Add TAP tests for contrib/sslinfo |
Date: | 2021-12-01 22:37:49 |
Message-ID: | Yaf5PUW259/e7W3D@paquier.xyz |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-committers |
On Wed, Dec 01, 2021 at 04:04:09PM -0500, Tom Lane wrote:
> I agree that some docs about this would be nice, but I think the rules
> for when to use perl2host are more complex than what you suggest here.
> Perhaps Andrew can clarify.
+When passing a path from the Perl test code to PostgreSQL, like for example
+the path to an SSL certificate, it must be converted for use on the host
+system by using PostgreSQL::Test::Utils::perl2host.
It seems to me that the rule applies to any paths passed down to the C
code of Postgres, then processed by a system call. It would be good
to add some examples of what that is, I guess. Until now, I think
that we have seen:
- Tablespace paths in SQL queries.
- Paths passed down as function arguments, processed in system calls
by frontends (like pg_basebackup and its tablespace map).
- Paths part of connection strings, processed in system calls in
libpq.
--
Michael
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2021-12-01 22:43:57 | Re: pgsql: Add TAP tests for contrib/sslinfo |
Previous Message | Tom Lane | 2021-12-01 21:04:09 | Re: pgsql: Add TAP tests for contrib/sslinfo |