Re: pgsql: Require version 0.98 of Test::More for TAP tests

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Andrew Dunstan <andrew(at)dunslane(dot)net>
Cc: Noah Misch <noah(at)leadboat(dot)com>, pgsql-committers(at)lists(dot)postgresql(dot)org
Subject: Re: pgsql: Require version 0.98 of Test::More for TAP tests
Date: 2021-11-21 17:46:54
Message-ID: 3468629.1637516814@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-committers

I wrote:
> Seems reasonable to me, although it's not entirely clear how to hook
> the output into configure's practices --- in particular, people using
> "./configure -q" might not be pleased by unwanted diagnostic output.
> But that could probably be dealt with.

Here's a draft patch based on Andrew's script. I'm not entirely
satisfied with it, because I couldn't get the output to go where
I wanted. I think that the reports of actual module versions should
go to config.log, and not appear in configure's user-visible output,
because (a) 99% of people won't care, and (b) otherwise we don't have
a good way to silence them under -q, where 100% of people won't care.

However, the only way I could get that to happen was to redirect
prove's stderr to config.log (&AS_MESSAGE_LOG_FD), which is not great
because it means that in the failure case the only place where any useful
info appears is config.log. I tried printing the version reports to
STDOUT instead of using diag(), but that seems to get redirected to
/dev/null somewhere. Anybody know how to get prove to not do that?

regards, tom lane

Attachment Content-Type Size
test-module-versions-under-prove-wip.patch text/x-diff 6.6 KB

In response to

Responses

Browse pgsql-committers by date

  From Date Subject
Next Message Tom Lane 2021-11-21 19:14:00 pgsql: pg_receivewal, pg_recvlogical: allow canceling initial password
Previous Message Tom Lane 2021-11-21 16:49:55 pgsql: Doc: update some things relevant to minimum Test::More version.