Re: Frontend error logging style

Andres Freund <andres@anarazel.de>

From: Andres Freund <andres@anarazel.de>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Peter Eisentraut <peter.eisentraut@enterprisedb.com>, Robert Haas <robertmhaas@gmail.com>, Michael Paquier <michael@paquier.xyz>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2022-03-22T02:13:48Z
Lists: pgsql-hackers
Hi,

On 2022-03-21 22:00:37 -0400, Tom Lane wrote:
> Andres Freund <andres@anarazel.de> writes:
> > This unfortunately has had some bitrot: http://cfbot.cputube.org/patch_37_3574.log
> 
> Yeah, I know.  That patch touches enough places that it's sure to
> break every few days during a commitfest.  I'm not real excited about
> maintaining it reactively.

Makes sense. I'd leave it on waiting-on-author then, so that reviewers looking
through the CF don't need to look at it?


> Maybe the plan could be to push it at the end of the commitfest?

Would make sense to me.

Arguably parts of it could be committed sooner than that, e.g. exit(-1). But
that'd not make it meaningfully easier to maintain, so ...


> But what I'd really like at the moment is buy-in or rejection of the whole
> concept, so I know whether it's worth spending more time on.

I'm +1. I've not carefully looked through every single changed callsite, but
IMO it looks like a clear improvement.

Greetings,

Andres Freund



Commits

  1. logging: Also add the command prefix to detail and hint messages

  2. Remove not-very-useful early checks of __pg_log_level in logging.h.

  3. Improve frontend error logging style.

  4. Apply PGDLLIMPORT markings broadly.