Re: Unified logging system for command-line programs - Mailing list pgsql-hackers

From Michael Paquier
Subject Re: Unified logging system for command-line programs
Date
Msg-id 20190402030514.GL16093@paquier.xyz
Whole thread Raw
In response to Re: Unified logging system for command-line programs  (Andres Freund <andres@anarazel.de>)
Responses Re: Unified logging system for command-line programs
List pgsql-hackers
On Mon, Apr 01, 2019 at 11:55:09AM -0700, Andres Freund wrote:
> A written upthread, I think this should have had a uniform interface
> with elog.h, and probably even share some code between the two. This is
> going in the wrong direction, making it harder, not easier, to share
> code between frontend and backend. While moving around as much code as
> we'd have had to do if we'd gone to error reporting compatible with
> elog.h.

Like Andres, I am a bit disappointed that this stuff is not reducing
the amount of diff code with ifdef FRONTEND in src/common/.  This
actually adds more complexity than the original code in a couple of
places, like this one which is less than nice:
+#ifndef FRONTEND
+#define pg_log_warning(...) elog(WARNING, __VA_ARGS__)
+#else
+#include "fe_utils/logging.h"
+#endif
--
Michael

Attachment

pgsql-hackers by date:

Previous
From: Masahiko Sawada
Date:
Subject: Re: Re: A separate table level option to control compression
Next
From: Michael Paquier
Date:
Subject: Re: Checksum errors in pg_stat_database