Date: Mon, 28 May 2018 11:01:25 -0400
From: Alvaro Herrera <alvherre@2ndquadrant.com>
To: Thomas Munro <thomas.munro@enterprisedb.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>,
	PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Subject: Re: Allowing printf("%m") only where it actually works
Message-ID: <20180528150125.j2x3mp62ssbeg3h5@alvherre.pgsql>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: 
 <CAEepm=1YLObd0aX9K4jqOHwQF3LQzSJB1VCNd-GstZhN67Q4CA@mail.gmail.com>
User-Agent: NeoMutt/20170306-137-4415bd-dirty (1.8.0)
Precedence: bulk

On 2018-May-27, Thomas Munro wrote:

> Out of curiosity I tried adding a GetLastError variable for Windows
> (to hide the function of that name and break callers) to the earlier
> experimental patch (attached).  I had to give it an initial value to
> get rid of a warning about an unused variable (by my reading of the
> documentation, __pragma(warning(suppress:4101)) can be used in macros
> (unlike #pragma) and should shut that warning up, but it doesn't work
> for me, not sure why).  Of course that produces many errors since we
> do that all over the place:
> 
> https://ci.appveyor.com/project/macdice/postgres/build/1.0.184

Ouch.

This seems to say that we oughta assign GetLastError() to saved_errno
during errstart, then use %m in the errmsg() instead.

-- 
Álvaro Herrera                https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services