Date: Thu, 1 Apr 2021 16:07:19 -0300
From: "'alvherre@alvh.no-ip.org'" <alvherre@alvh.no-ip.org>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: "iwata.aya@fujitsu.com" <iwata.aya@fujitsu.com>,
	"tsunakawa.takay@fujitsu.com" <tsunakawa.takay@fujitsu.com>,
	'Kyotaro Horiguchi' <horikyota.ntt@gmail.com>,
	"k.jamison@fujitsu.com" <k.jamison@fujitsu.com>,
	"pgsql-hackers@lists.postgresql.org" <pgsql-hackers@lists.postgresql.org>
Subject: Re: libpq debug log
Message-ID: <20210401190719.GA29711@alvherre.pgsql>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <3458958.1617302154@sss.pgh.pa.us>
User-Agent: Mutt/1.10.1 (2018-07-13)
Archived-At: 
 <https://www.postgresql.org/message-id/20210401190719.GA29711%40alvherre.pgsql>
Precedence: bulk

On 2021-Apr-01, Tom Lane wrote:

> So drongo is still failing, and after a bit of looking around at
> other code I finally got hit with the clue hammer.  Per port.h:
> 
>  * On Windows, setvbuf() does not support _IOLBF mode, and interprets that
>  * as _IOFBF.  To add insult to injury, setvbuf(file, NULL, _IOFBF, 0)
>  * crashes outright if "parameter validation" is enabled.  Therefore, in
>  * places where we'd like to select line-buffered mode, we fall back to
>  * unbuffered mode instead on Windows.  Always use PG_IOLBF not _IOLBF
>  * directly in order to implement this behavior.
> 
> You want to do the honors?  And do something about that shift bug
> while at it.

Ooh, wow ... now that is a silly bug!  Thanks, I'll push the fix in a
minute.

Andrew also noted that the use of command_ok() in the test file eats all
the output and is what is preventing us from seeing it in the first.
I'll try to get rid of that together with the rest.

-- 
Álvaro Herrera       Valdivia, Chile