Re: logicalrep_message_type throws an error - Mailing list pgsql-hackers

From Alvaro Herrera
Subject Re: logicalrep_message_type throws an error
Date
Msg-id 20230717142452.74yml7mfqkypfixp@alvherre.pgsql
Whole thread Raw
In response to Re: logicalrep_message_type throws an error  (Ashutosh Bapat <ashutosh.bapat.oss@gmail.com>)
Responses Re: logicalrep_message_type throws an error
List pgsql-hackers
On 2023-Jul-17, Ashutosh Bapat wrote:

> Prologue of psprintf() says
> 
> * Errors are not returned to the caller, but are reported via elog(ERROR)
> * in the backend, or printf-to-stderr-and-exit() in frontend builds.
> * One should therefore think twice about using this in libpq.
> 
> If an error occurs in psprintf(), it will throw an error which will
> override the original error. I think we should avoid any stuff that
> throws further errors.

Ooh, yeah, this is an excellent point.  I agree it would be better to
avoid psprintf() here and anything that adds more failure modes.  Let's
just do the thing in the original patch you submitted, to ensure all
these strings are compile-time constants; that's likely the most robust.

-- 
Álvaro Herrera         PostgreSQL Developer  —  https://www.EnterpriseDB.com/
"Pensar que el espectro que vemos es ilusorio no lo despoja de espanto,
sólo le suma el nuevo terror de la locura" (Perelandra, C.S. Lewis)



pgsql-hackers by date:

Previous
From: Aleksander Alekseev
Date:
Subject: Re: Getting rid of OverrideSearhPath in namespace.c
Next
From: Stéphane Tachoires
Date:
Subject: Re: [PATCH] Automatic HASH and LIST partition creation