Re: pgbench -f and vacuum

Tomas Vondra wrote:
> On 22.12.2014 07:36, Tatsuo Ishii wrote:
> > On 22.12.2014 00:28, Tomas Vondra wrote:

> >> (8) Also, I think it's not necessary to define function prototypes for
> >> executeStatement2 and is_table_exists. It certainly is not
> >> consistent with the other functions defined in pgbench.c (e.g.
> >> there's no prototype for executeStatement). Just delete the two
> >> prototypes and move is_table_exists before executeStatement2.
> >
> > I think not having static function prototypes is not a good
> > custom. See other source code in PostgreSQL.
>
> Yes, but apparently pgbench.c does not do that. It's strange to have
> prototypes for just two of many functions in the file.

Whenever a function is defined before its first use, a prototype is not
mandatory, so we tend to omit them, but I'm pretty sure there are cases
where we add them anyway. I my opinion, rearranging code so that called
functions appear first just to avoid the prototype is not a very good
way to organize things, though. I haven't looked at this patch so I
don't know whether this is what's being done here.

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