Re: should we add a XLogRecPtr/LSN SQL type?

On 2014-02-04 21:04:13 +0900, Michael Paquier wrote:
> On Tue, Feb 4, 2014 at 7:22 PM, Andres Freund <andres@2ndquadrant.com> wrote:
> > On 2014-02-04 19:17:51 +0900, Michael Paquier wrote:
> >> @@ -180,7 +175,7 @@ pg_get_replication_slots(PG_FUNCTION_ARGS)
> >>                 else
> >>                         nulls[i++] = true;
> >>                 if (restart_lsn != InvalidTransactionId)
> >> -                       values[i++] = CStringGetTextDatum(restart_lsn_s);
> >> +                       values[i++] = restart_lsn;
> >>                 else
> >>                         nulls[i++] = true;
> >
> > Isn't that missing a LSNGetDatum()?
> Oops yes. Will fix.
> 
> > Also, isn't it lacking the corresponding pg_proc change?
> restart_lsn is the 6th argument of pg_get_replication_slots, and the
> list of arguments of this function is already changed like that in my
> patch:
> {25,25,26,16,28,25} => {25,25,26,16,28,3220}
> Regards,

Ok.

I think the patch should also adapt pageinspect...

Greetings,

Andres Freund

-- Andres Freund                       http://www.2ndQuadrant.com/PostgreSQL Development, 24x7 Support, Training &
Services