Re: More protocol.h replacements this time into walsender.c

Dave Cramer

On Fri, 25 Jul 2025 at 04:11, Álvaro Herrera <alvherre(at)kurilemu(dot)de> wrote:

> On 2025-Jul-24, Dave Cramer wrote:
>
> > On Thu, 24 Jul 2025 at 17:05, Jacob Champion <
> > jacob(dot)champion(at)enterprisedb(dot)com> wrote:
> >
> > > On Thu, Jul 24, 2025 at 12:04 PM Dave Cramer <davecramer(at)gmail(dot)com>
> wrote:
>
> > > +/* Replication Protocol sent by the primary */
> > > +
> > > +#define PqMsg_XlogData 'w'
> > > +#define PqMsg_PrimaryKeepAlive 'k'
> > > +#define PqMsg_PrimaryStatusUpdate 's'
> > > +
> > > +
> > > +/* Replication Protocol sent by the standby */
> > > +
> > > +#define PqMsg_StandbyStatus 'r'
> > > +#define PqMsg_HotStandbyFeedback 'h'
> > > +#define PqMsg_RequestPrimaryStatus 'p'
> > >
> > > Since these are part of the replication subprotocol (i.e. tunneled,
> > > via CopyData) rather than the top-level wire protocol, do they deserve
> > > their own prefix? PqReplMsg_* maybe?
> >
> > I'm going to wait to see if there are any other opinions. Last time I did
> > this there were quite a few opinions before finally settling on the
> naming
>
> Count me in.
>

FYI, the reason I used XLogData is because the term is used multiple times
here https://www.postgresql.org/docs/devel/protocol-replication.html

Dave