Re: Using defines for protocol characters

On 2023-Aug-03, Dave Cramer wrote:

> New patch attached which uses PREPARED_SUB_COMMAND and
> PORTAL_SUB_COMMAND instead

Hmm, I would keep the prefix in this case and make the message type a
second prefix, with the subtype last -- PQMSG_CLOSE_PREPARED,
PQMSG_DESCRIBE_PORTAL and so on.

You define PASSWORD and GSS both with 'p', which I think is bogus; in
some place that isn't doing GSS, you've replaced 'p' with the GSS one
(CheckSASLAuth). I think it'd be better to give 'p' a single name, and
not try to distinguish which is PASSWORD and which is GSS, because
ultimately it's not important.

There are some unpatched places, such as basebackup_copy.c -- there are
several matches for /'.'/ that correspond to protocol chars in that file.
Also CopySendEndOfRow has one 'd', and desc.c has two 'C'.

I think fe-trace.c will require further adjustment of the comments for
each function. We could change them to be the symbol for each char, like so:

/* PQMSG_RESP_READY_FOR_QUERY */
static void
pqTraceOutputZ(FILE *f, const char *message, int *cursor)

Thanks

--
Álvaro Herrera 48°01'N 7°57'E — https://www.EnterpriseDB.com/
"Someone said that it is at least an order of magnitude more work to do
production software than a prototype. I think he is wrong by at least
an order of magnitude." (Brian Kernighan)