Re: Portal->commandTag as an enum

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Mark Dilger <mark(dot)dilger(at)enterprisedb(dot)com>
Cc: Alvaro Herrera <alvherre(at)2ndquadrant(dot)com>, John Naylor <john(dot)naylor(at)2ndquadrant(dot)com>, Andres Freund <andres(at)anarazel(dot)de>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Portal->commandTag as an enum
Date: 2020-02-29 01:42:02
Message-ID: 9854.1582940522@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Mark Dilger <mark(dot)dilger(at)enterprisedb(dot)com> writes:
>> On Feb 28, 2020, at 3:05 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> Is there a way to drop that logic altogether by making the tagname string
>> be "INSERT 0" for the INSERT case? Or would the zero bleed into other
>> places where we don't want it?

> In general, I don't think we want to increase the number of distinct
> tags. Which command you finished running and whether you want a
> rowcount and/or lastoid are orthogonal issues.

Well, my thought is that last_oid is gone and it isn't ever coming back.
So the less code we use supporting a dead feature, the better.

If we can't remove the special case in EndCommand() altogether, I'd be
inclined to hard-code it as "if (tag == CMDTAG_INSERT ..." rather than
expend infrastructure on treating last_oid as a live option for commands
to have.

regards, tom lane

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Nikita Glukhov 2020-02-29 02:13:40 Re: [PATCH] Opclass parameters
Previous Message Tom Lane 2020-02-29 01:35:33 Re: Allowing ALTER TYPE to change storage strategy