From: Tom Lane <tgl@sss.pgh.pa.us>
To: "David G. Johnston" <david.g.johnston@gmail.com>
cc: work.michael.2956@gmail.com,
        PostgreSQL Bug List <pgsql-bugs@lists.postgresql.org>
Subject: Re: BUG #18327: Column naming inconsistency for boolean literals in
 ELSE clauses of CASE expressions.
In-reply-to: 
 <CAKFQuwa6pmbdMDvjzkQijaimLwXXQYTGRReT-od0y-E42EjhxQ@mail.gmail.com>
References: <18327-f25f77a50d104716@postgresql.org>
 <CAKFQuwa6pmbdMDvjzkQijaimLwXXQYTGRReT-od0y-E42EjhxQ@mail.gmail.com>
Comments: In-reply-to "David G. Johnston" <david.g.johnston@gmail.com>
	message dated "Sun, 04 Feb 2024 12:01:15 -0700"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <3934129.1707076663.1@sss.pgh.pa.us>
Date: Sun, 04 Feb 2024 14:57:43 -0500
Message-ID: <3934130.1707076663@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/3934130.1707076663%40sss.pgh.pa.us>
Precedence: bulk

"David G. Johnston" <david.g.johnston@gmail.com> writes:
> This is indeed all consistent in that literals don't have names while other
> expression tend to.

Actually, pre-v15 the literal constants "true" and "false" do have
names for this purpose.  For historical reasons they were parsed
into the equivalent of 't'::bool and 'f'::bool, and then the name
selection rule for a typecast took effect.

Starting in v15 they're treated like other sorts of literals,
meaning they have no assigned name:

regression=# select true;
 ?column? 
----------
 t
(1 row)

Many of these other examples change as a consequence.

			regards, tom lane