From: | "David E(dot) Wheeler" <david(at)kineticode(dot)com> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Type Categories for User-Defined Types |
Date: | 2008-07-29 23:22:06 |
Message-ID: | 559754CA-52DC-4D03-8BB1-18EB0F673122@kineticode.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Jul 29, 2008, at 15:35, Tom Lane wrote:
> BTW, I did confirm (by temporarily hacking up TypeCategory()) that
> causing citext to appear to be of STRING category eliminates the need
> for the extra || operator and quote_literal() function that are in
> the current citext code. So the proposed solution really will work.
Nice! Does it also allow all of the casts to and from text to
implicitly work, e.g,
SELECT 4::int8::citext = '4' AS t;
SELECT '4'::citext::int8 = 4 AS t;
SELECT 4::bigint::citext = '4' AS t;
SELECT '4'::citext::bigint = 4 AS t;
> You might still want to keep the aliases in cases where the point is
> to
> have the function or operator output resolve as citext rather than
> text.
> I'm not sure how many of these cases that's really important for,
> though.
Not many, I should think.
Thanks,
David
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2008-07-29 23:45:11 | Re: Type Categories for User-Defined Types |
Previous Message | Tom Lane | 2008-07-29 22:35:10 | Re: Type Categories for User-Defined Types |