| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | "George Barbarosie" <george(dot)barbarosie(at)gmail(dot)com> |
| Cc: | pgsql-bugs(at)postgresql(dot)org |
| Subject: | Re: BUG #2375: ALTER COLUMN TYPE on composite types |
| Date: | 2006-04-04 19:45:32 |
| Message-ID: | 13727.1144179932@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-bugs |
"George Barbarosie" <george(dot)barbarosie(at)gmail(dot)com> writes:
> CREATE TYPE composite AS (
> label varchar(20),
> data varchar(1000) );
> CREATE TYPE composite2 AS (
> label varchar(20),
> data text );
> CREATE CAST (composite as composite2) WITHOUT FUNCTION AS IMPLICIT;
The above is invalid --- since values of composite types have their type
OID embedded in them, a coercion can't simply be a relabeling. You've
used your superuser powers to break the system, as is all too easy with
binary casts :-(
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Euler Taveira de Oliveira | 2006-04-05 02:10:35 | Re: NLS vs error processing, again (was Re: Composite Type |
| Previous Message | Tom Lane | 2006-04-04 14:41:13 | NLS vs error processing, again (was Re: Composite Type with Domain) |