From: | John DeSoi <jd(at)icx(dot)net> |
---|---|
To: | Yasir Malik <ymalik(at)cs(dot)stevens-tech(dot)edu> |
Cc: | PostgreSQL <pgsql-sql(at)postgresql(dot)org> |
Subject: | Re: Object-relational features |
Date: | 2004-03-15 13:28:50 |
Message-ID: | B21C3062-7684-11D8-A24F-000A95B03262@icx.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-sql |
On Mar 13, 2004, at 12:30 PM, Yasir Malik wrote:
> For
> example, using "create type as" is totally worthless because you can't
> use
> it as a field type in a table; you can't compose in another "create
> type
> as"; and you can't inherit another composite type. The only way to
> create
> a true type is to use "create type" and write C code as a shared
> object,
> so I'm basically doing everything C, which is not something I want to
> do.
I'm not sure if this is what you are looking for, but it shows how to
create a column type based on the text type. So your selects will
return the column type as your custom type and you can process the
content accordingly. From reading the docs (and asking on the list) I
did not think this was possible either without writing external code in
C. But a post about something else finally provided the clues I needed
to get it working.
Best,
John DeSoi, Ph.D.
====
test=# create or replace function lispin(cstring, oid, int4) returns
lisp as 'varcharin' language 'internal' immutable strict;
NOTICE: type "lisp" is not yet defined
DETAIL: Creating a shell type definition.
CREATE FUNCTION
test=# create or replace function lispout(lisp) returns cstring as
'varcharout' language 'internal' immutable strict;
NOTICE: argument type lisp is only a shell
CREATE FUNCTION
test=# create type lisp (input=lispin, output=lispout,
internallength=variable);
CREATE TYPE
test=# create table tst (a lisp);
CREATE TABLE
test=# insert into tst (a) values ('1');
INSERT 18499 1
test=# insert into tst (a) values ('(+ 5 5)');
INSERT 18500 1
test=# select * from tst;
a
---------
1
(+ 5 5)
(2 rows)
From | Date | Subject | |
---|---|---|---|
Next Message | Yasir Malik | 2004-03-15 14:06:39 | Re: Object-relational features |
Previous Message | Paul Thomas | 2004-03-15 11:42:23 | Re: Invalid Characters |