Message-ID: <3E7B3BA7.20709@web.de>
Date: Fri, 21 Mar 2003 17:19:51 +0100
From: Andreas Pflug <Andreas.Pflug@web.de>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.3a) Gecko/20021212
MIME-Version: 1.0
To: pgsql-odbc@postgresql.org
Subject: using domain types with ODBC, esp. lo
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit

In my schema definitions, I usually don't use basic datatypes, but self 
defined domains. This seems to work in general, but I encountered 
problems with blobs.
I'm using CREATE DOMAIN t_image AS lo;
While inserting data isn't sensitive to domains, selecting won't work. 
The reason is that the row's datatype obtained with CI_read_fields is 
compared to lo's oid. t_image's oid is different, so the column is 
handled wrong. This means, that at the moment domains are not usable on 
lo columns with ODBC.
As a workaround, the driver could maintain an internal list of domain's 
oids, mapping them to lo's oid when CI_read_fields detects one.

There might be another flaw inside CI_read_fields.
The well-known data types PG_TYPE_DATETIME, PG_TYPE_TIMESTAMP_NO_TMZONE, 
PG_TYPE_TIME, PT_TYPE_TIME_WITH_TMZONE are handled in a special way 
concerning the atttypmod field. This will fail for domain types of that 
base type.

Actually, best thing for this problems would be if the backend would 
deliver the base type oid, not the user type oid.
And lo could be hard-coded to the backend too... :-)

Regards,

Andreas