[Oracle-l] Re: invalid views

I was also not aware of this limitation w/ regard to JDBC Thin connections.

Tim,

Do you have more details on this, or a source?

I suppose it would be pretty easy to test. Perhaps I'll try that...(on Monday, *after* the holiday weekend!)

Have a great holiday, everyone,

-Mark

From: oracle-l-bounce_at_freelists.org On Behalf Of Hemant K Chitale
Sent: Wednesday, November 24, 2010 6:44 PM
To: tim_at_evdbt.com
Cc: oracle-l-freelists; Brian.Zelli_at_roswellpark.org
Subject: Re: invalid views

"older jdbc thin drivers will not perform the automagic compile" ?
I thought that compilation was always done by the server process; the client can only submit parse calls for a SQL statement ?

Hemant K Chitale
http://hemantoracledba.blogspot.com
sent from my smartphone
On Nov 25, 2010 3:55 AM, "Tim Gorman" > wrote:
Invalidated views don't affect the execution plan, because parsing stops at the invalidated object. If you are using one of the Oracle OCI drivers, then your session will automagically attempt to recompile the invalidated object transparently, then proceed with the parse operation. Some of the older, less-capable JDBC "thin" drivers will not perform the automagic recompile and will just return an error message.

Here is a simple test to run in SQL*Plus to illustrate...
create table xxxt (n1 number);
insert into xxxt values (1);
insert into xxxt values (2);
commit;
create view xxxv as select * from xxxt;
select * from xxxt;
select * from xxxv;
alter table xxxt add (n2 number);
select status from user_objects where object_name = 'XXXV';
select * from xxxv;
select status from user_objects where object_name = 'XXXV';

Hope this helps...

Tim Gorman

consultant -> Evergreen Database Technologies, Inc.

postal => 13087 West Cedar Drive #225, Lakewood CO 80228

website => http://www.EvDBT.com/

email => Tim_at_EvDBT.com

mobile => +1-303-885-4526

fax => +1-303-484-3608

Lost Data? => http://www.ora600.be/ for info about DUDE...
-- http://www.freelists.org/webpage/oracle-l