Re: View getting invalidated

Hi Yash,
I just did what you described as I understood it in your post...created a view based on a union query, then altered both tables (in this case added a column) and it returned to VALID status after the 'alter view compile'...also it returned to valid status after the first select on the view (which it also should do). The only difference to this behaviour that comes to mind right now is if you are compiling these tables remotely in which case the behaviour is different, or, if there is somehow another dependency in the mix (you can check USER_DEPENDENCIES to verify that)...

I hope this is of at least SOME help :)

Steve

"Yash R. Ganthe" <yash_ganthe_at_hotmail.com> wrote in message news:aa3rgr$t4s$1_at_news.vsnl.net.in...
> Hi,
> I have 2 tables A and B. The schema of both is identical. A view V is
> defined as ( select * from A union all select * from B).
> If I drop a particular column from both A and B, the view V is left
> invalidated. The Oracle documentation also says so. Any view that depends
on
> a table that has been altered, remains invalidated till it is recompiled.
>
> So I did a : alter view V compile.
> This command succeeded.But the view did not get validated.
>
> Can anyone guess why this happens? One would expect the query to be
> reevaluated after recompilation. Since there is a select *, it should work
> fine. But it doesn't. Can someone suggest something?
>
> Thanks
> Yash
>
>
Received on Tue Apr 23 2002 - 11:44:27 CDT