Re: Multiple Table Query

You might switch to a partitioned table: Each day add a new partition
Name the partitions in some 'date-related' way and generate a view of a single partition if you want to access the latest one particularly frequently

e.g,

    Alter table data_table
    add partition p_2000_02_14
    values less than (to_date('15-Feb-2000','dd-mon-yyyy));

    create or replace view data_view as
    select * from data_table partition (p_2000_02_14);

You will need a method for populating the range partitioning column with the correct date for the day, though.

You '%' query becomes:

    select count(*) from data_table;

--

Jonathan Lewis
Yet another Oracle-related web site:  http://www.jlcomp.demon.co.uk

M. Misener wrote in message <889r1m$55e$1_at_mohawk.hwcn.org>...


>I have an application that archives data each day into a table.

>Every day, and new table is created with same structure (same fields, same

>columns).

>The difference being, that the table_name is unique. ex: ARCH_020100,

>ARCH_020200, etc., base on Date.

>

>I'm not sure how construct a query to select count(*) where table_name

 "like


>ARCH_%" (something like that).

>

>Can anyone help?

>Thanks.

>

>

>