IMHO, I think there should be a way to retreive the information you need if you have all the unique keys that are necessary in your tables to identify the relations.
In your contact example you can do it even if you don't merge tables by adding a key like :


table NAME
fields : code, name
record1: 1, claude
record2: 3, Rocky

table ADDRESS
fields : code, contact_id, address
record1 : 1, 22, Montreal
record2 : 1, 55, Toronto
record3 : 3, 13, Toronto

table CONTACT
fields : code, contact_id, phone
record1 : 1, 22, 555-5555
record2 : 1, 55, 666-6666
record3 : 3, 13, 666-6666

and then

SELECT * FROM NAME, ADDRESS, CONTACT;
 WHERE NAME.CODE=ADDRESS.CODE AND ADDRESS.CONTACT_ID=CONTACT.CONTACT_ID

That's the idea of a relational database.

Maybe I miss something in your problem??
Someone else please advise. :)

CL


>Thanks for taking out time and replying me ...
>
>But see, the prob. remains same... Pls read my last msg. once again I have given the same problem in a similar kind of structure
>
>Take a case of Order system
>
>We get Export orders, so details abt. shipments, the major ones like order no, order quantity, description of goods, buyer, fob etc. etc. are filed in main file called order.
>
>The delivery dates and quantities of that order are filled in a seperate file called delivery....
>
>The LC info. of Orders is filled in a seperate file called LC
>
>Invoice which we raise are filled in a seperate file...
>
>and like that...there are more then 10-12 child files
>
>so, you see, in this type of structure, you can not merge 2 files like you did in contacts and names...
>
>here delivery dates are also multiple and lc's are also multiple, invoices are also multiple...
>
>And User requires, 1 field from parent, 1 field from 1 child and 1 field from another child....
>
>so one parent and lot of childs...
>
>But no child is realted with each other...
>
>Pls advs.. what will be the SQL in that case....
>
>
>