Stuart I guess if you wanted a separate file just holding IDs, you could create a file pointer direct to the index and use that..
Or treat it as a BTree file if you wanted to use partial searches using BSCAN.. I'm racking my brain to think of anything else. Brian I have found a file with an index on @ID. I know that it prevents the SAMPLE modifier from working on selects but does anyone know if there are any *good* reasons to have an index on the primary key of a file? Cheers, Stuart Boydell ------- u2-users mailing list [email protected] To unsubscribe please visit http://listserver.u2ug.org/
