Re: Using SQL to check for duplicate records

Well, folks, my third post, but if you're using an RPG
cursor you can "simulate" lookahead by reading into one data
structure, this makes *all* the fields into "lookahead"
fields, and after you do [whatever] you "move" it to the
other "current" data structure, prefixed differently. I do
this often, to help keep logic inside one program instead of
two (I've used QCMDEXC often enough but it's too ugly).

For RPG, Tom's "lookahead" works better than my array
example. It's just that after a series of programs where the
array was the only way to make it work properly
(multi-column output, getting a percentage over L1 sums,
etc., I got into the habit).

If it's a key field, and response time is the deciding
factor, it looks like lookahead to me.

- Alan