The reason being, I suppose, to avoid the tedium and error-prone activity of typing. Again, if that file structure were to change, I wouldn't be happy. Most of all, actually, a sense that dumb tasks like this should be automated. On the other hand, there's always the question of whether automating something is worth the effort.

Thanks for your help, Sergey. Have a nice weekend.

Jim


>Unless you've reason not to, I would just list 15 fields in the select and group on them.
>
>>
>>I'm not sure that I said I could use GROUP BY on all fields, and I certainly didn't mean to. Since I'm doing SUM(AMT), certainly I couldn't group by the AMT field, could I?
>>
>>The first time I stumbled over this problem, I had about 15 fields that were always identical for each occurence of "OrdNbr" (so these should have been in one table), and about 6 more that were really detail items (part, quantity, price, etc.) that should have been in a separate table. My ultimate goal is to get a table with all the 15 fields that belong in an "Order table" (one record per unique OrdNbr), along with one or more fields that are a summary from what should have been in the detail table.
>>
>>My apologies if my posts did not make this clear.
>>
>>Jim
>>
>>P.S.: As you've noted quite recently, it's almost impossible to pose a question in this forum and provide too much information about it. Being aware of that, I tried to be as explicit as possible in my post. Reminds me of an old algorithm for estimating the time to do a project ... take your best possible guess on how long it will take, and add 25%. As you can see, having done this, you now have a new best possible guess, to which you have to add ....
>>