Jason In retrospect I think I should perhaps have should have worded my disclaimer more along the lines of "I haven't bothered to read the question properly". In my haste, I read the requirement to be to split the digit strings into groups of two with any odd one being added to the final group - I believe the calc I posted does that (but I'm not rescinding my original disclaimer, just in case). As always, your eye and mind have proved to be more sensitive than mine cheers Tom On 8 Mar 2011, at 22:21, Jason DeLooze wrote: > John, > > Are you certain Tom's CF works as you want? Try some of the example you originally gave. For example > > TA123/TA-12-3 > or > 123TA45/12-3-TA-45 > > Tom's CF does not produce these results. [Sorry, Tom, I realize you said you didn't have to fully test your CF]. > ====== > > I was going to submit a CF, but realized we didn't have a complete spec for the CF. For example, what should be the output if the input is "TA1234567"? Should it be "TA-12-34-56-7" or "TA-12-34-567"? You have given us the spec for 5 or fewer consecutive numbers [1 ; 12 ; 12-3 ; 12-34 ; 12-345 ], but what about 6 or more numbers? Should they be > > 12-34-56 > 12-34-56-7 or 12-34-567 > 12-34-56-78 > 12-34-56-78-9 or 12-34-56-789 > etc > > With a complete spec, a CF can be produced (and fully tested). > > Regards, > Jason L. DeLooze > Annapolis, MD USA > > > > On 3/8/11 at 10:50 AM -0500, John Wenmeekers wrote: >> Thanks to all of you. >> >> Tom, it works. >> >> I had nearly the same formula as you, only, I put it on the wrong line >> inside the Case()....duh... >> >> Guess I was too long gazing at the screen....to see the obvious...while a >> couple of other eyes can see the problem right away... >> >> Thanks again.... >> >> JW >