Can anyone point me in the right direction? I am trying to do a running count/total for a list of items that will reset back to 1 after each group. For example. If I have a list of cars by country, I would like to list each alphabetically but to start from 1 through n for each list.

Each car would be listed like:


1 Mazda CX-9 2007
2 Nissan wb 2007
3 Subaru Outback 2006

Then for the next list/group it would begin again:

1 Buick Skylark 1980
2 Chevrolet Truck 2005
3 Ford GT 2005

So far I can only get the list to count in one direction using a
DEFINE
CNTR/I6 WITH ID = CNTR+1;

But I can't figure out how to get it to reset back to 1 after each grouping. Has anyone done anything like this?

simplest answer, Greg ...
Its all about the word LAST
lets assume your file is in COUNTRY order, the order you want to print ...
DEFINE FILE CAR
CNTR/I5 WITH ID = IF COUNTRY NE LAST COUNTRY THEN 1 ELSE 1 + LAST CNTR;
END

But that assumes that your records are already sorted in COUNTRY order.

What about LIST:

TABLE FILE CAR
LIST
CAR
BY COUNTRY
BY CAR
END

Francis,
One BY too many.

TABLE FILE CAR
LIST
CAR
BY COUNTRY
END

Greg,
When you use LIST, the numbering restarts automatically when the value of the OUTERMOST BY changes.

That works perfectly, susannah, thanks!

I have tried LIST before and that works (which can be a simple solution for anyone to try out) however I wasn't able to get rid of the "LIST" heading since this is a custom column. I did some searches on the forum for that and I heard that it is impossible to remove it without having to hold the file again and re-PRINT it (too many steps). But this takes care of it just fine. Unless someone can figure out how to supress the LIST header

Greg,
Here is your solution!

TABLE FILE CAR
BY COUNTRY
RANKED AS GREG BY CAR
END

Simple and elegant.