Re: CALLing other pgms - what is a good method of handling different parm lengths

On 2012-06-27, at 10:20 AM, Stone, Joel wrote:

I have a pgm ProperCase to change an UPPER-CASE string to Mixed-Case.

For example, "123 ELM ST" changes to "123 Elm St"

The field that I want to pass as a parm may be different lengths for example

CUST-NAME PIC x(50).

CUST-ADDRESS-1 PIC X(30).


I have been coding as follows:

move CUST-ADDRESS-1 of CUST-FILE to ws-string
call 'ProperCase' using ws-string
move ws-string to PRINT-LINE.



I would much prefer to do something like

Move ProperCase (CUST-ADDRESS-1) to PRINT-LINE

/|\
|
|------subprocedure name



Is this possible?

Nope - sadly when the COBOL gods introduced functions they never went beyond the equivalent of RPG BIFS - no way to write your own. If you wanted simple upper-casing for example you could do this:

move function upper-case(CUST-ADDRESS-1) to PRINT-LINE.

Surely not, but

My next choice, to make fewer lines of code and thus more readable code, would be to

move CUST-ADDRESS-1 to PRINT-LINE
call 'ProperCase' using PRINT-LINE

Problem is that in the pgm ProperCase, the parm is defined as very long x(1000) to accommodate a long string. But when I make this call, it wipes out everything in memory past the field that I am passing as a parm.

Any way to avoid this?

Thanks!

This would do the same in RPG surely? or are you using a varying string? I think ProperCase needs to be modified to take the length of the field it is handling. Then it could constrain its operation to the defined length. Simple pass the "length of" value to make sure the length is accurate.

Jon Paris

www.partner400.com
www.SystemiDeveloper.com