MVSFORUMS.com :: View topic - Convert external floating point number to COMP-2

Convert external floating point number to COMP-2

through

[^/[Print]\]

MVSFORUMS.com -> Application Programming

#1: Convert external floating point number to COMP-2 Author: jathampy, Location: UK

Posted: Wed Aug 02, 2023 11:13 am
—
I have an input file that contain amount fields in exponential form like "-2.2204462782314e-16". I want to convert this into COMP-2 format.

Tried NUMVAL-F intrinsic function and the cobol compiler (Z/OS 4.2.0) giving compilation error "expected a function name, but found NUMVAL-F". However NUMVAL and NUMVAL-C intrinsic functions are working.

Any idea how to convert the exponential representation of numbers to COMP-2 internal floating point ?

#2: Author: kolusu, Location: San Jose

Posted: Wed Aug 02, 2023 5:08 pm
—
jathampy,

Try this

Code:

WORKING-STORAGE SECTION.
01 WS-INP-FLNUM PIC +V9(16)E+99 GLOBAL.
01 WS-INP-FLCHAR REDEFINES WS-INP-FLNUM.
05 A PIC X(20).
01 WS-OUT-FL COMP-2.

PROCEDURE DIVISION.
MOVE "-2.2204462782314E-16" TO A.
DISPLAY 'CHAR VALUE : ' A
MOVE WS-INP-FLNUM TO WS-OUT-FL
DISPLAY 'COMP VALUE : ' WS-OUT-FL
GOBACK.

Result from 6.xx compiler

Code:

CHAR VALUE : -2.2204462782314E-16
COMP VALUE : -.22204462782314002E-15

#3: Author: jathampy, Location: UK

Posted: Sat Aug 05, 2023 3:47 am
—
Thanks Kolusu for providing the solution. I tried the solution but job failed with

Quote:

"IGZ0040S An invalid separate sign was detected in program"

while executing the move statement " MOVE WS-INP-FLNUM TO WS-OUT-FL"

We are using the old cobol compiler (Z/OS 4.2.0) and this will be upgraded only next year post Z/OS upgrade.

#4: Author: kolusu, Location: San Jose

Posted: Thu Aug 10, 2023 10:06 am
—
jathampy,

Can you give this a try ?

Code:

WORKING-STORAGE SECTION.
01 WS-INP-FLNUM PIC +V9(16)E+99 GLOBAL.
01 WS-INP-FLCHAR REDEFINES WS-INP-FLNUM.
05 A PIC X(21).
01 WS-OUT-FL COMP-2.

PROCEDURE DIVISION.

MOVE "-2.2204462782314E-16" TO A.
DISPLAY 'CHAR VALUE A : ' A

COMPUTE WS-OUT-FL = FUNCTION NUMVAL-F(A)
DISPLAY "COMP-2 Value is : " WS-OUT-FL

#5: Author: jathampy, Location: UK

Posted: Mon Aug 14, 2023 9:29 am
—
I tried NUMVAL-F before also and it's not working. The Other intrinsic functions NUMVAL and NUMVAL-C are working fine with the compiler version 4.2.0 V4R2 . When I use NUMVAL-F,it's giving compilation error:

Code:

IGYPS2130-S Expected a function-name, but found "NUMVAL-F". The "COMPUTE" statement was discarded.

MVSFORUMS.com -> Application Programming

output generated using printer-friendly topic mod. All times are GMT - 5 Hours

Page 1 of 1