One way to ensure that account numbers are input correctly is to use a check

digit. A check digit is normally a single digit calculated from an algorithm

applied to the original number. One of the most widely used check digit methods

is Modulus 10 (M10). The M10 algorithm works like this:

For each position in the base number, there is a weight factor. Positions are

counted from the farthest right digit (not including the check digit).

The Modulus 10 weight factor is 2 for all odd positions (1, 3, 5, ...) and 1

for all even positions (2, 4, 6, ...).

To calculate the Modulus 10 self-check digit, do the following:

1. Starting from the rightmost digit of the base number, multiply every other

digit by a factor of 2.

2. Add the digits in the products to the digits in the base number that were

not multiplied.

3. Subtract the sum from the next higher number ending in zero.

To create an M10 number, you append the result of step 3 to the end of the base

number.

For example, to calculate the check digit for the number 256, perform the

following steps:

1. Multiply the 1st rightmost digit (6) by 2 (2 * 6 = 12).

2. Multiply the 3rd digit from the right (2) by 2 (2 * 2 = 4).

3. Add each digit of the products and each digit of the base number that wasn't

multiplied by a factor (1+2+5+4 = 12).

4. Subtract the result from the next highest number that ends in a zero (20-12

= 8).

8 is the M10 check digit for the base number 256. The resulting M10 number

would be 2568.

IBM supports the validation of an M10 number through display files with the DDS

CHECK keyword with a parameter value of M10. However, there are no programs in

OS/400 that will generate an M10 check digit for a number. I wrote an RPG

program (M10001RG) to calculate the check digit of a number or, alternatively,

to validate a number that already has a check digit. This way I have the

ability to not only validate an M10 number, but I can also generate an M10

check digit; and, both can be done by calling the same program. 1

check digit; and, both can be done by calling the same program. Figure 1

contains the source code for RPG program M10001RG. The program uses two input

parameters and two output parameters:

Input parameter oneù15 character number to validate or to generate a check

digit for.

Input parameter twoùValidate flag (a value of 'Y' indicates that the program is

to validate the number).

Output parameter oneùContains calculated Modulus 10 check digit if input

parameter two is 'Y'.

Output parameter twoùIf validate flag parameter is set to 'Y', this parameter

will contain a 1 if the number is a valid M10 number; otherwise, it will

contain a value of 0.

--Andrew Longo

Modulus 10 Calculation/Validation Program

Figure 1RPG Program M10001RG

 *=============================================================== * To compile: * * CRTRPGPGM PGM(XXX/M10001RG) SRCFILE(XXX/QRPGSRC) * *=============================================================== *. 1 ...+... 2 ...+... 3 ...+... 4 ...+... 5 ...+... 6 ...+... 7 E NA 15 1 * C *ENTRY PLIST C PARM NBR 15 C PARM VALI 1 C PARM CKDIG 1 C PARM VALID 1 C MOVE *BLANK VALID C MOVE *BLANK CKDIG * C MOVEANBR NA * C NA,2 IFNE *BLANK C NA,15 ANDEQ*BLANK C ' ' CHEKRNBR I 20 C VALI IFEQ 'Y' C MOVE NA,I CKDIG C SUB 1 I C ENDIF * C MOVE *ZEROS TOGLE 10 C MOVE *ZEROS CHSUM 70 * C I DOWGE1 C MOVE NA,I HOLD 10 C TOGLE IFEQ *ZERO C HOLD MULT 2 WRK 20 C ELSE C Z-ADDHOLD WRK C ENDIF C ADD 5 TOGLE C WRK IFGT 9 C SUB 9 WRK C ENDIF C ADD WRK CHSUM C SUB 1 I C ENDDO * C MOVE CHSUM NXTHI 20 C 100 SUB NXTHI CHDIG 10 C MOVE CHDIG CHDIC 1 C VALI IFEQ 'Y' C CHDIC IFEQ CKDIG C MOVE '1' VALID C ELSE C MOVE '0' VALID C ENDIF C ENDIF C MOVE CHDIC CKDIG C ENDIF * C RETRN *. 1 ...+... 2 ...+... 3 ...+... 4 ...+... 5 ...+... 6 ...+... 7