IBM Mainframe Forum

Hi,

I am in a situation where i need to mask some data.

Say, input data is like
000001 ABCDEF____NEW______DAY
000002 GHIJ______DEVELOPT__NIGHT
000003 KLMNO____OLD_______EVENING
000004 PQRSTUVW_MEDIUM___MORNING
000005 XYZ_______SAFE______AFTERNOON

Output required is like
000001 AXXXXX____NYY_______DTT
000002 GXXX______DYYYYYYY__NTTTT
000003 KXXXX_____OYY_______ETTTTTT
000004 PXXXXXXX__MYYYYY____MTTTTTT
000005 XXX_______SYYY______ATTTTTTTT

[I have added dashes here to get the correct alignment of data. The file is FB and LRECL is 30. There are 3 fields of length 10 each in the file. The numbers at the beginning are only line numbers].


That is, other than the first letter of each field, all the other characters need to be changed to a pre-decided character, here 'X', 'Y' and 'T'. I tried using ALTSEQ and succeeded. However, the trouble is am able to use just one ALTSEQ. Is there any way i can use multiple ALTSEQ, one to replace each field ?? Or what is the other way of doing this ??

I would be glad if you can assist me on this.

Thanks,
Sam.

samurai007 wrote:I have added dashes here to get the correct alignment of data.

More work than it's worth, and it's still sloppy...

Say, input data is like
000001 ABCDEF____NEW______DAY
000002 GHIJ______DEVELOPT__NIGHT
000003 KLMNO____OLD_______EVENING
000004 PQRSTUVW_MEDIUM___MORNING
000005 XYZ_______SAFE______AFTERNOON

Output required is like
000001 AXXXXX____NYY_______DTT
000002 GXXX______DYYYYYYY__NTTTT
000003 KXXXX_____OYY_______ETTTTTT
000004 PXXXXXXX__MYYYYY____MTTTTTT
000005 XXX_______SYYY______ATTTTTTTT

Why didn't you just wrap 'code' around it and save yourself the time?

Sam,

Which characters are you trying to replace? Is it any non-blank, or just A-Z, or something else?

@ CICS Guy

Why didn't you just wrap 'code' around it and save yourself the time?

I didnt know this. Will use it going forward. Thanks.
But this was not the problem to which i needed a solution.

@ Frank

Which characters are you trying to replace? Is it any non-blank, or just A-Z, or something else?

For every field, we have a character between [A-Z] assigned to it.
Suppose a field, NAME, has the value as 'FRANK' and if the character assigned to it is 'S', i would like the field to appear as 'FSSSS' in my output file.
Suppose a field, COUNTRY, has the value as 'AMERICA' and if the character assigned to it is 'M', i would like the field to appear as 'AMMMMMM' in my output file.

In essence, other than the first letter of each field, all the other non-blank letters in the field have to be replaced with the pre-defined character. This pre-defined character varies from field to field, say, for NAME it is 'S', for COUNTRY it is 'M' and so on.
Now if you can compare this with the data I gave you initially, i believe you would get a fair idea of how each field would appear at the output.

Using an OVERLAY in the OUTREC replaces ALL the letters in the field, including blanks, with the pre-defined character. I would like the blanks to remain unchanged.
Using ALTSEQ allows me to replace the contents of the entire file with only one character. I would want to replace each field with its corresponding pre-defined character.


Let me know if this is sufficient explanation.

Thanks,
Sam.

You can use a DFSORT job like the following to do what you asked for: