IBM Mainframe Forum

by **akanchu** » Wed Jan 13, 2010 11:10 am

I have a file that about 400 bytes of data.
Few fields in this file has data in the following format.

1799 ARBYDALE LN
FFFF4CDCECCDC4DD444444444444444444444444444444444444444444
1799019284135035000000000000000000000000000000000000000000
----------------------------------------------------------
1234 HILTON WHITEMALL RD
FFFF4CCDEDD4ECCECDCDD4DC4444444444444444444444444444444444
1234089336506893541330940000000000000000000000000000000000
----------------------------------------------------------
2025 S DUMPBELL AVE APT 317
FFFF4E4CEDDCCDD4CEC4CDE4FFF4444444444444444444444444444444
2025020444725330155017303170000000000000000000000000000000

I have to convert these trailing spaces of the field to low values.

Select all

1799 ARBYDALE LN..........................................
FFFF4CDCECCDC4DD000000000000000000000000000000000000000000
1799019284135035000000000000000000000000000000000000000000
---------------------------------------------------------
1234 HILTON WHITEMALL RD..................................
FFFF4CCDEDD4ECCECDCDD4DC0000000000000000000000000000000000
1234089336506893541330940000000000000000000000000000000000
---------------------------------------------------------
2025 S DUMPBELL AVE APT 317...............................
FFFF4E4CEDDCCDD4CEC4CDE4FFF0000000000000000000000000000000
2025020444725330155017303170000000000000000000000000000000

The length of different fields vary.
eg the above is an address line, similarly there could be a name in the same record, which also needs to fixed.

can we handle this via sort or fileaid

Thanks in advance.
akanchu

by **Alissa Margulies** » Wed Jan 13, 2010 10:49 pm

Hello Akanchu.

Here is a sample SyncSort for z/OS 1.3.2 job tht will do what you asked:

Select all

//S1 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//SORTOUT DD DSN=OUTPUT.FILE
//SORTIN DD *
1799 ARBYDALE LN
1234 HILTON WHITEMALL RD
2025 S DUMPBELL AVE APT 317
//SYSIN DD *
SORT FIELDS=COPY
OUTFIL IFTHEN=(WHEN=INIT,FINDREP=(INOUT=(X'4040',X'0000'))),
IFTHEN=(WHEN=(80,1,CH,EQ,X'40'),OVERLAY=(80:X'00'))
/*

by **akanchu** » Thu Jan 14, 2010 12:20 am

Thanks Alissa.

just ran into trouble with that because we have "SYNCSORT FOR Z/OS 1.3.1.0R"
so it errored out on FINDREP... so close.

just out of curiosity and to understand FINREP :
if there is data starting at position 80, what will need to be done?
eg say like

Select all

1799 ARBYDALE LN Miami
1234 HILTON WHITEMALL RD Niagara
2025 S DUMPBELL AVE APT 317 Buffalo

Also does FINDREP take care of the odd number(say 11 or 13) spaces.

Thanks

by **Alissa Margulies** » Thu Jan 14, 2010 1:03 am

If you only want to change the trailing blanks to low values in the first field, then you can code the following:

Select all

//SYSIN DD *
SORT FIELDS=COPY
OUTFIL IFTHEN=(WHEN=INIT,FINDREP=(INOUT=(X'4040',X'0000'),
STARTPOS=1,ENDPOS=79)),
IFTHEN=(WHEN=(79,1,CH,EQ,X'40'),OVERLAY=(79:X'00'))

However, if you want to change the trailing blanks in both fields, the code can be modified as follows (assuming the second field is 10 bytes):

Select all

//SYSIN DD *
SORT FIELDS=COPY
OUTFIL IFTHEN=(WHEN=INIT,FINDREP=(INOUT=(X'4040',X'0000'))),
IFTHEN=(WHEN=(79,1,CH,EQ,X'40'),OVERLAY=(79:X'00'),HIT=NEXT),
IFTHEN=(WHEN=(89,1,CH,EQ,X'40'),OVERLAY=(89:X'00'))
/*

by **akanchu** » Thu Jan 14, 2010 11:44 am

Thanks Alissa.
So with this option, I will have to know the length of each field that may have trailing spaces, to take care of the last byte.
Do you think there is a work around to ignore the check for the last byte.
I guess, will have to do another round of sort to take care of them.

Since I cannot use the FINDREP option, I will wait for any input from anybody else.

Thanks for your help.

by **dick scherrer** » Fri Jan 15, 2010 12:15 am

Hello,

If you need this any time soon, you could actually write a bit of code - which might take 15 minutes. . .

IBM Mainframe Forum

Conversion from trailing spaces of text to low-values

Conversion from trailing spaces of text to low-values

Re: Conversion from trailing spaces of text to low-values

Re: Conversion from trailing spaces of text to low-values

Re: Conversion from trailing spaces of text to low-values

Re: Conversion from trailing spaces of text to low-values

Re: Conversion from trailing spaces of text to low-values