Delete duplicate records

[Newbie2u]

Have a huge file that with record length around 366.
One field colume 1-20 has the name of a ticket in it. If the name of the ticket appears more than once in the file I want to delete all occurances but one. I know this seems fairly easy but I cannot seem to get anything to work. Can someone point me in the right direction. Thank you so much.

 

[Tzu]

Hello Newbie2u,

You can just use a SORT jcl like :

//*
//*===================================================================
//*
//* EXCLUDE DOUBLES IN CONTROL FILE.
//*
//*===================================================================
//SORT EXEC PGM=SORT
//SORTIN DD DSN=Your.File.IN,DISP=OLD
//SORTOUT DD DSN=Your.File.OUT,UNIT=DISK,VOL=SER=XXXXXX,
// SPACE=(CYL,(150,200),RLSE),DISP=(NEW,CATLG),
// DCB=(LRECL=10004,BLKSIZE=20000,RECFM=VB)
//SORTWK01 DD UNIT=3390,SPACE=(CYL,(500,50)),VOL=SER=XXXXXX
//SORTWK02 DD UNIT=3390,SPACE=(CYL,(500,50)),VOL=SER=XXXXXX
//SYSOUT DD SYSOUT=*
//STATOUT DD SYSOUT=*
//SYSIN DD *
SORT FIELDS=(1,20,CH,A),EQUALS
RECORD TYPE=VB,LENGTH=10004
SUM FIELDS=NONE
/*

Regards,

Tzu.

 

[KiwiREXXDude]

If you wanted a simple REXX solution you could try putting the value into a STEM and then checking for the existance of the stem. For example (assuming LINES. is your input);

NewCnt = 0
drop DONE. NEWLINE.
do I = 1 to LINES.0
ticket = left(LINES.I, 20)
if Value('DONE.'ticket) == 1 then iterate I
DONE.ticket = 1
NewCnt = NewCnt + 1
NEWLINE.NewCnt = LINES.I
end