|
View previous topic :: View next topic
|
| Author |
Message |
pushpagiri
New User
.jpg)
Joined: 07 Jul 2005
Posts: 51
|
|
|
|
Hi,
This is my problem :
Expanation:
I want to remove the character '$' from each line of file1.
But while doing so there should be any space introduced( instead of '$')
file1:
| Code: |
aaaa bbbbb$$$$$$xxxxx xx$$$$
bbb bbbb$$$$$$$$$ccccc cccccc$$
|
Outfile:
| Code: |
aaaa bbbbbxxxxx xx
bbb bbbbccccc cccccc
|
Please guide me to some solution.
Thanks in advance. |
|
| Back to top |
|
 |
priyesh.agrawal
Senior Member
Joined: 28 Mar 2005
Posts: 1448
Location: Chicago, IL
|
|
|
|
Give this command at command line in EDIT MODE....C $ '' ALL ...
Regards,
Priyesh. |
|
| Back to top |
|
|
priyesh.agrawal
Senior Member
Joined: 28 Mar 2005
Posts: 1448
Location: Chicago, IL
|
|
|
|
The command above will delete all the occurrences of $ from the file....
If you want to insert an SPACE in place of $... give this..C $ ' ' ALL..
Regards,
Priyesh. |
|
| Back to top |
|
|
pushpagiri
New User
Joined: 07 Jul 2005
Posts: 51
|
|
|
|
Hi Priyesh,
Is it possible through any JCL.Because this is a step required in a process
which should be done through JCL.
Also I want the output in a dataset. |
|
| Back to top |
|
|
priyesh.agrawal
Senior Member
Joined: 28 Mar 2005
Posts: 1448
Location: Chicago, IL
|
|
|
|
Thru INREC statement with CHANGE option it should be possible to achieve what you want in DFSORT's way.
Is $ occurring at any specific location...else it would be bit difficult checking each & every character in the file...Frank will be knowing the better way...
Well, What is REC LENGTH & format of your Input file...
Regards,
Priyesh. |
|
| Back to top |
|
|
pushpagiri
New User
Joined: 07 Jul 2005
Posts: 51
|
|
|
|
hi,
It is FB of length 160.
Yes priyesh.The character '$' can occur at any position in the input file. |
|
| Back to top |
|
|
Frank Yaeger
DFSORT Developer
Joined: 15 Feb 2005
Posts: 7129
Location: San Jose, CA
|
|
|
|
Push,
DFSORT's TRAN=ALTSEQ feature can be used to replace each $ with a blank. But I believe you want to actually remove the $ characters and shift the other characters to the left - is that correct? |
|
| Back to top |
|
|
pushpagiri
New User
Joined: 07 Jul 2005
Posts: 51
|
|
|
|
Frank,
Sorry for the small mistake in my post:
{Expanation:
I want to remove the character '$' from each line of file1.
But while doing so there should be any space introduced( instead of '$')
}
Correction:
But while doing so there should not be any space introduced( instead of '$')
Yes Frank.I want to remove $ and shift other characters to left.
But with ALTSEQ the characters will not be shifted and spaces will be
inserted.
Is there any way to implement the method of Priyesh though any JCL.
(CHANGE ALL '$' '').Can this be done through any of these -DFSORT,ICEMAN,ICETOOL? |
|
| Back to top |
|
|
priyesh.agrawal
Senior Member
Joined: 28 Mar 2005
Posts: 1448
Location: Chicago, IL
|
|
|
|
Hi Push,
To solve your problem by JCL way, here is another solution.
REXX SCRIPT....
| Code: |
ADDRESS TSO
'PROFILE NOPREFIX'
SAY '- BEGIN ---------------------------------'
"EXECIO * DISKR FILELIST (STEM FILELIST. FINIS"
DO ILIST = 1 TO FILELIST.0
I = 0
TABTMP. = ''
TABTMP.0 = 0
FILELIST.ILIST = STRIP(FILELIST.ILIST,T)
SAY ' FILE IN PROGRESS: ' FILELIST.ILIST
"ALLOC DA("FILELIST.ILIST") F(FILEIN) SHR"
"EXECIO * DISKR FILEIN(STEM TABTMP. FINIS"
"FREE F(FILEIN)"
"ALLOC DA("FILELIST.ILIST") F(FILEOUT) OLD"
/* DEFINE STRING TO BE CHANGED */
CHANGED = 0
PAR1 = '$'
PAR2 = ''
CALL CHANGE
"EXECIO * DISKW FILEOUT(STEM TABTMP. FINIS"
"FREE F(FILEOUT)"
END
SAY '- END -----------------------------------'
EXIT RC
CHANGE:
DO I = 1 TO TABTMP.0
IF INDEX(TABTMP.I,PAR1) > 0 THEN DO
CHANGED = CHANGED + 1
STRAPP = DELSTR(TABTMP.I,INDEX(TABTMP.I,PAR1),LENGTH(PAR1))
TABTMP.I = INSERT(PAR2,STRAPP,INDEX(TABTMP.I,PAR1)-1)
END
END
RETURN |
Job to run this script
| Code: |
//STEPREXX EXEC PGM=IKJEFT01,PARM='MEMBERNAME'
//SYSEXEC DD DISP=SHR,DSN=PDS.HAVING.MEMBER.OF.REXX,
//SYSTSPRT DD SYSOUT=*
//SYSTSIN DD DUMMY
//FILELIST DD *
INPUT.NAME.DSN1
//* |
You need to run this job step in order to remove $ from the file.
Put this REXX Script in a PDS. member name to be specified in PARM parameter.
Check & let us know the results.
Regards,
Priyesh. |
|
| Back to top |
|
|
pushpagiri
New User
Joined: 07 Jul 2005
Posts: 51
|
|
|
|
Hi priyesh,
Actually I don't have much knowledge in REXX.Am a <beginner in that.
I tested your code.But the output remains the same.
I don't know where it got problem.
Am trying to undersatnd your code.
| Code: |
STRAPP = DELSTR(TABTMP.I,INDEX(TABTMP.I,PAR1),LENGTH(PAR1))
TABTMP.I = INSERT(PAR2,STRAPP,INDEX(TABTMP.I,PAR1)-1) |
Is there any need to insert PAR2 after deleting the PAR1?
Am not sure about it.
(but thankz for your great effort) |
|
| Back to top |
|
|
priyesh.agrawal
Senior Member
Joined: 28 Mar 2005
Posts: 1448
Location: Chicago, IL
|
|
|
|
Hi Push,
Actually this code is supposed to change PAR1 to PAR2...
For which I kept PAR1='$' & PAR2='' (means nothing for PAR2).
So no need to insert PAR2 seperately...
| Quote: |
| (but thankz for your great effort) |
I forgot to mentione earlier... this to be lead to MGIndaco, who earlier provided me this REXX Code for one of my problem.
I been the communicator in this case.
Regards,
Priyesh. |
|
| Back to top |
|
|
priyesh.agrawal
Senior Member
Joined: 28 Mar 2005
Posts: 1448
Location: Chicago, IL
|
|
|
|
Push,
I just now tested & it is working fine for me.
Regards,
Priyesh. |
|
| Back to top |
|
|
Frank Yaeger
DFSORT Developer
Joined: 15 Feb 2005
Posts: 7129
Location: San Jose, CA
|
|
|
|
| Quote: |
| Can this be done through any of these -DFSORT,ICEMAN,ICETOOL? |
Sort products do not have any built-in features to do this kind of thing.
FYI, PGM=ICEMAN is just a way to invoke your sort product (e.g. DFSORT). |
|
| Back to top |
|
|
pushpagiri
New User
Joined: 07 Jul 2005
Posts: 51
|
|
|
|
Hi Priyesh,
I have solved it finally( but not without your help)
The REXX code changed the first occurance of '$' only.
So the following IF should be changed to DO WHILE.
| Code: |
IF INDEX(TABTMP.I,PAR1) > 0 THEN DO
CHANGED = CHANGED + 1
STRAPP = DELSTR(TABTMP.I,INDEX(TABTMP.I,PAR1),LENGTH(PAR1))
TABTMP.I = INSERT(PAR2,STRAPP,INDEX(TABTMP.I,PAR1)-1)
END |
| Code: |
DO WHILE INDEX(TABTMP.I,PAR1) > 0
CHANGED = CHANGED + 1
STRAPP = DELSTR(TABTMP.I,INDEX(TABTMP.I,PAR1),LENGTH(PAR1))
TABTMP.I = INSERT(PAR2,STRAPP,INDEX(TABTMP.I,PAR1)-1)
END |
Thank you very much Priyesh(and MGIndaco too).Without your help,I couldnot have solved this problem.
Thanks and Regards,
Push |
|
| Back to top |
|
|
Frank Yaeger
DFSORT Developer
Joined: 15 Feb 2005
Posts: 7129
Location: San Jose, CA
|
|
|
|
With z/OS DFSORT V1R5 PTF UK90007 or DFSORT R14 PTF UK90006 (April, 2006), we can now do this kind of thing with DFSORT. This particular case is a bit tricky because it has '$' characters that should be removed and multiple blanks that should not be removed. But here's a DFSORT job that will do it for the RECFM=FB/LRECL=160 input data set:
| Code: |
//S1 EXEC PGM=ICEMAN
//SYSOUT DD SYSOUT=*
//SORTIN DD DSN=... input file (FB/160)
//SORTOUT DD DSN=... output file (FB/160)
//SYSIN DD *
OPTION COPY
ALTSEQ CODE=(40FF,FF40)
* Replace blanks with X'FF's to protect them.
INREC IFTHEN=(WHEN=INIT,
BUILD=(1,161,TRAN=ALTSEQ)),
* Change '$'s to blanks so squeeze can remove them.
IFTHEN=(WHEN=INIT,
BUILD=(1,161,SQZ=(SHIFT=LEFT,PREBLANK=C'$'))),
* Replace X'FF's with blanks to get the blanks back.
* Note that blank padding at end of each record will be
* replaced with X'FF's (we can't avoid this).
IFTHEN=(WHEN=INIT,
BUILD=(1,161,TRAN=ALTSEQ)),
* Extract the bytes before the first X'FF' into the
* 160-byte %nn field (this will get the original bytes
* padded with blanks, not X'FF's).
IFTHEN=(WHEN=INIT,
PARSE=(%00=(ENDBEFR=X'FF',FIXLEN=160)),
* Put the %nn field into the output record. It will
* have blanks in the right places.
BUILD=(%00))
/*
|
This is what the data looks like at each stage with a '-' shown for each X'FF' to make it easier to understand and the record truncated where it doesn't matter:
After first WHEN=INIT with BUILD:
| Code: |
aaaa--bbbbb$$$$$$xxxxx---xx$$$$--------------------
bbb--bbbb$$$$$$$$$ccccc----cccccc$$----------------
|
After second WHEN=INIT with BUILD:
| Code: |
aaaa--bbbbbxxxxx---xx--------------------
bbb--bbbbccccc----cccccc----------------
|
After third WHEN=INIT with BUILD:
| Code: |
aaaa bbbbbxxxxx xx ----------
bbb bbbbccccc cccccc -----------
|
After fourth WHEN=INIT with PARSE and BUILD:
| Code: |
aaaa bbbbbxxxxx xx
bbb bbbbccccc cccccc
|
For complete details on all of the new DFSORT and ICETOOL functions available with the April, 2006 PTFs, see:
Use [URL] BBCode for External Links |
|
| Back to top |
|
|
|
|
