IBM Mainframe Forum

I'm going to combine 3 datasets with the SORT step. On succesful execution of this job I am able to see the total record count of the output file (which is of all 3 I/P files) in the spool. But I would like to know is there any possiblity to get the record count of each input file, and the same should be displayed in the spool. Is there is any method to log this?

Thanks,
Vinoth R

Run JCL like this --
There should be a blank character before the REPRO command, which this dumb board deleted. The DCB parameters on the OUTPUT DD statement must match, within reason, the actual DCB parameters of the input dataset/

You will get output like this -
There was a blank character before the REPRO command, ...

Vinoth R,

It's not clear what you're trying to do. If you are asking can you use copy three concatenated data sets to an output data set in one pass AND get the count of each individual input data set, then the answer is no (I don't believe IDCAMS can do that either). For concatenated data sets, DFSORT wouldn't know where each data set starts and ends.

If you want to create an output data set consisting of three input data sets and you want a count of each input data set, you could do that with ICETOOL in one step with a COPY operator for each input data set and a MOD output data set, e.g.

COPY FROM(IN1) TO(OUT)
COPY FROM(IN2) TO(OUT)
COPY FROM(IN3) TO(OUT)

where OUT is a MOD data set. You would get messages in DFSMSG giving the record count for each input file.

If you want to do something else, you need to explain more clearly exactly what you want to do. Show an example of your input files and expected output.

Actually this is my JCL
XXPS010 EXEC PGM=SORT
XXSORTIN DD DUMMY,
XX DCB=(RECFM=FB,LRECL=344,BLKSIZE=0,BUFNO=5)
X\SORTIN DD DSN=xxx00000.xxxxxxx.G0004V00,
X\ DISP=SHR,DCB=BUFNO=50
X/SORTIN DD DCB=(BUFNO=50,RECFM=FB,LRECL=344,BLKSIZE=0),DISP=SHR,
X/ DSN=xxx00000.xxxxxxx.G0004V00
X@ DD DSN=xxx00000.xxxxxxx.G0005V00,
X@ DISP=SHR,DCB=BUFNO=50
SORTOF02 DD DSN=XXX00000.xxxxxxxx.BKUP(+1),
DISP=(NEW,CATLG,DELETE),

I need to print the record count of the inout files "xxx00000.xxxxxxx.G0004V00 and xxx00000.xxxxxxx.G0005V00" in spool. Is it possible with SORT utility? Please advice

What did you not understand about Frank's answer? When sort runs it puts diagnostics of the run to SYSOUT on the spool. Theses diagnostics include count of records read and records written. However, if you concatenate the files SORT cannot tell where one file ends and the next starts. But if you use the COPY function against the individual files then you will get the counts.

I am using SORT FIELDS=COPY. Is this right?

Wrong - you should be using the COPY function of DFSORT - not the SORT function (nor the MERGE function). Check out the manual.

you should be using the COPY function of DFSORT

SORT FIELDS=COPY does perform the copy function. SORT FIELDS=(p,m,f,s,...) would perform the SORT function.

Of course, you don't need any SORT statement with the COPY operator of DFSORT's ICETOOL since it does a COPY function by default.