IBM Mainframe Forum

by **indra_nsec** » Fri Feb 08, 2013 1:00 am

Hi,

I am trying to read a flat file which has been formed from a csv file and populate a data area

eq,

csv file as below

abc,1.2,rr,,7.89,qqq

flat file is formed as

abc00001.2rr00007.8qqq

now i need above data to filled in data area

as fields like A(3),N(3.3),A(2),N(4.2),A(3)

is it possible anyway?

by **RGZbrog** » Mon Feb 11, 2013 9:55 pm

For Natural under Windows or Unix it's a simple matter of reading the CSV directly. On the mainframe the CSV must be converted to a flat file, as you have done, but the format could be simplified to make your job much easier, because the record definition must match the data byte for byte.

If you changed the record format to abc001200rr000780qqq you would have the simplest solution.

Select all

DEFINE DATA LOCAL
1 #IN
2 #A1 (A3)
2 #N1 (N3.3)
2 #A2 (A2)
2 #N2 (N4.2)
2 #A3 (A3)
END-DEFINE
*
READ WORK 1 #IN
DISPLAY #A1 #N1 #A2 #N2 #A3
END-WORK
END

In the data, note the implied decimal points and the number of digits before and after the decimal.

Because the record includes the decimal point and fewer decimal positions

Select all

DEFINE DATA LOCAL
1 #IN
2 #A1 (A3)
2 #N1 (A7)
2 #A2 (A2)
2 #N2 (A7)
2 #A3 (A3)
1 #OUT
2 #N3 (N5.1)
2 #N4 (N5.1)
2 #N5 (N3.3)
2 #N6 (N4.2)
END-DEFINE
*
READ WORK 1 #IN
MOVE EDITED #N1 TO #N3 (EM=99999.9)
MOVE EDITED #N2 TO #N4 (EM=99999.9)
MOVE #N3 TO #N5
MOVE #N4 TO #N6
DISPLAY #A1 #N1 #N3 #N5 #A2 #N2 #N4 #N6 #A3
END-WORK
END

You need to remove the decimal (MOVE EDITED) and then convert to the proper number of digits (MOVE/ASSIGN).

by **indra_nsec** » Wed Feb 13, 2013 12:06 am

Hi,

Thanks for the answer.
Infact thats what I also decided to do later on ........

by **boby125** » Mon Sep 15, 2014 10:27 am

Text removed as it was copied from RGZbrog's reply, above.

by **enrico-sorichetti** » Mon Sep 15, 2014 11:15 am

On the mainframe the CSV must be converted to a flat file, as you have done, but the format could be simplified to make your job much easier, because the record definition must match the data byte for byte.

any reason to clone a previous reply ???

IBM Mainframe Forum

reading flat file and populating data area

reading flat file and populating data area

Re: reading flat file and populating data area

Re: reading flat file and populating data area

Re: reading flat file and populating data area

Re: reading flat file and populating data area