I have converted delimited files into nice mainframe centric fixed
position datasets. Here's an skeleton of an example of taking a file
that came from EXCEL as TXT tab delimited and became a friendly dataset.
In this case I wanted each field to be 30 bytes just for my visual ease
but each field can be whatever you think its max length might be.
//SORTIN DD *
SORT FIELDS=COPY
OUTFIL PARSE=(%01=(ENDBEFR=X'05',FIXLEN=30), COL.A
%02=(ENDBEFR=X'05',FIXLEN=30), COL.B
%03=(ENDBEFR=X'05',FIXLEN=30), COL.C
%04=(ENDBEFR=X'05',FIXLEN=30), COL.D
%05=(ENDBEFR=X'05',FIXLEN=30)), COL.E
BUILD=(%01,5X,%02,5X,%03,5X,%04,5X,%05)
LRECL will be sum of (30x5) + 20 blanks.
The dates manipulation I have to defer to Sri.
On 7/1/2015 5:33 PM, Sri h Kolusu wrote:
For some reason this topic did not show up in my mail box but is available
here
https://groups.google.com/d/msg/bit.listserv.ibm-main/p4tGxAcXyPQ/uyoz0k7_y4oJ
June,
DFSORT can definitely convert the comma delimited file into a fixed length
file. As for the conversion of the dates and padding you need to have a
second parse to get the values from ZM/ZD/YYYY birth date that needs to be
converted to MM/DD/YYYY .
Send me a sample of your input and desired output as text files to our
hotline at [email protected] or send it to me directly and I will show you
a way to do it.
Thanks,
Kolusu
DFSORT Development
IBM Corporation
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
