Prev: Mainframe Question
Next: Search for a string using SORT
From: Michael Mattias on 5 Feb 2006 10:52
<pottmi(a)gmail.com> wrote in message
news:1139101225.085526.164350(a)g44g2000cwa.googlegroups.com...
>
> Back to the problem...
>
> I am trying to read the data using existing COBOL programs without
> modifing the existing COBOL programs. I can do this as long as I
> convert the data to the microfocus format (w/128 byte header).

This truly IS a problem.

You have an existing program designed to read data provided in a certain
format; you have data in another format; there is no way you are going to
get this to work. This is a dream and you WILL wake up.

You need to either convert the data to the expected format, or change the
format in which the existing program expects the data.

Others have made some good suggestions to do the former; that's that path
down which I'd set out.

MCM



From: pottmi on 5 Feb 2006 17:57
I know that I can convert the data to make the problem "go away", in
fact I am already doing that as an (hopefully) interim measure. Adding
the conversion step increases IO and I want to avoid the IO by having
the COBOL program read the data without a conversion.

I tried to make that clear in my original post so no one would have to
waste their time answering a question that I did not ask.

By conversion, I mean the format of the meta data in the sequential
file, not the data itself. By meta data I mean the 128 byte header
record that microfocus requires, the header on each record that
indicates length, and sometimes padding bytes to bring new records to
the appropriate byte boundary.

--
potter

From: pottmi on 5 Feb 2006 18:04
Thanks Colin,

I think I will stick with my own conversion program as I will not have
to run anything on the mainframe to make it work. Plus, I am using
microfocus SE on AIX.

Your "Final note" is quite helpful I will interested in finding out
what those extra bytes are used for. I have alway been curious, but
never had a need to figure it out.

From: pottmi on 5 Feb 2006 18:12
I am not really considering changing the COBOL programs, there are
about 400 of them reading about 400 different sequential files. My
worst case right now is having to run my conversion program on the data
before the program runs.

Never the less, I am interested in your sample programs. I will send a
separate email to you so all you have to do is reply.

From: on 5 Feb 2006 21:40
In article <1139180224.894240.81710(a)g44g2000cwa.googlegroups.com>,
<pottmi(a)gmail.com> wrote:
>I know that I can convert the data to make the problem "go away", in
>fact I am already doing that as an (hopefully) interim measure. Adding
>the conversion step increases IO and I want to avoid the IO by having
>the COBOL program read the data without a conversion.

1) What is the anticipated transaction volume involved (millions of
records per unit time)?

2) What controls exist to insure that the format does not change, again,
as it is changing now?

>
>I tried to make that clear in my original post so no one would have to
>waste their time answering a question that I did not ask.

The fact that you did not ask it does not necessarily generate the
conclusion that the question you did not ask is irrelevant.

DD
First  |  Prev  |  Next  |  Last
Pages: 1 2 3
Prev: Mainframe Question
Next: Search for a string using SORT