Getting data to the IFS

Looks like a lot of us are using the IFS! :)

What I'm doing:

We use a package that creates a text "flat file" that is used for EDI.
In creating this file, the package just slaps all the fields from 2
other files into this file as such:

Header record---------------------------
Detail record-----------------
Detail record-----------------
Header record---------------------------
Detail record-----------------

Etc... The file is defined to be the length of the Header file record
format (it's the longest of the two.)

When I try to view this file using SQL, some data is not viewable. This
appears to be some of the numeric data, either defined as Packed or
Signed.

I use CPYTOIMPF to get this "flat file" to the IFS. My assumption was
that the EDI tool we use (not iSeries) would be able to convert the
fields itself, but as we are finding out, it can't. Numbers are wrong,
in some places, we see extra linefeeds that throw the edi translator
off.

I think that my CPYTOIMPF is not working correctly because the fields
have already been written "badly" into text format.

I'm looking for suggestions on how to handle this. Do I need to
translate the fields back to their original formats and then recreate
the text string? This would suck, because there are a LOT of fields.

While researching this, I've looked all over. I've found suggestions on
how to use CPYTOIMPF and CPYTOSTMF, but nothing that has yet to help me.
Any help is GREATLY appreciated!

Also, I'm travelling this week and my response time might be a bit slow.


TIA, Mark