Hi Scott,
First, welcome to the JDEList Forums!
Just some addition after Philippe's reply.
I prefer to use Fixed Field Length records instead of comma (or other) delimited in the input (and output too) file and I use our "flexible" data type conversion Business Function to convert string parts into date or number, so this way I can handle all type of errors, e.g. invalid number or date too.
In the past, I also used Table Conversion but it placed the flat file only into a workfile and made only some preparation and I placed the most part of our logic into a "main" UBE where I could write more flexibly ER logic and make complicated processes and produce a report about the input as well about errors and statistical data.
Recently I have created 3 new interfaces (2 output + 1 input) and I omitted to use Table Conversion and I used Business Functions to Open, Read, Write and Close the flat file. This is a bit flexible and simple solution, because:
* I do not have to setup OneWorld to access the Flat File (ODBC Data Source, OneWorld Data Source, OneWorld Environment to acces flat files, etc.)
* there are easy ways to deal with the file names and locations
On the other hand, in the mentioned input interface, I used first time the transaction processing with succes to perform a "nothing or all" input.
Hope, could help a bit.
Regards,
Zoltán
P.S.1: I (and many of us) have already sent a lot of post in the flat file and interface themes. Please, check the archives if you are interested in it.
P.S.2.: Next times please, attach your systmconfiguration information too. Thanks.
B7332 SP11, ESU 4116422, Intel NT4, SQL 7 SP1
(working with B7321, B7331, XE too)