DSXchange

My answer assumes they actually do need to 'loop' because the files need to be processed individually to completion in the proper order.

BradMiller wrote:I am setting APT_STRING_PADCHAR to "0*20".

Surely you mean "0x20", yes?

You need to find the decimal value of that character and then use CHAR(xxx) in your Convert() function rather than the actual character.

I... don't believe there is one.

Let's see what others have to say on the subject.

Build a job to redact/mask the PHI elements and create a new file. Verify new file has the same issue. Send new file to IBM Support.

After the sort.

Have you looked here yet?

They're already doing the NEXTVAL part upstream... hopefully there's a good reason for that, parent / child load or some such. Still won't fix the fundamental binding problem, however.

Doesn't "your client" have support? If not, then the whole question of patches is moot.

You'd need to post examples of your actual data for anyone to provide more cogent help.

Arnd, your syntax example is... lacking.

Those really should be questions to your official support provider.

Ray already did. The stage wants at least one column defined and yet you need to leave the ID field out of the stage since it won't be bound to anything.

No. The client perhaps but only that.

Be careful with all of the "drop all indexes and constraints" suggestions. Not always as easy as it sounds (RI) and can be problematic if any other process is accessing your table while it is being loaded. Never mind the fact that it is not guaranteed to be any kind of faster overall, you'...

Yes, use the Loop stages in a Sequence job to run your file loading job. Use the UserVariables stage to retrieve the filenames in the proper order in a delimited format.