DSXchange

No, it's either one job or all jobs. You'll need another solution, perhaps something like the DSX Cutter posted here or an iterative loop, for category exports.

It's always a good idea to post complete and unedited error messages rather than something that may or may not be... close.

No because this kind of backup needs to happen on the client, hence the Windows batch file. There are more options available now if you are on the 8.x release, however.

Lots of info out there on this, for example here. Scroll down to the 5. References section in particular the Character Predeclared Entity bit.

Next time, don't post it again, simply go back and edit your post for clarity. Nothing here is cast in stone.

Don't make people figure out what you need from just an example. Either show us the case statement or explain in words the transformation needed, the business rule.

Your example makes no sense on its own.

But i am not very clear with the answer.Are you saying 8.1 version uses both Universe repository and Oracle repository. I thought 8.1 no longer uses universe database. You thought wrong. :wink: As I noted, the 8.x release uses both repositories - so everything is still in the 'Universe' repository ...

I don't think anyone here would recommend an 'upgrade in place' from 7 to 8, although that was common (and painless) in previous versions. General advice is to create a new, clean 8.x installation and then import your project from the previous release into it. That regardless of the fact of location...

The 8.x versions use both repositories, you have no choice in the matter, nor do you have to worry about transferring anything between the two as it will manage all that automagically during an upgrade or import.

You've done all you can at this point. As far as I know, that access grant is still a manual process and seeing how this is a weekend and today Father's Day to boot, I would guess it won't happen until tomorrow at the earliest.

Have you been through the docs on the SCD stage? There's a chapter for it in the Parallel Job Developer Guide pdf and a section in that chapter specific to 'Surrogate keys in a Slowly Changing Dimension stage'. It should explain how to tell the stage what your Surrogate Key is, either through the Pu...

A minor detail you failed to mention in the original post. I would suggest using the Field() function like you were for a single character, say the pipe as the field delimiter, and then strip the extraneous second character from the field. That or making a pass to convert those pairs to a single cha...

In all honesty, Arnie, I'm a wee bit confused as well. We started off talking about using an IDENTITY column and now you are generating the sequence numbers yourself rather than letting the database do it for you? Is there any particular reason for that change? I'm also unclear how you implemented t...

Just set the stage property to Delimiter String rather than Delimiter (as in single character) and give that value.

Still, why the worry about uniqueness in this application? All you care about is if the MD5 hash for any given record changes over time and who cares if the hash it carries is the same as any other record? Or are you thinking that the field values could change to some odd combination of values that ...