Hi @mzuer​ 

It all depends upon how many conditions you have.

You could quite easily setup a Tester, make it Case Sensitive or Insensitive depending upon your requirements.

Filter the attributes accordingly, then use String Replacer or Bulk Attribute Replacer as necessary:

The example above, I created an attribute called AttributeX, with a value of Starter and starter.

I filtered it via the Tester, make sure it was Case Sensitive, to show if it begins with S then using String Replacer, I replaced Start with End.

I'm not sure i'd ever try to do it this way, but i think in theory it should be possible. If the prefix is abc, remove all attributes that start with abc and send this feature one direction, add an abc prefix to a unique id, then remove all attributes that don't start with abc. Convert all remaining attributes to uppercase, featuremerge to aggregate the features into a single feature with all attributes.

I think this is what you mean:

All features go trough the top and bottom and are merged to themself at the end.

Bottom features keep the lowercase attributes, top features keep the uppercase attributes that are converted to uppercase.

Ha! we all had the same kind of idea. Amazing how annoying it was when it would have been much easier to just be able to specify a regex to determine which attributes should upper/lowercase.

It was what I expected to do in my first try.

Also tried @Uppercase(@CurrentAttribute()) in the Rename. But this gave an Attribute with that name.

not as straightforward as I imagined it could be possible to be, but thanks for sharing this solution

Thanks for the information