But I really am looking to split by whitespace, with an unknown of how many tokens in a label.  My example of human names was just to simplify, but could be anything... not just human names.  Any Wikidata QID.

I would want 6 columns automatically created for that. Or in JSON terms.. An array of string objects.

{

"Castle",

"of",

"Saint",

"Pée",

"sur",

"Nivelle", 

}

This has to do with a use case of pre-processing the label names for data ingestion into further analysis workflows.

I was hoping that I could easily leverage a bit of horsepower for free from the WDQS for this (splitting label names)...perhaps even using the Label service itself to do the splitting.

The indexing service behind the scenes already stores much of this, and stores those tokens for each label.

The problem is that we don't currently have a way to get the tokens of a label for any particular QID and its labels in various languages.

And that's what I want to solve, either through SPARQL or an enhancement to the Label service or something else.

If the answer is that I will have to resort to my own programmatic methods via the dump files then so be it, I guess, but I'd rather not have to put in the work for something that is done already behind the scenes.