Here's one way using two JSONFragmenters in parallel:

You'll maybe want to add e.g. an AttributeCreator after each JSONFragmenter to identify if each line is from EP1 or EP2.

Yes, but would basically have the same effect as using a bunch of ListExploders, which would also require manually inserting one for each "EP". I'd kinda need something like.. ["W-Number"]["EP"[*]][*]?

I'd use a JSON fragmenter to break into EP fragments then a list exploder



It can probably be done in xquery using the XMLXQueryExtractor, however I dropped into python instead.

import json
 
class FeatureProcessor(object):
    def __init__(self):
        pass
    def input(self,feature):
        j = json.loads(feature.getAttribute('json'))
        w = j["W-Number"]
        for k,v in w.items():
            feature.setAttribute("W-Number",k)
            for x in v:
                for l,m in x.items():
                    feature.setAttribute(l,m)
                self.pyoutput(feature)
                
    def close(self):
        pass

For some reason when I originally tried this, the JSONFragmenter rejected the feature.

Trying it again, it's now working, and the best solution. All you need to do afterwards is rename the json_index attribute to W-Number.

I don't understand the comparison with ListExploders. I understand wanting to avoid them, as they are very slow if you have lots of data, but the JSONFragmenter should be much faster.

Wonderful, this got rid of the JSONFlattener and AttributeExposer as well 👍

The output (and effort to make/edit the workspace) would be the same. I got kinda stuck trying to use the AttributeExposer output, but it wasn't even necessary in the end (see the other answer).