Additionally, I wanted to separate each set of 60 attributes into different tables for processing within FME before reinserting them into the same document (e.g. using a single writer).
Hi @je6thoma, if all the Curve #1 - #18 features have a common attribute that can be used as the Join On parameter, send all the curve features to the Supplier port and check the "Generate List" option in the FeatureMerger parameters dialog. With this option, the transformer adds a list attribute to the resulting features, which stores all merged attributes from multiple supplier features. Then you can explode the features on the list with the ListExploder. Note: Make sure that "Using Incoming List" is set to the "Conflict Resolution" parameter of the ListExploder.
Additionally, I wanted to separate each set of 60 attributes into different tables for processing within FME before reinserting them into the same document (e.g. using a single writer).
you can do that with setting the "Feature Type Fanout", See here: Setting Feature Type Fanout
I think you can win here but maybe you can help us a bit more so we can see what is going on.
From the example, we see 60 requestors. Can you give us a tiny sample of what data is on a requestor and most importantly what "key" attribute it has that will connect with the suppliers.
Similarly, what is the data model of the suppliers. What Keys to do they have.
And lastly, what would you like the output to be. Do you need the requestor duplicated for each supplier it matches?
i.e.
Requestors look like:
keyAttr Val1 Val2
--------- ----- -----
3 a b
Suppliers look like:
...
Output should look like:
...
If you can give us that we can help.
Note -- I doubt you'll need the testfilter ultimately. I suspect that Group-ing in the FeatureMerger may accomplish the same thing...if it is necessary at all.
Good scenario!
also conditional attribute creator might be helpfull.
Depending on what it actually is you are using as criterium.
If there is no criteria, yu can just join em unconditionally.
