Skip to main content

I want this feature set to be processed per tracknumber. So first all the features with tracknumer 1 (2 features), then with tracknumer 2 (3 features), then trackunmer 3 (2 features).

From each subset only one feature has to be selected, based on 2 rules

  • if all features have 'yes' in their decision attributes, select the one with the most recent date
  • if one of the features in the subset has a 'no' in one of the decision attributes, select (one of) the no fields, doesn't matter which one.

The resulting featureset is seen in knipsel1.png

 

I have tested a Matcher , but then I am stuck more or less. I put the subsets in a list in the Singlematch output port. If I use a Listexploder, I get all the features back in one set. I want to take each feature of the singlematch output port and process the Listelements in a seperate process (in a workspacerunner), but how do I feed one feature (with a list) to the workspacerunner?

Other lines of thought are an Inlinequerier attached tot the Matcher or at the inputtabel, but that could be too complex for SQL to handle.

​I would probably aggregate by tracknumber using the concatenate option on both decision attributes - put the date into a list attribute.

Your output should be one feature per track with a two comma separated yes,no decison attri​bites and a list of dates.

Then you can use a tester on both decion attributes to see if either contain "no's".

For the features which only have "yes" use a list explorer, then sort by date and then a Sampler to sample first 1 features (group by track number) to get the ​most recent feature for each track.

For the no's perhaps is doesn't matter?​

There might be a better way but this I think would work​


Reply