Skip to main content

Good day,

 

I would like to ask. I am currently solving the tracks in the WS in the attachment. I ran into a data "loss" issue happening on featureMerger. Points from the polygon of interest go to the Requestor's input, and points with the same ID as the requestor's input go to the Supplier's input - these points are part of the RouteSegment. Now to my problem. I believe that I don't have all the data on the FeatureMerger output because I only have each point once on the requestor input, however, I can have a point multiple times on the suplier input and it can be part of different routeSegments. At the moment when a point at the requestor input is "reserved" by the first point of the same ID on the supplier, at that moment the rest of the data goes to the Rejected output.

 

I can think of creating X featuremergers, where I would send the rejected data to the input of the supplier and the original polygon_selection_topsky source to the requestor. But I find it quite daunting. I vaguely remember that we used a "data flow repeater" somewhere, where it was possible to set the branch to stop if a certain number of repetitions or the expected output was reached. Could you give me some solution please?

Because the attachment limit is 50MB, below I am sending a link to download the workspace:

 

https://www.uschovna.cz/zasilka/BP9TMMJHFD2S53W3-GIP/

 

Thank you

I have not downloaded the workspace but...

 

Check under the Attribute Accumulation part of the FeatureMerger parameters, if there's a 1-to-many relationship you can have it generate a list on the output features.


I have not downloaded the workspace but...

 

Check under the Attribute Accumulation part of the FeatureMerger parameters, if there's a 1-to-many relationship you can have it generate a list on the output features.

Hi, thanks for the reply. But I don't see such a setting on the featuremerger.


Hi, thanks for the reply. But I don't see such a setting on the featuremerger.

@fmesafe.podpora​ 

image


Hi, yes I found that, but I can't set up a 1:N session anywhere.


@fmesafe.podpora​ FeatureMerger doesn't use that terminaology. You either use Process Duplicate Suppliers or Generate List.

But I have a feeling that the FeatureJoiner might be a more appropriate transformer in this case - although I did not download the workspace to check


Reply