Hi! I need to find Neighbor Lines from Blue (Base) to Red ones (Candidates) and have the average distance between these neighbors. So i splited lines into vertex and for each genererated find the closest line and then calculate average with a StatisticsCalculator. It worked graat but only for one Neighbor. I tried use NeighborPairFinder but it doesn't work to, becouse it found neighbors when Blue lines have more than one neighbor.
Any suggestion?
If you use the neighborpair finder, then the outer blue line will only find pairs if the range is equal or larger then the distance between the outer red and outer blue line. ( as the help says..."
If no, or only one, Candidate feature is found to be within the maximum distance, then the Base feature will be output unchanged via the UnmatchedBase port.")
Best to build an iterative neighborfinder, with keep candidates ex_or base option.
You can set the max distance to a large value and add a list to the results see attribute accumulation section in the NeighborFinder.
Naturally the number of neighbors to find should be more than 1.
You can then explode the list elements or search in the list for more neighbors.
Hope this helps, Itay
If you use the neighborpair finder, then the outer blue line will only find pairs if the range is equal or larger then the distance between the outer red and outer blue line. ( as the help says..."
If no, or only one, Candidate feature is found to be within the maximum distance, then the Base feature will be output unchanged via the UnmatchedBase port.")
Best to build an iterative neighborfinder, with keep candidates ex_or base option.
@tram
Hi I tried your setup with the neighborpairfinder.
And get same result as you do. It will only do 1 base in this setup.
Because the separation angle is the limiter?
If you use a second neighborpairfinder on the unmatchedbase with a negative separation angle, it will match it. (though only to closest line twice ?!?.)
Don't know if this is intended behavior....maybe safe can say?
You need 2 neighborfinders for to do it. Or a iterative NBF with options.
You can set the max distance to a large value and add a list to the results see attribute accumulation section in the NeighborFinder.
Naturally the number of neighbors to find should be more than 1.
You can then explode the list elements or search in the list for more neighbors.
Hope this helps, Itay
That can be solved by making a (nested) custom transformer to iterate neighbor finding.
One can then have options: "keep Candidates" and/or "Keep bases". In case "and" you can control the iteration count.
(is what I did anyways). I use it very often.
Safe should built a general one.
A
That can be solved by making a (nested) custom transformer to iterate neighbor finding.
One can then have options: "keep Candidates" and/or "Keep bases". In case "and" you can control the iteration count.
(is what I did anyways). I use it very often.
Safe should built a general one.
A