Note, aware of issue #5 which reports the same error message. Seeing slightly different behaviour though, in that calling setPrecision causes the error, and then solves it. Meaning:
- no setPrecision: works
- 1e-12: fails
- 1e-16: works
Test polys are shown below and geojson here. The blue line is a polygon - imagine a rubber band stretched around the red pin. The green pin is the output ring start from the error message, and the red pin the last matching segment found. At the time of the error the algo is stepping counter-clockwise around the yellow poly.
Given a lot of the "Unable to complete output ring ..." errors seem to happen with close together points, is it possible segments with duplicate (or near duplicate) points are being mixed up when producing output rings in geom-out?
Is this theoretically possible to fix? Happy to try if so. Appreciate any insights you can provide @luizbarboza 🙏
Note, aware of issue #5 which reports the same error message. Seeing slightly different behaviour though, in that calling setPrecision causes the error, and then solves it. Meaning:
Test polys are shown below and geojson here. The blue line is a polygon - imagine a rubber band stretched around the red pin. The green pin is the output ring start from the error message, and the red pin the last matching segment found. At the time of the error the algo is stepping counter-clockwise around the yellow poly.
Given a lot of the "Unable to complete output ring ..." errors seem to happen with close together points, is it possible segments with duplicate (or near duplicate) points are being mixed up when producing output rings in geom-out?
Is this theoretically possible to fix? Happy to try if so. Appreciate any insights you can provide @luizbarboza 🙏