+44Not sure if this is a known issue or not, this is in FME 2020.2
GeoJSON processed by the GeometryReplacer results in a line with 0 coordinates, it is not rejected which is what i would expect to happen
e.g.
{"type":"LineString","coordinates":[[[0,0],[100,0]]]}Processing this with the GeometryReplacer, results in this geometry
+63Your coordinates array has one too many pairs of square brackets
{"type":"LineString","coordinates":[[0,0],[100,0]]}
+44Your coordinates array has one too many pairs of square brackets
{"type":"LineString","coordinates":[[0,0],[100,0]]}
Yes, that's the point. It's invalid geojson and I would expect it to be rejected by the geometry replacer but instead a line with zero coordinates is created, it is not rejected.
+63Ah, I follow. Yes, as its specified as a line, I'd expect it to fail... as you can't have a line with 1 coordinate. It also "passes" if you just specify
{"type":"LineString","coordinates":[[0,0]]}It seems it might not be validating the geom type vs the output geom.
This is also happening in 2022.1
+19Yes, that's the point. It's invalid geojson and I would expect it to be rejected by the geometry replacer but instead a line with zero coordinates is created, it is not rejected.
@ebygomm @hkingsbury FYI I've filed FMEENGINE-77079 about this issue.
+44Yes, that's the point. It's invalid geojson and I would expect it to be rejected by the geometry replacer but instead a line with zero coordinates is created, it is not rejected.
Thanks, I wasn't 100% sure whether this would be expected behaviour or not.
