Skip to main content
  • EN

AreaGapAndOverlapCleaner has turned on itself, see examples below. The changes occur within tolerances, but the transformer should only be modifying gaps between neighbouring polygons, not itself? I have checked incoming features with the GeometryValidator but can not find anything wrong with them.

What is going on here?

 

Before AreaGapAndOverlapCleaner (this is a river)

 

Skärmavbild 2023-09-13 kl. 11.32.46 

After (AreaGapAndOverlapCleaner thinks there is a drought and has turned the river into puddles)

Skärmavbild 2023-09-13 kl. 11.32.56 

Before and after combined

 

Skärmavbild 2023-09-13 kl. 11.32.38 

Another example: Here AreaGapAndOverlapCleaner (light brown) has attacked corners and bends. See the tip of the left area, and the small slivers in the bend on the far right have been introduced by the transformer.

 

Skärmavbild 2023-09-13 kl. 12.02.47 

 

Are you able to share the original geometries?

Sure, here come the offenders.

Sure, here come the offenders.

And you're saying that the neighbouring polygons do not overlap these? Because the AreaGapAndOverlapCleaner will address the overlaps as well as gaps

And you're saying that the neighbouring polygons do not overlap these? Because the AreaGapAndOverlapCleaner will address the overlaps as well as gaps

I suspect it’s some strange bug in my system. But if I run the AreaGapAndOverlapCleaner on the sample I shared I get the strange result pictured in the three first images of my first post. There are only two features in the sample, and they are well out of tolerances reach from each other. It’s like AreaGapAndOverlapCleaner has started to do a little self generalisation on the side.

I have now tested this sample on an other installation of FME (separate machine) as well as both 2022 and 2023. The problem persists. There could of course be some irregularity in the input data, but I have not been able to find any.

I have now tested this sample on an other installation of FME (separate machine) as well as both 2022 and 2023. The problem persists. There could of course be some irregularity in the input data, but I have not been able to find any.

FYI, I have filed FMEENGINE-80398 about this issue.

I have the same problem, with geometry downloaded from the dutch BGT (https://app.pdok.nl/lv/bgt/download-viewer/) and read as citygml. The BGT is already mostly topologically correct so in and out should be the same, but for example here it damaged the corner of a house (in green) image

Reply


Community Stats

30,108
Posts
114,295
Replies
37,745
Members
Terms & ConditionsCookie settings
A product of
Safe Software respectfully acknowledges that we live, learn and work on the traditional and unceded territories of the Kwantlen, Katzie, and Semiahmoo First Nations.