Skip to main content
Question

How to split supersmall intersected parts of line


Forum|alt.badge.img

This kind of line intersections are in source data and can be visible in ArcGIS only at very large scale (1:1 or closer) and Intersector doesnt split them. There are no option to set tolerance in Geometry validator or Intersector, nor other transformers, that are involved.

9 replies

erik_jan
Contributor
Forum|alt.badge.img+17
  • Contributor
  • July 11, 2017

Could the Snapper transformer help in this case (set snapping tolerance to snap the short endpoints to the intersecting line)?


takashi
Influencer
  • July 12, 2017

Hi @is, unlike ArcGIS, FME computes geometries with full precision of double precision floating point number. I therefore think it could also be possible that the intersection shown on the display was just a computational error of ArcGIS and the lines aren't intersecting in the full precision actually.

Have you confirmed that there is an actual intersection by seeing the full precision coordinate values of relevant vertices? You can see every coordinate values with the Feature Information window of FME Data Inspector.


Forum|alt.badge.img
  • Author
  • July 12, 2017

Yes, intersections are real - ArcGIS Check Geometry tool shows this parts as error geometries.


takashi
Influencer
  • July 12, 2017

I don't think the ArcGIS Check Geometry tool could detect intersections between different lines. If you need to split a line at its self-intersections, set Yes to the "Split Self-Intersecting Features" parameter in the Intersector transformer.


Forum|alt.badge.img
  • Author
  • July 12, 2017

Thanks for opinion, but the problem is that "Yes" to the "Split Self-Intersecting Features" parameter in the Intersector transformer is already set, but supersmall loops stays unsplitted as they are.

Looks like Snapper (vertex snapping mode) could help, like @erik_jan suggests.


Forum|alt.badge.img
  • Author
  • July 12, 2017
erik_jan wrote:

Could the Snapper transformer help in this case (set snapping tolerance to snap the short endpoints to the intersecting line)?

It's worked- result is right! Thank You!

 

 


ebygomm
Influencer
Forum|alt.badge.img+32
  • Influencer
  • July 12, 2017

It sounds like the intersections are not true intersections but are lines which are within a certain distance of each other which is smaller than a specified tolerance? FME will not report these in a straightforward way as the lines do not truly intersect


Forum|alt.badge.img
  • Author
  • July 13, 2017
ebygomm wrote:

It sounds like the intersections are not true intersections but are lines which are within a certain distance of each other which is smaller than a specified tolerance? FME will not report these in a straightforward way as the lines do not truly intersect

It is one line with supersmall loop - one record in databese, not two intersected lines or more.

 

 


ebygomm
Influencer
Forum|alt.badge.img+32
  • Influencer
  • July 13, 2017
is wrote:
It is one line with supersmall loop - one record in databese, not two intersected lines or more.

 

 

Can you share the geometry of one of these lines?

 

 


Cookie policy

We use cookies to enhance and personalize your experience. If you accept you agree to our full cookie policy. Learn more about our cookies.

 
Cookie settings