1

(It goes without saying, I am a beginner with all of this!)

I have a QGIS (version 3.28.11-Firenze) file into which I imported a KMZ from Google Earth. The purpose of this is to produce a shapefile that I'll use to import into a 3rd party pipeline modeling program. One of the resulting issues is the conversion leaves 'dangling ends'. I've figured out how to resolve the dangling ends in a single pipeline to connect the separated ends, but when I attempt to resolve a dangling end for a pipeline that connects to/forms a branch with another pipeline no matter what I do I have a dangling end there.

I've enabled the snap functions and used trim/extend feature. I've tried editing verticies. I've merged, cleaned, extracted (extracted Z values), and simplified as much as is feasible. I posted something similar before in this thread: How to extend a line to Intersect another line but I felt while related to the previous question this probably should be its own thread. That previous post did show me how to get the two lines to connect visually, but when I import the shapefile to the 3rd party program, the lines are not connected.

To make sure this was actually possible, to have a polyline with a branch, I exported a shapefile from the 3rd party program (using geopandas, if that's important) and imported into QGIS and sure enough, no dangling end at that intersection.

So I'm stuck. How do I resolve the dangling end? The screenshot attached is a full zoom; I can't zoom in any further. I don't see a gap, but the Topology Checker Panel says one is there, confirmed after the import to the other software program.

Pipeline With Branch - Red Dot from Topology Checker - Dangling End

Vince
  • 20,017
  • 15
  • 45
  • 64
Matt Greer
  • 53
  • 5
  • After additional searching, I'm fairly certain the answer lies within the v.clean function, but I have not yet been able to determine the correct selections and parameters to resolve the issues throughout. Perhaps there isn't one perfect set of parameters. – Matt Greer Oct 12 '23 at 17:43
  • 1
    Maybe they have different Z values? I dont know if topology checker can detect this. Do you need the Z:s? If not try dropping them – BERA Oct 12 '23 at 17:47
  • I misunderstood what I was doing (extracting != removal) so I exported a layer without the z dimension. That doesn't seem to have helped any.

    What I have determined through running v.clean in batch is that neither the 'break' nor 'rmdangle' has no effect on the number of dangles reported. For this file, 'snap' set to either 0.0001 or 0.00001 does "OK", so I kind of understand what's going on. However, the higher I set that setting, I start to lose resolution. It's like running 'snap' also runs 'simplify', which I don't want.

     – Matt Greer Oct 12 '23 at 18:31
  • 1
    If it's possible could you upload a sample of your problem data (geometry only is fine) to an external file sharing site and link it here? Perhaps others might be able to look closely at the actual geometry/topology and see what the issue is. It may be a very technical problem that is best solved by inspecting the data itself. Between the 3rd party program/geopandas, the KMZ format and QGIS there may be different levels of precision in determining whether lines are connected or dangling. – she_weeds Oct 13 '23 at 01:40

0 Answers0