3

I'm working on determining relationships (boundary/interior intersections) between two 3D objects (triangular faces) and stumbled on Shapely, which I am interested in using instead of implementing my own point/segment/ray/triangle intersection functions.

However, I'm running into the following problem:

>>> from shapely.geometry import Polygon
>>> poly = Polygon([(0,1,1),(1,-1,1),(-1,-1,1)])
>>> poly2 = Polygon([(0,1,0),(1,-1,0),(-1,-1,0)])
>>> poly.intersects(poly2)
True
>>> poly.equals(poly2)
True

The problem I seem to be running into is that the two polygons are equal in their 2D orthogonal projections (same triangle), but in different planes (one's at Z=1, other at Z=0), but Shapely is saying they're equal and intersect.

Is there some magic I'm missing to make shapely think in 3 dimensions? I've been googling, but every example I've seen so far is only in two dimensions.

Georgy
  • 12,464
  • 7
  • 65
  • 73
squishy
  • 33
  • 1
  • 3

1 Answers1

6

According to the Shapely manual, it states that the following for the z coordinate plane for geometric objects:

A third z coordinate value may be used when constructing instances, but has no effect on geometric analysis. All operations are performed in the x-y plane.

If your calculations require the z coordinate plane, then Shapely might not be for you. Of course, you could try to get the points of the polygon as a list and compare it to other polygons. However, if you want to have a Python geometric library that can handle the z dimension, you can find some here.

Community
  • 1
  • 1
enderskill
  • 7,354
  • 3
  • 24
  • 23