Select to view content in your preferred language

Geometry: Don't believe what you see ... # 6

677
0
05-08-2019 05:18 PM
Labels (1)
DanPatterson_Retired
MVP Emeritus
1 0 677

Export a single shape from a featureclass.  

Looks the same ... right?

It isn't, necessarily.

Here is an example:

A single multipart shape.

The point order would be the same, wouldn't it?

Not sure any more... some different rule set seems to be applied when a shape leaves its 'nest' and ventures out into the world on its own.

I can understand point order being influenced by the construction method and its association with other shapes.  In the case of the shape 'in-group', the outer part is ordered from the top right(10, 20) and clockwise.  The second, smaller part is the listed second (line 13).

When Shape 1, ventures out on its own, the inner part is now first (line 2), almost like freedom provide a new rule-set during its transition

Shape 1... its in-group pointsShape 1... out on its own
array([
array([[10., 20.],
       [10., 10.],
       [ 0., 10.],
       [ 0., 20.],
       [10., 20.],
       [nan, nan],
       [ 2., 18.],
       [ 2., 12.],
       [ 8., 12.],
       [ 8., 18.],
       [ 2., 18.]]),
array([[ 7., 17.],
       [ 7., 13.],
       [ 3., 13.],
       [ 3., 17.],
       [ 7., 17.],
       [nan, nan],
       [ 5., 16.],
       [ 4., 14.],
       [ 6., 14.],
       [ 5., 16.]])
], dtype=object)‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍
array([
array([[ 7., 17.],
       [ 7., 13.],
       [ 3., 13.],
       [ 3., 17.],
       [ 7., 17.],
       [nan, nan],
       [ 5., 16.],
       [ 4., 14.],
       [ 6., 14.],
       [ 5., 16.]]),
array([[10., 20.],
       [10., 10.],
       [ 0., 10.],
       [ 0., 20.],
       [10., 20.],
       [nan, nan],
       [ 2., 18.],
       [ 2., 12.],
       [ 8., 12.],
       [ 8., 18.],
       [ 2., 18.]])
], dtype=object)‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍

So, there is probably a logical explanation for what is seen, BUT, if you were relying on the initial point ordering when the shape lived at home, to apply when it moved out on its own, then you would be in for a surprise.

Conclusions:

  • Don't rely on what you see.
  • Examine what you have.
  • Be prepared to 'deal'
  • Not all shapes behave the same way when they move out.
  • Any rules that you come up with probably have a corner-case
  • At least it didn't take anything that didn't belong to it, when it moved out  
About the Author
Retired Geomatics Instructor at Carleton University. I am a forum MVP and Moderator. Current interests focus on python-based integration in GIS. See... Py... blog, my GeoNet blog...
Labels