signed_distance returns different answer if polygon is explicitly closed or not

[skygering]

import GeometryOps as GO
import GeoInterface as GI

pt = (1.1, 1.0)
rect_open = GI.Polygon([[(1.0, 1.0), (1.0, 2.0), (2.0, 2.0), (2.0, 1.0)]])
rect_closed = GI.Polygon([[(1.0, 1.0), (1.0, 2.0), (2.0, 2.0), (2.0, 1.0), (1.0, 1.0)]])

GO.signed_distance(pt, rect_open) # return 0.1
GO.signed_distance(pt, rect_closed) # return 0.0

[asinghvi17]

Ah oops! I imagine we should probably filter that based on first + last coordinate...

[skygering]

It's annoying because GI.Polygon doesn't enforce repeated points, but we should still assume they're closed. This is related to #46, but an example of it biting me in the butt 😆

[rafaqz]

Not so easy to enforce repeats because we can wrap other polygon types

We could have a repeat bit we set at construction that adds an extra point to getpoint...

[asinghvi17]

Or just add the extra point to the end if necessary

[rafaqz]

Hmm this means both GI.getpoint and GI.npoint for GI.LinearRing do a comparison of the first and last points and add the extra point or the +1 when necessary?

Wonder how expensive that is - mostly what it does to the iterator GI.getpoint(::LineString) having a conditional in there for every point access. Probably significant compared to directly accessing a vector.

[asinghvi17]

Ah yeah, that would be expensive...I was thinking we could do this at construction, but it wouldn't solve the problem for external types. I suppose it would add a bit more complexity since most methods would be expecting a closed polygon as well, so they'd have to check for that each time they were called, which is suboptimal.

[rafaqz]

Probably we can add a close argument to apply that will close any LineaRingTrait it encounters (and LineString when its in a Polygon for some extra complexity)?

Then we could to tuples(obj; close=true) and just take the conversion hit.