[Community] WKT Z Values (Shapely-1.0.11)
sgillies at frii.com
Tue Mar 17 20:07:03 EET 2009
I'm happy to hear from happy users :)
Short story: the GEOS WKB and WKT writers don't write out Z. I'm open
to reimplementing them in Python, and the Shapely 1.1 architecture (in
trunk) will make that fairly easy.
Long story: GEOS isn't a 3D library at all. To GEOS (and PostGIS) the
LineString([ (0, 0, 0), (1, 1, 1) ])
is a short, slanting wall 0 units high at one end and 1 at the other.
This is commonly referred to in FOSS4G circles as a 2.5D treatment.
The heights of these walls are not considered during any spatial ops.
On Mar 17, 2009, at 11:52 AM, moosilauke at gmail.com wrote:
> Hello --
> I just wanted to say that I'm psyched to be using Shapely. I'm very
> tired of ArcObjects, and I'd rather support free software anyway.
> Combining Shapely with PyShapelib and PyShapeTree saves me so much
> I have a question about something I've encountered using
> Shapely-1.0.11. I checked the archives a year back to see if someone
> else has posted something about an issue I'm seeing, but I didn't
> find anything. When I create a point object containing a z value, it
> seems that the z value is lost when I generate the WKT
> representation of that point. For example,
> >>> p = geometry.Point(1,2,3)
> >>> p.wkt
> 'POINT (1.0000000000000000 2.0000000000000000)'
> The z is missing from this output. Should it return something like
> Point Z (1.0000000000000000 2.0000000000000000 3.0000000000000000) ?
> Community mailing list
> Community at lists.gispython.org
More information about the Community