fix: explicitly extract xy coords from 3D-LineString

[AlFontal]

I've been dealing with some trajectory calculations which end up producing 3D LineStrings as geometry.

I am, however, unable to use geom_map with this kind of GeoDataFrame since plotnine expects xy coordinates and fails when the geometry.coords object contains tuples with a z dimension.

It would also be nice to be able to use the z dimension for some other aesthetics, such as color, for instance. My solution for now has been to generate Point geometries from the LineStrings and use geom_path instead of geom_map, but it would be great if this could be handled directly when passing a LineString geometry.

This little fix makes 3D linestrings not throw an error, but does so by completely disregarding the z dimension. I am not completely sure of how to fix this to be able to use z as an extra aesthetic without explicitly reverting the LineString into Points...

[codecov[bot]]

Codecov Report

Merging #570 (cb9fb6b) into master (a81b8f9) will decrease coverage by 0.01%. The diff coverage is 33.33%.

@@            Coverage Diff             @@
##           master     #570      +/-   ##
==========================================
- Coverage   85.23%   85.21%   -0.02%     
==========================================
  Files         154      154              
  Lines        9930     9932       +2     
  Branches     1768     1770       +2     
==========================================
  Hits         8464     8464              
- Misses        949      950       +1     
- Partials      517      518       +1     

Impacted Files	Coverage Δ
plotnine/geoms/geom_map.py	74.35% <33.33%> (-1.96%)	:arrow_down:

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update a81b8f9...cb9fb6b. Read the comment docs.

[has2k1]

There isn't a way to map to the z coordinate. The key issue is, the mapping translates to one aesthetic value for each entry in the dataframe, but for 3-d lines then each entry has a variable number of z coordinates!

Though I think with some good thought you may be able create a non-standard custom geom that does that.