initial pass at docs

[jGaboardi]

• This is an initial pass at docs; nowhere near complete.
• @martinfleis & @anastassiavybornova let me know how this is looking so far
• Also, what other sections to include / exclude etc.
• Need to think about what / how many notebooks and what data to include for that
  • perhaps use Apalachicola (or some other small place)?
  • either way, we might want to move apalachicola_original.parquet from neatnet/tests/data/ into a neatnet /data/ directory

xref:

• formal API documentation #5
• logo for neatnet #153

[martinfleis]

We need the dependencies defined somewhere, possibly in pixi env. Had to add like 6 packages to make it run :).

Let's talk about the structure in person.

[jGaboardi]

We need the dependencies defined somewhere, possibly in pixi env. Had to add like 6 packages to make it run :).

I think I was probably doing this in a conda environment when I did initially. Still haven't fully adopted the pixi paradigm, very cumbersome – but do not lose faith in me.

Let's talk about the structure in person.

So how do we want to proceed?

[martinfleis]

I have pushed an update to ensure all deps come with pixi tests env and in 0759d77 a proposal on what I think could be exposed as a public API.

[martinfleis]

Furthermore, I would move close_gaps, extend_lines and FaceArtifacts from momepy here and make them a part of public API as well.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 98.8%. Comparing base (cca5abf) to head (b732b5c).

Additional details and impacted files

@@          Coverage Diff          @@
##            main    #160   +/-   ##
=====================================
  Coverage   98.8%   98.8%           
=====================================
  Files          6       6           
  Lines        998     998           
=====================================
  Hits         986     986           
  Misses        12      12           

Files with missing lines	Coverage Δ
neatnet/__init__.py	100.0% <100.0%> (ø)
neatnet/artifacts.py	97.9% <ø> (ø)
neatnet/simplify.py	98.2% <ø> (ø)

[martinfleis]

perhaps use Apalachicola (or some other small place)?

I'd rather not as the Apalachicola case only works if you include sidewalks as we do in our tests. But normally, you should start with data without. Some other small-ish place?

[jGaboardi]

perhaps use Apalachicola (or some other small place)?

I'd rather not as the Apalachicola case only works if you include sidewalks as we do in our tests. But normally, you should start with data without. Some other small-ish place?

Since we depend on momepy why not Bubenec? Or is that too small?

[martinfleis]

There's nothing to simplify.

[jGaboardi]

• pixi updates – Thanks!
• public API – I'm OK with that
• transfered momepy functions – Sure. Probably FaceArtifacts in artifacts.py and the others in geometry.py?
• small demo data – Any other ideas? Maybe @anastassiavybornova has a nice idea?