You could just do everything in tabular land using array database tools.

Joris is going to give a talk about it next week at the GeoPython conference in Basel. If you go there, don’t miss it!

2025.geopython.net

The development was supported by @numfocus.bsky.social (small development grant for the #GeoPandas project). A lot of the credit goes to Joris Van den Bossche and @paleolimbot.bsky.social (through S2Geography). Also many thanks to the new contributors!

github.com/benbovy/sphe...

It still misses a lot of features but already provides all the basic functionality: IO functions (WKT, WKB, GeoArrow), most predicates / measurement / overlay operations as well as a few constructive operations... All exposed as vectorized functions.

spherely.readthedocs.io

I also did `pixi global install micromamba`, because why not?

I still find the "conda workflow" (explicit environments + manual activation) relevant for some of my uses cases.

I use Jupyter a lot for testing, prototyping, etc., and #pixi-kernel works pretty nice!

github.com/renan-r-sant...

Also managed to easily discover and run tasks in emacs, using `pixi task list --json` with github.com/mohkale/comp...

It leverages Xarray indexes so the CRS information is propagated seamlessly through operations and is used for auto-alignment of Xarray objects.

It also provides some mechanisms to integrate it with other Xarray extensions (TBD).

2/2

👋 @echarles.net

Nice! We should provide similar s2 cell utility functions in Spherely as well to allow such things! 🙂

Amazing! It definitely deserves more views.

I suggest to settle all this with a paintball battle.

I have to admit I'm a big fan of @fabiocrameri.ch colormaps!