use eurec4a-intake requirements

eurec4a / how_to_eurec4a

Code examples to get you started with EUREC⁴A data.

https://howto.eurec4a.eu

MIT License

6 stars 20 forks source link

use eurec4a-intake requirements #109

Closed observingClouds closed 7 months ago

observingClouds commented 7 months ago

ensures up-to-date requirements for eurec4a-intake

fixes #107

d70-t commented 7 months ago

Ah, nice hack :-) I kind of like it.

I'm now just wondering if we might want to try going even more sustainable, and define all those requirements in the eurec4a python package? Maybe as an optional dependency like eurec4a[data] to emphasize that we want not only the dependencies which are required to execute the eurec4a package, but also those to access all the data? This would then also simplify end user installs.

observingClouds commented 7 months ago

Yeah, that sounds great. You wanna do this? ;P

observingClouds commented 7 months ago

@d70-t, I wanted to do this mostly quickly because I wanted to merge another PR to the book which should be visible online tomorrow...

d70-t commented 7 months ago

I'd be fine with merging this for now. But probably the eurec4a package approach doesn't take too long as well.

observingClouds commented 7 months ago

I'm a bit confused why I get here for the ICON-LES notebook the error

File [0;32m/usr/share/miniconda3/envs/how_to_eurec4a/lib/python3.12/site-packages/dask/dataframe/__init__.py:24[0m, in [0;36m_dask_expr_enabled[0;34m()[0m
[1;32m     22[0m     [38;5;28;01mimport[39;00m [38;5;21;01mdask_expr[39;00m  [38;5;66;03m# noqa: F401[39;00m
[1;32m     23[0m [38;5;28;01mexcept[39;00m [38;5;167;01mImportError[39;00m:
[0;32m---> 24[0m     [38;5;28;01mraise[39;00m [38;5;167;01mValueError[39;00m([38;5;124m"[39m[38;5;124mMust install dask-expr to activate query planning.[39m[38;5;124m"[39m)
[1;32m     25[0m [38;5;28;01mreturn[39;00m [38;5;28;01mTrue[39;00m

[0;31mValueError[0m: Must install dask-expr to activate query planning.

but the tests in eurec4a-intake work just fine without installing intake[dataframe] (which I just removed here now).

d70-t commented 7 months ago

Maybe we don't need [dataframe] anymore 🤷‍♂️

d70-t commented 7 months ago

Maybe this has something to do with the latest release of dask. Apparently they made something based on dask-expr the default.

observingClouds commented 7 months ago

close in favour of #111