pyiron conda builds

[landinjm]

For pyiron it's best practice to have a uniform build/package management stack. Since they mainly use conda, it would be a good idea to migrate our existing Julia dependencies onto conda.

This is a longer term project though.

[j-fu]

Ok I made a small test in a clean environment:

• requirements.txt:

juliacall>=0.9.0
click>=8

• juliapkg.json:

{
   "julia": "1.12",
   "packages": {
                 "ExtendableGrids": {
                 "uuid": "cfc395e8-590f-11e8-1f13-43a2532b2fa8",
                  "version": "1"
                    }
                }
}

(this could be our solver once registered instead of ExtendableGrids)

• test.py

from juliacall import Main as jl
jl.seval('using ExtendableGrids')
X=[float(x) for x in range(10)]
g=jl.simplexgrid(X)
print(jl.num_nodes(g))

I could run these under a clean user id without previous julia installation.

conda create -p condaenv pip
conda activate ./condaenv
pip install -r requirements.txt
python -m juliapkg resolve
python test.py

@jan-janssen: I am not sure how to move the juliapkg step to conda