Old Conda environment in Julia

sbacelar · May 18, 2024, 12:15pm

It seems that I have an old Conda environment in Julia: /Users/sbac/.julia/conda/3/lib/python3.7.

I am using now other Python environments with Python version 3.12. I can’t find the name of this environment. How can I (should I) update this Julia environment?

I am having a lot of problems with Quarto in Julia and I suspect that is due to the fact of not having pyyaml installed in that environment.

mkitti · May 18, 2024, 12:51pm

This is a base environment installed by PyCall.jl using miniconda or miniforge.

If you still need PyCall.jl for something, you can try a few approaches. There is probably a conda binary located here, which you could run.

/Users/sbac/.julia/conda/3/bin/conda

You could also just move the .julia/conda to .julia/conda.bak. Then do

using Pkg
Pkg.build("PyCall")

That should download a fresh conda install into that directory. The problem is that you may need to install individual Python packages again.

An alternative is that you can configure PyCall.jl to use another Python executable via the the Python environment variable PYTHON (ENV["PYTHON"] inside Julia). See the instructions in the README below:

github.com

JuliaPy/PyCall.jl/blob/master/README.md#specifying-the-python-version

# Calling Python functions from the Julia language

[![Test with system Python](https://github.com/JuliaPy/PyCall.jl/workflows/Test%20with%20system%20Python/badge.svg)](https://github.com/JuliaPy/PyCall.jl/actions?query=workflow%3A%22Test+with+system+Python%22)
[![Test with conda](https://github.com/JuliaPy/PyCall.jl/workflows/Test%20with%20conda/badge.svg)](https://github.com/JuliaPy/PyCall.jl/actions?query=workflow%3A%22Test+with+conda%22)
[![PkgEval](https://juliaci.github.io/NanosoldierReports/pkgeval_badges/P/PyCall.named.svg)](https://juliaci.github.io/NanosoldierReports/pkgeval_badges/P/PyCall.html)
[![Coverage](https://codecov.io/gh/JuliaPy/PyCall.jl/branch/master/graph/badge.svg)](https://codecov.io/gh/JuliaPy/PyCall.jl)

This package provides the ability to directly call and **fully
interoperate with Python** from [the Julia
language](https://julialang.org/).  You can import arbitrary Python
modules from Julia, call Python functions (with automatic conversion
of types between Julia and Python), define Python classes from Julia
methods, and share large data structures between Julia and Python
without copying them.

## Installation

Within Julia, just use the package manager to run `Pkg.add("PyCall")` to
install the files. Julia 0.7 or later is required.

This file has been truncated. show original

The newer package PythonCall.jl behaves differently and provides individual conda environments for your Julia projects. You might need to use PythonPlot.jl rather than PyPlot.jl for example though.