Documentation Link

https://matplotlib.org/devdocs/devel/development_setup.html#create-a-dedicated-environment

Problem

Actually, I run into a little bit into trouble to get the c_internals compiled. After reading the docs, it sounds like that the Codespace environment is set up in way that out of the box it can be used. So I am wondering if there can be:

1. provide further instructions how to compile the c_internals
2. or update the .devcontainer/setup.sh

I know about this section create-a-dedicated-environment, but I haven't managed to get it running in Codespace.

Other users might face similar struggles. If there's an option in the shell to autocompile during the CodeSpace setup, it would make this environment very convenient.

Actually, I was planning to address this blind spot in the coverage rate of the PR #29520 , but I wasn't able to get pytest running.

ERROR lib/mpl_toolkits/axes_grid1/tests - ImportError: cannot import name '_c_internal_utils' from 'matplotlib' (/workspaces/matplotlib/lib/matplotlib/__init__.py)
ERROR lib/mpl_toolkits/axisartist/tests - ImportError: cannot import name '_c_internal_utils' from 'matplotlib' (/workspaces/matplotlib/lib/matplotlib/__init__.py)
ERROR lib/mpl_toolkits/mplot3d/tests - ImportError: cannot import name '_c_internal_utils' from 'matplotlib' (/workspaces/matplotlib/lib/matplotlib/__init__.py)

Further Sources:

• https://discourse.matplotlib.org/t/cannot-import-name-c-internal-utils-from-matplotlib/22644
• https://github.com/matplotlib/matplotlib/pull/29520/files#annotation_31349915228

Suggested improvement

• Include procedure for compiling c_internal into .devcontainer/setup.sh
• Provide some details about how to gcc via conda.