Conda Meep on Windows, Mac, and Linux with Docker

[theogdoctorg]

Hello Meep community!

I labeled this as a "Show and tell" post to introduce a quick-start method to install & run Python and Meep multi-platform (including Windows) using Docker containers. I hope this might make Meep a bit more accessible to those just starting out, as well as those looking for a sandboxed development environment to try some code changes without borking the whole system in case of a bad configuration as might happen on a bare-metal install. If it seems helpful, I'd humbly suggest that the Docker container method be added as an alternative in the official Meep documentation.

https://github.com/theogdoctorg/docker-conda-meep/

Installation for use with Jupyter notebooks is really as simple as installing Docker, running a single command (docker run....) and copy-pasting a link into a browser window. Windows complicates things only a bit by requiring activation of WSL, which on Windows 11 was a single command (wsl --install) at a command prompt followed by creation of a username & password.

Once the container is running, there's a few different choices for how to run/access python & Meep.

Full disclosure: this is my first time using Docker at all, but so far I've been pleased by the cross-platform compatibility it allows. I have tested basic functionality of the single-threaded Jupyter notebook option so far on an Intel Mac laptop, Windows 11 PC (via WSL), and PC running Ubuntu. Things get a bit more complicated trying to foward a GUI from inside the container, which I've only tested that on Ubuntu at this point. All of the options do work on Ubuntu.

Any thoughts/suggestions?

BTW, for more development-focused efforts, I intend to update/develop Scimax's docker image here:
https://github.com/theogdoctorg/docker-meep-and-mpb
So far the only changes are to update to Ubuntu 24.10 and some of the packages, but so far everything seems to be compiling. Not sure how to run the tests to confirm a good build that I remember y'all mentioning in the past.