Miscellaneous Refactoring (Discussion)

[zandi]

As I'm getting familiar with the codebase to make some of my own changes, I'm wanting to make some bigger refactors for things. Like, replace the verboseprint functions with python's logging module, adding some unit tests, separate scenario setup logic from the MeshNode class, move the default config values for the Config class to a file, etc.

These feel like bigger changes so I'm wondering if there's any way you'd like things split up or coordinated? I want to make PRs that are easy to review and merge, and also not spend a lot of time on things that may not get merged.

[GUVWAF]

These all seem reasonable changes to me. Indeed it would be good to separate them into multiple PRs, I would say for the things you mentioned it would be good to have a PR for each. Although if the unit tests are related to a specific change they could be combined in there.

Thanks in advance for your effort!

[zandi]

Sounds good! I don't want to overpromise that I'll have a lot of time to work on these, but I do want to get at least some of these done.

[zandi]

FYI I think I'm going to make some issues & assign them to myself, to better plan out & keep track of the refactoring I want to do. Feel free to comment on/close them as you see fit.

[GUVWAF]

That's fine to me!

[zandi]

OK, so as mentioned in my other comment:

I want to do some rigorous testing of the setting where there are a smaller number of fixed location solar nodes that are higher in elevation and with higher gain antennas, and many more potentially mobile lower-elevation nodes with compromised antennas. I saw the stale PR (#33), and saw a lot of refactoring I'd like to have before implementing those changes, so I went down a refactoring rabbit hole. I'm getting close to testing that heterogeneous network situation so I figure there's a few more things I'd like to get in to support that testing, and then I'll be mostly content for my purposes.

Off the top of my head, and trying to avoid flights of fancy, I'm thinking:

• existing issues: burning man based optimizations, MeshNode refactoring, etc. (Refactor MeshNode to separate sim plumbing from node configuration #58, Update pip requirements #51, Bring in relevant refactors & optimizations from stale PR #33 #50)
• Add batch testing class to automate running tests & collecting results for a list of config objects & list of node config collections
• add ability to detect & measure suboptimal routes (portable node with compromised antenna winning race & rebroadcasting rather than solar node with higher-gain antenna)
  • add message/packet tracing? Track specifics of rebroadcast cancellations in solar nodes?
• Misc improvements left in my TODOs or necessary for extended code (eg: completeness check of node graph given to a discrete sim)

I'll just make issues for these like the last batch. If you want to assign them to me that's great, and if you think it's getting too far afield for what the focus is for the project and don't want the work that's fine too, not trying to run off the rails or anything.

[zandi]

Just chiming in here. I wrapped up the simulations I was planning to do that led to all the refactoring in the first place. I wrote it up here if you're curious: https://nopswamp.net/posts/2026/05/meshtastic-network-simulation-improve-your-meshtastic-networks/

If you're interested in having any of the changes upstreamed, this is my branch: https://github.com/zandi/Meshtasticator/tree/feature/heterogenous-network-sim

It's somewhat hacky, so if you want anything particular upstreamed it may take me some time to clean it up. Some of the things like new metrics may be wanted, while other things may not. Just check out the diff and let me know, no rush.

[GUVWAF]

Thanks for sharing the write-up, very interesting! Good to see some simulated evidence that infrastructure nodes do help, and muted nodes can help too. I shared it on the Meshtastic Discord too.

I'll have a look later if there's anything still in your branch I would like to include.