Visualization of the backtrader codebase

[ivanmilevtues]

This PR contains diagrams which visualize the backtrader codebase. You can see how they render here:
https://github.com/CodeBoarding/GeneratedOnBoardings/blob/main/backtrader/on_boarding.md

The aim of these diagrams is to help people get up-to-speed with the codebase. This said we don't want to introduce something which again will require you to maintain, so that is why we generated the diagrams with the help of static analysis and LLMs.

Backtrader has over 4.4K forks, which shows there's a strong community of developers engaging with the codebase—not just as users, but as contributors as well. From what we've seen, most of them are typically interested in modifying one or two components that are relevant to their specific needs.

With that in mind, we believe having a high-level diagram makes it much easier to navigate the system and quickly understand how a specific component fits into the broader architecture. To support this, we've also recently introduced a GitHub Action that keeps the diagrams automatically up-to-date with the latest changes.

We’d love to hear your thoughts on this approach. Also, how do you currently onboard new contributors to the codebase? If there's any way we can help streamline that process, we'd be happy to collaborate.

Full disclosure: we're trying to turn this into a startup, but we're still in a very early stage and figuring out what will actually be useful for people.

I would usually start by opening a discussion, but they are not enabled for this repository so I opened a PR instead!

[GoldenWind8]

Very cool

[ivanmilevtues]

Thanks man, appreciate it @GoldenWind8

[ivanmilevtues]

Hey dropped by to mention that we are now open-source so if you are intereted you can check the repo: https://github.com/CodeBoarding/CodeBoarding @GoldenWind8

[femtotrader]

https://deepwiki.com/mementum/backtrader provides an interesting visualization also

[ivanmilevtues]

Yes, for sure I think the goals are different. As we are trying to do a pure visualization which is solely based on the source code - that is we base our approach in Static Analysis. While as you can see the deepwiki approach references the readme quite often and their goal is more the have a nice wiki-page, and visuals are not first class citizens :D

[Youcef3939]

it's really awesome!