Hi! This is a proof-of-concept to explore the concept of "code as data". Your codebase could tell you so much only if it was easier to get at. It's generally talked about in abstract terms, but I had a specific problem.
Elasticsearch is a piece of software with a lot of different settings you can tweak and the documentation is pretty good if you know what to look for. As someone operating a cluster, there is no one place that lists all the settings! I wanted to compile a full list of Elasticsearch settings to have a quick lookup and to help discover settings I didn't know about or might even be undocumented.
So, I used bblfsh to analyze the entire Elasticsearch codebase and get all the settings. Take a look at the code! This project was done as an experiment and I currently don't plan on investing a ton of time into it, but if you see any issues or want to help out, please open an issue on the repository.