Just in case, pinging @bbatsov :)

I don't understand what should be written in the last section "Semantic Interpretation in clojure-ts-mode".

Basically it should be an explanation what gets treated a deserving special indentation and font-locking compared to regular function call forms or data forms. (e.g. how do we decide that something a macro call that has a body) In our case that's done via the configuration you've recently introduced, but we could have also extracted this information from parsed definitions - e.g. if a form takes a body param or special metadata we might generate on the fly some special font-locking and indentation rules for it. (although I guess we won't be working in such a direction)

but we could have also extracted this information from parsed definitions - e.g. if a form takes a body param or special metadata we might generate on the fly some special font-locking and indentation rules for it. (although I guess we won't be working in such a direction)

I was thinking about it and I couldn't come up with a design for such a feature. We would need a way to describe the entire form's syntax somehow, produce a query and inject it to treesit-font-lock-settings somehow. All the queries must be precompiled, otherwise it will be really slow, so I'm not sure there is an easy and efficient way to implement this.

Once it's merged, I think #73 can be closed.

I was thinking about it and I couldn't come up with a design for such a feature. We would need a way to describe the entire form's syntax somehow, produce a query and inject it to treesit-font-lock-settings somehow. All the queries must be precompiled, otherwise it will be really slow, so I'm not sure there is an easy and efficient way to implement this.

I was thinking something similar, that's why I have doubts we'll ever work in this direction. It's much easier to extract such information from the runtime (as CIDER does).