MoonkaleRust files show up in the graph as their fn, struct, impl and the edges between them; Julia files were dots with no connections — is an add-on needed?
Done 2026-09-21. No add-on: symbols come from the index’s tree-sitter extractors, and only Rust had one. Two more now, bundled in moonkale-index (native, so desktop and server): Julia (tree-sitter-julia, symbols_julia.rs) — function and short-form f(x) = …, struct / mutable struct, abstract type, primitive type, module, macro, const; definitions inside a module link to it with Contains; dotted names such as Base.show are kept — and Python (tree-sitter-python, symbols_python.rs) — def and class, decorated or not, methods inside a class linked to it. Each symbol is a node with a Defines edge from its file, exactly like Rust’s. What “interconnections” means today, for every language: file → symbol (Defines) and container → member (Contains); calls and references are not extracted for any language (that is LSP territory or a later tree-sitter pass — see Core Languages). The other core languages (C/C++, Go, TypeScript, Lean, Nix, …) still show as dots. Unit tests parse a module with every form.