perl-lsp — a fast Perl language server in Rust with cross-file type inference, completion, goto-def, and rename. Built on tree-sitter-perl and tower-lsp.
Witness bags with graph edges handle Perl's late binding better than Perl::LanguageServer.
Perl developers
Perl::LanguageServer · Pyright · Intelephense
How it works: - builds witnesses and scopes by walking the parse tree. Fully known values get concrete types, otherwise they get an edge pointing to the last witness we have for this symbol - a worklist runs to a fixed point: reducers fold each symbol's witnesses and chase edges. At this point, the witness bag is complete and ready to use - all downstream features are queries into the bag, following along the graph edges that are relevant
Extensible - you can bring your own Rhai plugin to shortcut common patterns (exporters and OOP frameworks), or even add new types of definitions (worker tasks and web routes).
The build/query timing distinction with edges lets you late bind types, which is duly similar to how Perl itself works: `Foo->new` isn't known if it's the constructor for Foo or calling a method on the sub name `Foo`; late binding lets you make that call after the worklist folds.
Tree-sitter + Rhai scripts replace opinionated formatters, but beta stability and language coverage remain questions.
Simpler than neovim but helix already does this with better LSP support.
Beats CodeQL on CWE-Bench with optional LLM enhancement mode.
Neatify exposes AST-level formatting as live Rhai scripts, so you write real code to shape output instead of tweaking a dozen JSON flags. The Tree-sitter backbone means any language with a grammar can be targeted, and the repo-first flavoring plus LLM-assisted script generation is a smart, practical twist. It's clearly early-stage (defaults modify files in-place and coverage is limited), but the architecture is an interesting alternative to opinionated, black-box formatters.
Korean keywords for programming when English dominates every other language.
LLM-assisted compiler frontend for 15 languages, but unclear if it outperforms static AST analysis alone.
