Tree-sitter (parser generator)

This article is about the parser generator. For the form of political protest, see Tree sitting.

In computing, Tree-sitter is a parser generator and incremental parsing library.

Tree-sitter

Original author(s)	Max Brunsfeld
Initial release	2018; 7 years ago (2018)
Stable release	0.25.3[1]  / 5 March 2025; 2 months ago (5 March 2025)
Repository	github.com/tree-sitter/tree-sitter
Written in	Rust, C
Platform	Cross-platform
Type	Parser generator
License	MIT License
Website	tree-sitter.github.io/tree-sitter/

Details

It is used to parse source code into concrete syntax trees usable in compilers, interpreters, text editors, and static analyzers.^[2][3] It is specialized for use in text editors, as it supports incremental parsing for updating parse trees while code is edited in real time,^[4] and provides a built-in S-expression query system for analyzing code.^[5]

Text editors which have official integrations with Tree-sitter include Atom,^[6] GNU Emacs,^[7] Neovim,^[8] Lapce,^[9] Zed,^[10] and Helix.^[11] Language bindings allow it to be used from programming languages including Go, Haskell, Java, JavaScript (with Node.js and WASM), Kotlin, Lua, OCaml, Perl, Python, Ruby, Rust, and Swift. Tree-sitter parsers have been written for these languages and many others.^[12] GitHub uses Tree-sitter to support in-browser symbolic code navigation in Git repositories.^[13]

Tree-sitter uses a GLR parser, a type of LR parser.^[14][15][13]

Tree-sitter was originally developed by GitHub for use in the Atom text editor, where it was first released in 2018.^[16][6]

See also

• Comparison of parser generators