// the take
A Rust-based linter/formatter that fixes spacing and punctuation issues in mixed CJK/English text — the classic 'ChineseEnglish' problem with no space between scripts. Built for teams writing bilingual docs, comments, or UI copy who want this enforced in CI the way ESLint enforces JS style. Ships as a CLI plus bindings for Node, Python, Ruby, Java, and a WASM build.
Genuinely broad language coverage (28+ file types) via per-language AST/PEG grammars, so it only touches strings and comments rather than naively regexing whole files. The rule set is configurable per-rule with inline disable comments, which matters a lot for code with legitimate license headers or non-prose strings. It's been in production use since 2013 (originally Ruby, rewritten in Rust) at real sites like Ruby China and V2EX, and the MDN Translated Content integration is a solid maturity signal. Fast enough that the 30K-file MDN repo runs in ~8 seconds.
Scope is narrow and specific to CJK/English mixing — if that's not your problem, this tool does nothing for you. Spellcheck is explicitly marked experimental and only available through the LSP, not the main CLI lint/fix path. Single maintainer (huacnlee) for a project with this many language bindings is a bus-factor risk, especially for the native extensions (Ruby, Java, Python) that need rebuilding against new Rust ABI/runtime versions. The per-language grammars are hand-written PEG files, which means new or unusual syntax in any of the 28 languages requires grammar changes rather than falling back gracefully.
We read the README, the directory tree and a key source file in October 2026. Nobody pays to be listed. How we judge →