Overview of cpd-tokenizer
masterThe cpd-tokenizer crate provides language-aware tokenization for over 200 programming languages. It generates token streams designed specifically for duplicate detection in copy/paste detection workflows.
Key features include:
- Generic tokenizer: Handles comments, strings, keywords, and punctuation.
- JS/TS/JSX/TSX tokenizer: Uses an Oxc-based implementation.
- Markdown tokenizer: Supports code fences, front matter, and embedded languages.
- SFC tokenizer: Supports Vue, Svelte, and Astro files.
- Embedded language detection: Enables cross-format tokenization for languages embedded within other formats.