Adding a language

Extractors are plugins. Each is one object with:

• globs: the files it claims
• matches(file): whether it handles a given file
• analyze(abs, rel): returns symbols, imports and re-exports

Everything downstream (linking, snapshotting, diffing, reviewing, the viewer) works on that shape. A new language is one new file in src/extractors/, registered in src/extractors/index.ts.

An extractor with a fixed cost per call can also implement analyzeMany(files). A scan then calls it once for all its files instead of once per file. That is how the Python extractor runs a single interpreter for the whole project.

Development setup

git clone https://github.com/kosminog/hiarky.git
cd hiarky
npm install
npm run build
npm link   # makes `hiarky` available globally from this checkout
npm test

npm run dev rebuilds on change and npm run typecheck type-checks without emitting.

Roadmap

• A tree-sitter backend to deepen the shallow-scanned languages (call graphs, nested declarations)
• Per-symbol history and compare-any-two-snapshots in the viewer