Skip to content

Reference

Facts about what the packages contain and how they behave. For tasks, see the how-to guides; for the reasoning behind any of it, see Explanation.

Packages

Package Import Contents Environment
@idfkit/core @idfkit/core Parsing, writing, the object model. Synchronous, no I/O. Anywhere
@idfkit/core/node @idfkit/core/node loadIdf, saveIdf, schema discovery from disk Node
@idfkit/schemas @idfkit/schemas Schema bundle, Schema, httpSource Anywhere
Generated types @idfkit/core/types/v26-1 Per-version interfaces and TypeMap Types only

@idfkit/core re-exports Schema, SchemaBundle, and httpSource from @idfkit/schemas, so a browser application usually needs only the one import.

Other reference

  • Supported versions — the 17 bundled releases and how a version string in a file is matched to one of them.
  • Slim schema format — what the bundle keeps, what it drops, and the single-letter keys.

How the API pages are produced

The three API pages are generated from the TypeScript sources by TypeDoc, read into the site through mkdocstrings-typescript. Nothing on them is hand-maintained, so a symbol appears because it is exported and its description is the TSDoc comment above it.

That handler is at an early stage and renders less structure than its Python counterpart: prose, headings, and the symbol tree are faithful, but parameter lists and full type signatures are sparse. Where a signature matters, this documentation states it in prose rather than relying on the generated page.