@chestnutlabs/gcode-parser

Worker-safe G-code parse core and session client for the Chestnut Labs G-code Preview stack (design: DD-003). Parses .gcode / .gcode.3mf off the main thread into ToolpathIR, with streaming input, progressive partial previews for large files, resource limits, cancellation, and an adversarial-input corpus behind it.

import { GcodeParseSession } from '@chestnutlabs/gcode-parser';

const session = new GcodeParseSession(); // batteries worker: all dialects + .gcode.3mf
const result = await session.parse(bytes, {
  onPartial: (ir) => render(ir), // progressive preview
  onProgress: (p) => bar(p)
});
render(result.ir);
session.dispose();

Bytes are handed to the worker, not copied. A Uint8Array passed to parse() has its buffer transferred to the worker, so it is empty (byteLength === 0) afterwards. To parse the same file again, for example to switch plates with { plate: 1 }, keep a copy and pass bytes.slice() each time. Passing the already-transferred bytes rejects with E_POST_FAILED; the session stays usable.

Workers

• Batteries default (zero setup): the session creates its own worker via the bundler-native new Worker(new URL('./worker.js', import.meta.url)) pattern — all supported dialect adapters plus .gcode.3mf container support. Vite resolves it out of the box (CI-tested).
• Slim / custom: pass a worker factory for the slim build (no dialects/containers), your own adapter set via createWorkerHandler(...), other bundlers, or strict-CSP environments.

Wire options accept dialect selection, container/plate selection (plate for multi-plate .gcode.3mf), and resource limits. Most applications should consume this through @chestnutlabs/gcode-preview-core or a framework adapter rather than directly.

Part of Chestnut Labs G-code Preview · MIT