vscode-ajsbutler

The vscode-ajsbutler extension offers a convenient way to display definition information for JP1/AJS3 from Hitachi in a human-readable format.

Features

• Displays JP1/AJS3 definition information in an easily readable format.
• Set the editor’s language mode to ‘jp1ajs’ to view the formatted definition items.
• Switch the editor to enable a list-style display of the information.
• Select which items to display in the unit list.
• View unit definitions.
• Save data in CSV format.
• Switch the editor to enable a flow-style display of the information.
• Highlights the current node, ancestor path, and root jobnet more clearly in the flow viewer.
• Supports web extensions.

Recent Updates

• Flow viewer now supports inline nested jobnet expansion, current-scope flow search, and list/flow bridge navigation.
• Flow viewer visuals and layout behavior are more stable in both desktop and web hosts.

Extension Settings

This extension does not require any specific configuration settings.

Usage

To use this extension:

1. Install the extension.
2. Set the editor’s language mode to ‘jp1ajs’ to view the formatted definitions.
3. Switch the editor to enable either a list-style or flow-style display.

Development

This repository uses Specification-Driven Development (SDD) as the standard process for non-trivial changes.

• SDD workflow and document roles: docs/specs/README.md
• Agent-facing repository rules and routing: AGENTS.md
• Durable behavior contracts: docs/requirements/use-cases/
• Copilot CLI entry point: .github/copilot-instructions.md
• Codex workflows: .codex/skills/

Keep README as an overview and command reference. Detailed development rules, approval gates, feature artifact responsibilities, and agent routing live in the documents above.

Browser-based extension testing uses @vscode/test-web, which currently requires Node.js 20 or later. If you manage Node with nodebrew, switch to a Node 20 release before running browser tooling:

nodebrew install-binary v20.19.0
nodebrew use v20.19.0
hash -r
corepack enable
pnpm install
pnpm exec playwright install chromium-headless-shell

After switching, you can run:

pnpm run qlty
pnpm run test:full
pnpm test
pnpm run test:web
pnpm run build

ANTLR parser generation is explicit. Normal build and test commands consume the committed parser artifacts in src/generate/parser. When changing src/antlr/*.g4, ANTLR command options, or the generator version, run:

pnpm run antlr4ts

Then commit any generated parser artifact changes with the grammar change.

Desktop and web test commands are compatible from a clean checkout:

• pnpm test prepares the desktop extension and editor bundles, compiles tests, then runs the desktop extension tests.
• pnpm run test:web prepares the web extension and editor bundles, compiles tests, then runs the web extension smoke tests.
• pnpm run test:full prepares all development bundles and compiled tests once, then runs both desktop and web test runners.

The sample/ directory contains reusable JP1/AJS definition files for parser, normalization, unit-list, and flow-graph regression tests. Prefer those shared fixtures over ad hoc large inline definitions when adding broader coverage.

For manual browser-side verification during development, use the VS Code launch configuration Launch Extension(web) in .vscode/launch.json.

pnpm run test:web runs the extension test suite against VS Code for the Web in headless Chromium. pnpm run test:full prepares development bundles and compiled tests once before running both desktop and web extension tests. GitHub Actions also runs pnpm run lint:md, pnpm run build, pnpm run test:compile, pnpm run test:desktop:run, and pnpm run test:web:run on pull requests. The workflow caches Playwright browser downloads while still running the Playwright install command, so cache misses remain valid.

For AI Agents

Agent development rules are centralized in AGENTS.md and docs/specs/README.md. README intentionally stays at overview level to avoid duplicating the SDD workflow.

Telemetry

This extension collects telemetry data to improve the experience of using this extension with VS Code. We collect anonymous operational metadata such as extension lifecycle, command usage, viewer readiness or close events, and anonymous viewer action and search outcomes such as CSV export, navigation, or no-match searches, anonymous diagnostic categories and hover outcomes, and coarse performance buckets for operations such as parsing, unit-list build, flow graph build, and CSV export. We do not collect names, addresses, paths, definition contents, diagnostic messages, hover token text, search text, commands, server names, credentials, or other personal identifiers. The extension respects the telemetry.enableTelemetry setting. You can learn more about that setting in our FAQ.

License

MIT