@subjektifylabs/subjektify-build

subjektify-build is a core package of the Subjektify toolkit, designed to facilitate the build process the Subjekt model. This package provides tools and utilities to parse, validate, and compile Subjekt model files, ensuring a seamless and efficient build workflow.

Features

• Model Parsing: Parse Subjekt model files (.subjekt or JSON) to generate an Abstract Syntax Tree (AST) or a SubjektifyModel.
• Validation: Validate the structure and semantics of Subjekt models to ensure correctness.
• Compilation: Compile Subjekt models into executable code or other target formats.

Installation

subjektify-build is already included in the subjektify package.

Tasks

npx subjektify build

Environment Extensions

The plugin adds a model object to the Subjektify Runtime Environment.

This object contains both the AST model and the semantic model for the Subjekt files configured.

Usage

subjektify-build is configurable from the subjektify.config.(js|ts). Here’s a basic example of how to use subjektify-build to parse and build a Subjekt model:

import { SubjektifyConfig } from "subjektify";

const config: SubjektifyConfig = {
  namespace: "my.namespace",
  version: "0.1.0",
  license: "MIT",
  build: {
    sources: ["subjects"],
    output: "build",
    projections: [
      {
        abstract: false,
        transformations: [
          {
            type: "excludeByTrait",
            args: { traits: ["internal"] },
          },
        ],
      },
    ],
  },
};

export default config;

How it Works

1. Model Parsing: The build process starts by parsing Subjekt model files from the specified sources. These files should be in .subjekt format.
2. Validation: Once parsed, the models are validated to ensure they adhere to the expected syntax and semantics.
3. Projections and Transformations: The build configuration allows defining projections and transformations that modify or generate different views of the model. These transformations are applied based on the specified projections.
4. Output: Finally, the transformed models are written to the specified output directory.