📘 EndpointExtractor

English

English Version

🔹 Overview

EndpointExtractor is a Node.js library built on top of Tree-sitter and a custom ApiGenerator. It scans Java(etc.) source code and extracts API endpoint information (e.g., method, URL, parameters, request/response schema), making it useful for API documentation, reverse engineering, and testing automation.

✨ Features

• 🚀 Automatic Endpoint Extraction – Supports HTTP annotations (@GetMapping, @PostMapping, etc.)
• 📦 Structured Metadata – Extracts parameters from headers, query, path, cookies, and body
• 🧩 Java Type → JSON Schema conversion
• 📑 Stable Unique Endpoint ID generation
• ⚡ Robust & Scalable – Designed for large Java projects

📥 Installation

npm install endpoint-extractor

⚡ Usage Example

const { EndpointExtractor } = require("endpoint-extractor");

(async () => {
  const extractor = new EndpointExtractor({ rootDir: "./my-java-project" });
  const { list, stats, issues } = await extractor.extractEndpoints();

  console.log("Extracted Endpoints:", list);
  console.log("Project Stats:", stats);
  console.log("Issues:", issues);
})();

📊 Example Output

{
  "name": "save",
  "method": "POST",
  "url": "/role/save",
  "target_id": "1839471982734",
  "request": {
    "query": { "parameter": [] },
    "body": {
      "mode": "json",
      "raw": "{\"id\":1,\"name\":\"Admin\"}",
      "raw_schema": { "type": "object", "properties": { ... } }
    }
  }
}

🔧 Options

| Option | Type | Default | Description | | ----------- | ------ | --------------- | ---------------------------------- | | rootDir | string | process.cwd() | Root directory of the Java project | | concurrency | number | 32 | File parsing concurrency level |

📜 License

MIT License © 2025