# ShittyServerless Execution Instance 🛠️

**A lightweight Node.js server for executing serverless functions locally or via S3 storage.**  

This project allows developers to deploy and run serverless functions by simply placing files in a storage backend (local filesystem or S3/MinIO). It handles streaming headers, request data, and safe child process execution.

Most notable use being ChattedRooms' newest API rendition.

---

## Features ✨

- Execute serverless functions in **Python, Node.js, or Bash** automatically.
- Supports **local filesystem** or **S3-compatible storage** (e.g., AWS S3, MinIO).
- **Safe child process execution** with timeout handling.
- **Streaming response support** with header parsing and body streaming.
- Flexible **file lookup by extension** or brute-force search.
- Easy integration with **existing serverless workflows**.

---

## Prerequisites 📝

- Node.js v18+  
- NPM (Node Package Manager)  
- Optional: AWS credentials and S3 bucket if not using local storage  

---

## Installation ⚡

1. Clone the repository:

```bash
git clone https://github.com/yourusername/shitty-serverless.git
cd shitty-serverless
```

2. Install dependencies:

```bash
npm install
```

3. Configure your environment by editing `config.sample.js` and then renaming it to `config.js`:

```bash
cp config.sample.js config.js
# Then edit config.js to match your setup
```

> ⚠️ Using local storage is **not recommended for production**.  

---

## Usage 🚀

1. Start the server:

```bash
node index.js
```

2. Open your browser or use `curl`:

```bash
curl http://localhost:3000/yourFunction.py
```

- The server will attempt to execute the requested file if it exists in storage.
- Supported file extensions:  
  - `.sh` → Bash  
  - `.py` → Python 3  
  - `.js` → Node.js

3. If the requested file does not exist, the server will try a **brute-force search** by appending supported extensions.

---

## Example Function (Python) 🐍

```python
import json
import sys

print("Content-Type: application/json")

req = json.loads(input('!!RECVDATA').strip())  # You can only ask for JSON data before body output

print("!!STARTBODY", end="")  # No newline, avoids blank line before body
print(json.dumps(req), end="")  # Echo back request data

sys.exit(2)  # Exit code 2 means 403 Forbidden, but here it is just an example
```

- Place this file in `localstorage/` or S3 bucket.
- Request via HTTP:

```bash
curl http://localhost:3000/example.py
```

- Response will include headers and body streamed directly from the function.

---

## Exit Code Mapping 📊

The server maps child process exit codes to HTTP status codes:

| Exit Code | HTTP Status |
|-----------|-------------|
| 0         | 200 OK      |
| 1         | 500 Internal Server Error |
| 2         | 403 Forbidden |
| 3         | 401 Unauthorized |
| 4         | 404 Not Found |

Other codes default to **500 Internal Server Error**.

---

## Development Tips 💡

- For **debugging**, all child process `stdout` and `stderr` are logged to the server console.  
- Ensure your serverless scripts **emit the `!!STARTBODY` token** to separate headers from body content.  
- Use `!!RECVDATA` in scripts to access the incoming HTTP request object.

---

## License 🛡️

This project is provided **as-is**, without warranty. Use at your own risk.