# msghandler - Cross-Platform Bi-Directional Data Bridge

A high-performance, bi-directional data bridge for **Julia**, **JavaScript**, **Python**, and **MicroPython** applications using NATS (Core & JetStream), implementing the Claim-Check pattern for large payloads.

[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](https://opensource.org/licenses/MIT)
[![NATS](https://img.shields.io/badge/NATS-Enabled-green.svg)](https://nats.io)

---

## Table of Contents

- [Overview](#overview)
- [Cross-Platform Support](#cross-platform-support)
- [Features](#features)
- [Quick Start](#quick-start)
- [API Reference](#api-reference)
- [Payload Types](#payload-types)
- [Cross-Platform Examples](#cross-platform-examples)
- [Testing](#testing)
- [Documentation](#documentation)
- [License](#license)

---

## Overview

msghandler enables seamless communication across multiple platforms through NATS, with intelligent transport selection based on payload size:

| Transport | Payload Size | Method |
|-----------|--------------|--------|
| **Direct** | < 500KB | Sent directly via NATS (Base64 encoded) |
| **Link** | ≥ 500KB | Uploaded to HTTP file server, URL sent via NATS |

### Use Cases

- **Chat Applications**: Text, images, audio, video in a single message
- **File Transfer**: Efficient transfer of large files using claim-check pattern
- **IoT/Embedded**: Sensor data, telemetry, and analytics pipelines (MicroPython)
- **Cross-Platform Communication**: Interoperability between Julia, JavaScript, Python, and MicroPython systems

---

## Cross-Platform Support

| Platform | Implementation | Features |
|----------|----------------|----------|
| **Julia** | [`src/msghandler.jl`](src/msghandler.jl) | Full feature set, Arrow IPC, multiple dispatch |
| **JavaScript (Node.js)** | [`src/msghandler_ssr.js`](src/msghandler_ssr.js) | Node.js, async/await, Arrow IPC |
| **JavaScript (Browser)** | [`src/msghandler_csr.js`](src/msghandler_csr.js) | Browser, WebSocket NATS, async/await, JSON table only |
| **Python** | [`src/msghandler.py`](src/msghandler.py) | Desktop Python, asyncio, type hints, Arrow IPC |
| **MicroPython** | [`src/msghandler_mpy.py`](src/msghandler_mpy.py) | Memory-constrained, synchronous API |

### Platform Comparison

| Feature | Julia | JavaScript | JavaScript (Browser) | Python | MicroPython |
|---------|-------|------------|----------------------|--------|-------------|
| Multiple Dispatch | ✅ Native | ❌ | ❌ | ❌ | ❌ |
| Async/Await | ❌ | ✅ Native | ✅ Native | ✅ Native | ⚠️ (uasyncio) |
| Type Safety | ✅ Strong | ⚠️ (TypeScript) | ⚠️ (TypeScript) | ✅ (Type hints) | ❌ |
| Arrow IPC | ✅ Native | ✅ Native | ❌ (Browser incompatible) | ✅ Native | ❌ |
| JSON Table | ✅ | ✅ | ✅ (Only table type) | ✅ | ⚠️ (Limited) |
| Direct Transport | ✅ | ✅ | ✅ | ✅ | ✅ |
| Link Transport | ✅ | ✅ | ✅ | ✅ | ⚠️ (Limited) |
| Handler Functions | ✅ | ✅ | ✅ | ✅ | ✅ |
| Cross-Platform API | ✅ | ✅ | ✅ | ✅ | ✅ |
| WebSocket NATS | ❌ | ❌ | ✅ | ❌ | ❌ |

---

## Features

- ✅ **Cross-platform messaging** for Julia, JavaScript, Python, and MicroPython applications
- ✅ **Bi-directional messaging** with request-reply patterns
- ✅ **Multi-payload support** - send multiple payloads with different types in one message
- ✅ **Automatic transport selection** - direct vs link based on payload size
- ✅ **Claim-Check pattern** for payloads ≥ 500KB
- ✅ **Apache Arrow IPC** support for tabular data (Desktop: Julia/Python/Node.js)
- ✅ **JSON Table** support for tabular data (All platforms including Browser)
- ✅ **Exponential backoff** for reliable file server downloads
- ✅ **Correlation ID tracking** for message tracing
- ✅ **Reply-to support** for request-response patterns
- ✅ **Handler function abstraction** - pluggable file server implementations (Plik, AWS S3, custom)

---

## Quick Start

### Prerequisites

1. **NATS Server** - Install and run a NATS server:
   ```bash
   docker run -p 4222:4222 nats:latest
   ```

2. **HTTP File Server** (optional, for large payloads) - Install and run a file server:
   ```bash
   # Using Plik
   docker run -p 8080:8080 -v /tmp/fileserver:/var/lib/plik -e PLIK_ADMIN_PASSWORD=admin plik/plik
   
   # OR using simple Python HTTP server
   mkdir -p /tmp/fileserver
   python3 -m http.server 8080 --directory /tmp/fileserver
   ```

### Send Your First Message

#### Julia

```julia
using msghandler, NATS

payload_1 = ("test_message", "Hello World", "text")

file_path_large_image = "./test/large_image.png"
file_data_large_image = read(file_path_large_image)
filename_large_image = basename(file_path_large_image)
payload_2 = (filename_large_image, file_data_large_image, "binary")

payloads = [payload_1, payload_2]  # payloads to be send format is [(dataname, data, type), ...] tuples.
envelope, envelope_json_str = msghandler.smartpack("test.topic", 
                                                  payloads; 
                                                  broker_url="nats.yiem.cc",
                                                  fileserver_url="http://192.168.88.104:8080")
conn = NATS.connect("nats.yiem.cc")  # Create NATS connection
NATS.publish(conn, "test.topic", envelope_json_str)  # Publish message to NATS
NATS.drain(conn)
```

#### JavaScript (Node.js)

```javascript
import msghandler from './src/msghandler_ssr.js';

const data = [["message", "Hello World", "text"]];
const [env, env_json_str] = await msghandler.smartpack(
    "/chat/room1",
    data,
    { broker_url: "nats://localhost:4222" }
);
console.log("Message sent!");
```

#### JavaScript (Browser)

```javascript
import msghandler from './src/msghandler_csr.js';

const data = [["message", "Hello World", "text"]];
const [env, env_json_str] = await msghandler.smartpack(
    "/chat/room1",
    data,
    { broker_url: "ws://localhost:4222" }
);
console.log("Message sent!");
```

#### Python

```python
from msghandler import smartpack

data = [("message", "Hello World", "text")]
env, env_json_str = await smartpack(
    "/chat/room1",
    data,
    broker_url="nats://localhost:4222"
)
print("Message sent!")
```

#### MicroPython

```python
from msghandler import smartpack

data = [("message", "Hello World", "text")]
env, env_json_str = smartpack(
    "/chat/room1",
    data,
    broker_url="nats://localhost:4222",
    size_threshold=100000  # 100KB for MicroPython
)
print("Message sent!")
```

### Receive Your First Message

#### Julia

```julia
using msghandler, NATS

conn = NATS.connect("nats.yiem.cc")
NATS.subscribe(conn, "test.topic") do msg
  println("Received message on $(msg.subject)")
  payloads = String(msg.payload)
  
  # Use NATSBridge.smartreceive to handle the data
  # API: smartreceive(msg, download_handler; max_retries, base_delay, max_delay)
  envelope = msghandler.smartunpack(
    payloads;
    max_retries = 5,
    base_delay = 100,
    max_delay = 5000
  )
  println(envelope.payloads[1])
end
```

---

## API Reference

### Unified API Standard

All platforms use the same input/output format for payloads:

**Input format for `smartpack`:**
```
[(dataname1, data1, type1), (dataname2, data2, type2), ...]
```

**Output format for `smartunpack`:**
```json
{
  "correlation_id": "...",
  "msg_id": "...",
  "timestamp": "...",
  "send_to": "...",
  "msg_purpose": "...",
  "sender_name": "...",
  "sender_id": "...",
  "receiver_name": "...",
  "receiver_id": "...",
  "reply_to": "...",
  "reply_to_msg_id": "...",
  "broker_url": "...",
  "metadata": {...},
  "payloads": [(dataname1, data1, type1), (dataname2, data2, type2), ...]
}
```

### smartpack

Sends data either directly via NATS or via a fileserver URL, depending on payload size.

#### Julia

```julia
using msghandler

env, env_json_str = msghandler.smartpack(
    subject::String,
    data::AbstractArray{Tuple{String, Any, String}};
    broker_url::String = "nats://localhost:4222",
    fileserver_url = "http://localhost:8080",
    fileserver_upload_handler::Function = plik_oneshot_upload,
    size_threshold::Int = 500_000,
    correlation_id::String = string(uuid4()),
    msg_purpose::String = "chat",
    sender_name::String = "msghandler",
    receiver_name::String = "",
    receiver_id::String = "",
    reply_to::String = "",
    reply_to_msg_id::String = "",
    is_publish::Bool = true,
    NATS_connection::Union{NATS.Connection, Nothing} = nothing,
    msg_id::String = string(uuid4()),
    sender_id::String = string(uuid4())
)
# Returns: ::Tuple{msg_envelope_v1, String}
```

#### JavaScript (Node.js)

```javascript
import msghandler from './src/msghandler_ssr.js';

const [env, env_json_str] = await msghandler.smartpack(
    subject,
    data,  // Array of [dataname, data, type] tuples
    {
        broker_url: 'nats://localhost:4222',
        fileserver_url: 'http://localhost:8080',
        fileserver_upload_handler: msghandler.plikOneshotUpload,
        size_threshold: 500_000,
        correlation_id: uuidv4(),
        msg_purpose: 'chat',
        sender_name: 'msghandler',
        receiver_name: '',
        receiver_id: '',
        reply_to: '',
        reply_to_msg_id: '',
        is_publish: true,
        nats_connection: null,
        msg_id: uuidv4(),
        sender_id: uuidv4()
    }
);
// Returns: Promise<[env, env_json_str]>
```

#### JavaScript (Browser)

```javascript
import msghandler from './src/msghandler_csr.js';

const [env, env_json_str] = await msghandler.smartpack(
    subject,
    data,
    {
        broker_url: 'ws://localhost:4222',
        fileserver_url: 'http://localhost:8080',
        fileserver_upload_handler: msghandler.plikOneshotUpload,
        size_threshold: 500_000,
        correlation_id: uuidv4(),
        msg_purpose: 'chat',
        sender_name: 'msghandler',
        receiver_name: '',
        receiver_id: '',
        reply_to: '',
        reply_to_msg_id: '',
        is_publish: true,
        nats_connection: null,
        msg_id: uuidv4(),
        sender_id: uuidv4()
    }
);
// Returns: Promise<[env, env_json_str]>
```

#### Python

```python
from msghandler import msghandler

env, env_json_str = await msghandler.smartpack(
    subject: str,
    data: List[Tuple[str, Any, str]],
    broker_url: str = "nats://localhost:4222",
    fileserver_url: str = "http://localhost:8080",
    fileserver_upload_handler: Callable = plik_oneshot_upload,
    size_threshold: int = 500_000,
    correlation_id: str = None,
    msg_purpose: str = "chat",
    sender_name: str = "msghandler",
    receiver_name: str = "",
    receiver_id: str = "",
    reply_to: str = "",
    reply_to_msg_id: str = "",
    is_publish: bool = True,
    nats_connection: Any = None,
    msg_id: str = None,
    sender_id: str = None
)
# Returns: Tuple[Dict, str]
```

#### MicroPython

```python
from msghandler import msghandler

# Limited to direct transport (< 100KB threshold)
env, env_json_str = msghandler.smartpack(
    subject,
    data,  # List of (dataname, data, type) tuples
    broker_url="nats://localhost:4222",
    size_threshold=100000  # Lower threshold for memory constraints
)
# Returns: Tuple[Dict, str]
```

### smartunpack

Receives and processes messages from NATS, handling both direct and link transport.

#### Julia

```julia
using msghandler

env = msghandler.smartunpack(
    msg::NATS.Msg;
    fileserver_download_handler::Function = _fetch_with_backoff,
    max_retries::Int = 5,
    base_delay::Int = 100,
    max_delay::Int = 5000
)
# Returns: ::JSON.Object{String, Any}
```

#### JavaScript (Node.js)

```javascript
import msghandler from './src/msghandler_ssr.js';

const env = await msghandler.smartunpack(
    msg,
    {
        fileserver_download_handler: msghandler.fetchWithBackoff,
        max_retries: 5,
        base_delay: 100,
        max_delay: 5000
    }
);
// Returns: Promise<env_object>
```

#### JavaScript (Browser)

```javascript
import msghandler from './src/msghandler_csr.js';

const env = await msghandler.smartunpack(
    msg,
    {
        fileserver_download_handler: msghandler.fetchWithBackoff,
        max_retries: 5,
        base_delay: 100,
        max_delay: 5000
    }
);
// Returns: Promise<env_object>
```

#### Python

```python
from msghandler import msghandler

env = await msghandler.smartunpack(
    msg,
    fileserver_download_handler=fetch_with_backoff,
    max_retries=5,
    base_delay=100,
    max_delay=5000
)
# Returns: Dict with "payloads" key
```

#### MicroPython

```python
from msghandler import msghandler

env = msghandler.smartunpack(
    msg,
    fileserver_download_handler=_sync_fileserver_download,
    max_retries=3,
    base_delay=100,
    max_delay=1000
)
# Returns: Dict with "payloads" key
```

---

## Payload Types

| Type | Julia | JavaScript | Python | MicroPython | Description |
|------|-------|------------|--------|-------------|-------------|
| `text` | `String` | `string` | `str` | `str` | Plain text strings |
| `dictionary` | `Dict`, `NamedTuple` | `Object`, `Array` | `dict`, `list` | `dict` | JSON-serializable dictionaries |
| `arrowtable` | `DataFrame`, `Arrow.Table` | ❌ (Browser), ✅ (Node.js) | `pandas.DataFrame` | ❌ | Tabular data (Arrow IPC) |
| `jsontable` | `DataFrame`, `Vector{NamedTuple}` | `Array<Object>` | `list[dict]` | ⚠️ | Tabular data (JSON) - **Only table type in Browser** |
| `image` | `Vector{UInt8}` | `Uint8Array`, `Buffer` | `bytes` | `bytearray` | Image data (PNG, JPG) |
| `audio` | `Vector{UInt8}` | `Uint8Array`, `Buffer` | `bytes` | `bytearray` | Audio data (WAV, MP3) |
| `video` | `Vector{UInt8}` | `Uint8Array`, `Buffer` | `bytes` | `bytearray` | Video data (MP4, AVI) |
| `binary` | `Vector{UInt8}`, `IOBuffer` | `Uint8Array`, `Buffer` | `bytes`, `bytearray` | `bytearray` | Generic binary data |

---

## Cross-Platform Examples

### Example 1: Chat with Mixed Content

Send text, image, and large file in one message.

#### Julia

```julia
using msghandler

data = [
    ("message_text", "Hello!", "text"),
    ("user_avatar", image_data, "image"),
    ("large_document", large_file_data, "binary")
]

env, env_json_str = smartpack("/chat/room1", data; fileserver_url="http://localhost:8080")
```

#### JavaScript (Node.js)

```javascript
import msghandler from './src/msghandler_ssr.js';

const data = [
    ["message_text", "Hello!", "text"],
    ["user_avatar", imageData, "image"],
    ["large_document", largeFileData, "binary"]
];

const [env, env_json_str] = await msghandler.smartpack(
    "/chat/room1",
    data,
    { fileserver_url: 'http://localhost:8080' }
);
```

#### JavaScript (Browser)

```javascript
import msghandler from './src/msghandler_csr.js';

const data = [
    ["message_text", "Hello!", "text"],
    ["user_avatar", imageData, "image"],
    ["large_document", largeFileData, "binary"]
];

const [env, env_json_str] = await msghandler.smartpack(
    "/chat/room1",
    data,
    { broker_url: 'ws://localhost:4222', fileserver_url: 'http://localhost:8080' }
);
```

#### Python

```python
from msghandler import msghandler

data = [
    ("message_text", "Hello!", "text"),
    ("user_avatar", image_data, "image"),
    ("large_document", large_file_data, "binary")
]

env, env_json_str = await msghandler.smartpack(
    "/chat/room1",
    data,
    fileserver_url="http://localhost:8080"
)
```

### Example 2: Dictionary Exchange

Send configuration data between platforms.

#### Julia

```julia
using msghandler

config = Dict(
    "wifi_ssid" => "MyNetwork",
    "wifi_password" => "password123",
    "update_interval" => 60
)

data = [("config", config, "dictionary")]
env, env_json_str = smartpack("/device/config", data)
```

#### JavaScript (Node.js)

```javascript
import msghandler from './src/msghandler_ssr.js';

const config = {
    wifi_ssid: "MyNetwork",
    wifi_password: "password123",
    update_interval: 60
};

const [env, env_json_str] = await msghandler.smartpack(
    "/device/config",
    [["config", config, "dictionary"]]
);
```

#### Python

```python
from msghandler import msghandler

config = {
    "wifi_ssid": "MyNetwork",
    "wifi_password": "password123",
    "update_interval": 60
}

data = [("config", config, "dictionary")]
env, env_json_str = await msghandler.smartpack("/device/config", data)
```

### Example 3: Table Data (Arrow IPC)

Send tabular data using Apache Arrow IPC format.

#### Julia

```julia
using msghandler
using DataFrames

df = DataFrame(
    id = [1, 2, 3],
    name = ["Alice", "Bob", "Charlie"],
    score = [95, 88, 92]
)

data = [("students", df, "arrowtable")]
env, env_json_str = smartpack("/data/analysis", data)
```

#### JavaScript (Node.js)

```javascript
import msghandler from './src/msghandler_ssr.js';

const df = [
    { id: 1, name: "Alice", score: 95 },
    { id: 2, name: "Bob", score: 88 },
    { id: 3, name: "Charlie", score: 92 }
];

const [env, env_json_str] = await msghandler.smartpack(
    "/data/analysis",
    [["students", df, "arrowtable"]]
);
```

#### Python

```python
from msghandler import msghandler
import pandas as pd

df = pd.DataFrame({
    "id": [1, 2, 3],
    "name": ["Alice", "Bob", "Charlie"],
    "score": [95, 88, 92]
})

data = [("students", df, "arrowtable")]
env, env_json_str = await msghandler.smartpack("/data/analysis", data)
```

#### JavaScript (Browser)

```javascript
import msghandler from './src/msghandler_csr.js';

// Browser uses jsontable (JSON array of objects) instead of arrowtable
// Apache Arrow is not compatible with browsers
const df = [
    { id: 1, name: "Alice", score: 95 },
    { id: 2, name: "Bob", score: 88 },
    { id: 3, name: "Charlie", score: 92 }
];

const [env, env_json_str] = await msghandler.smartpack(
    "/data/analysis",
    [["students", df, "jsontable"]],  // Use jsontable for browser
    { broker_url: 'ws://localhost:4222' }
);
```

### Example 4: Request-Response Pattern

Bi-directional communication with reply-to support.

#### Julia

```julia
using msghandler

# Requester
env, env_json_str = smartpack(
    "/device/command",
    [("command", Dict("action" => "read_sensor"), "dictionary")];
    broker_url="nats://localhost:4222",
    reply_to="/device/response"
)

# Receiver (in separate application)
msg = NATS.subscription.next()
env = smartunpack(msg)
# Process request and send response
response_env, response_json = smartpack(
    "/device/response",
    [("result", Dict("value" => 42), "dictionary")],
    reply_to="/device/command",
    reply_to_msg_id=env["msg_id"]
)
```

#### JavaScript (Node.js)

```javascript
import msghandler from './src/msghandler_ssr.js';

// Requester
const [env, env_json_str] = await msghandler.smartpack(
    "/device/command",
    [["command", { action: "read_sensor" }, "dictionary"]],
    { broker_url: 'nats://localhost:4222', reply_to: '/device/response' }
);

// Receiver (in separate application)
// const msg = await natsConsumer.next();
// const env = await msghandler.smartunpack(msg);
// Process request and send response
// const response_env, response_json = await msghandler.smartpack(
//     "/device/response",
//     [["result", { value: 42 }, "dictionary"]],
//     { reply_to: '/device/command', reply_to_msg_id: env.msg_id }
// );
```

#### Python

```python
from msghandler import msghandler

# Requester
env, env_json_str = await msghandler.smartpack(
    "/device/command",
    [("command", {"action": "read_sensor"}, "dictionary")],
    broker_url="nats://localhost:4222",
    reply_to="/device/response"
)

# Receiver (in separate application)
# msg = await nats_consumer.next()
# env = await msghandler.smartunpack(msg)
# Process request and send response
# response_env, response_json = await msghandler.smartpack(
#     "/device/response",
#     [("result", {"value": 42}, "dictionary")],
#     reply_to="/device/command",
#     reply_to_msg_id=env["msg_id"]
# )
```

---

## Testing

### Test File Organization

| Platform | Sender Tests | Receiver Tests |
|----------|--------------|----------------|
| **Julia** | `test/test_julia_*_sender.jl` | `test/test_julia_*_receiver.jl` |
| **JavaScript** | `test/test_js_*_sender.js` | `test/test_js_*_receiver.js` |
| **Python** | `test/test_py_*_sender.py` | `test/test_py_*_receiver.py` |

### Run Tests

#### Julia

```bash
# Text message exchange
julia test/test_julia_text_sender.jl
julia test/test_julia_text_receiver.jl

# Dictionary exchange
julia test/test_julia_dict_sender.jl
julia test/test_julia_dict_receiver.jl

# File transfer
julia test/test_julia_file_sender.jl
julia test/test_julia_file_receiver.jl

# Mixed payload types
julia test/test_julia_mix_payloads_sender.jl
julia test/test_julia_mix_payloads_receiver.jl

# Table exchange
julia test/test_julia_table_sender.jl
julia test/test_julia_table_receiver.jl
```

#### JavaScript (Node.js)

```bash
# Text message exchange
node test/test_js_text_sender.js
node test/test_js_text_receiver.js

# Dictionary exchange
node test/test_js_dictionary_sender.js
node test/test_js_dictionary_receiver.js

# Binary transfer
node test/test_js_binary_sender.js
node test/test_js_binary_receiver.js

# Table exchange
node test/test_js_table_sender.js
node test/test_js_table_receiver.js
```

#### Python

```bash
# Text message exchange
python3 test/test_py_text_sender.py
python3 test/test_py_text_receiver.py

# Dictionary exchange
python3 test/test_py_dictionary_sender.py
python3 test/test_py_dictionary_receiver.py

# Binary transfer
python3 test/test_py_binary_sender.py
python3 test/test_py_binary_receiver.py

# Table exchange
python3 test/test_py_table_sender.py
python3 test/test_py_table_receiver.py
```

---

## Browser Deployment

### Using with Node.js Build Tools

The browser implementation (`src/msghandler_csr.js`) can be bundled for production deployment using modern JavaScript build tools.

#### Prerequisites

```bash
# Install the browser-compatible NATS client
npm install nats.ws
```

#### Vite (Recommended)

```bash
npm create vite@latest my-app -- --template vanilla
cd my-app
npm install nats.ws
```

In `vite.config.js`:
```javascript
import { defineConfig } from 'vite';
export default defineConfig({
  resolve: {
    alias: {
      'nats.ws': 'nats.ws/dist/esm/browser.js'
    }
  }
});
```

Build command:
```bash
npm run build  # Outputs to dist/ folder
```

#### Webpack

```bash
npm install webpack webpack-cli --save-dev
npm install nats.ws
```

In `webpack.config.js`:
```javascript
module.exports = {
  entry: './src/index.js',
  output: {
    filename: 'bundle.js',
    path: __dirname + '/dist'
  },
  resolve: {
    alias: {
      'nats.ws': 'nats.ws/dist/esm/browser.js'
    }
  }
};
```

Build command:
```bash
npx webpack
```

#### esbuild (Simple & Fast)

```bash
npm install esbuild nats.ws --save-dev
```

Create `build.js`:
```javascript
import esbuild from 'esbuild';

esbuild.buildSync({
  entryPoints: ['src/msghandler_csr.js'],
  bundle: true,
  outfile: 'dist/msghandler-csr-bundle.js',
  format: 'esm',
  platform: 'browser',
  target: 'es2020'
});
```

Build command:
```bash
node build.js
```

### Using in Your HTML

```html
<!DOCTYPE html>
<html>
<head>
  <title>My App</title>
</head>
<body>
  <script type="module" src="dist/msghandler-csr-bundle.js"></script>
  <script type="module">
    import msghandlerCSR from './dist/msghandler-csr-bundle.js';
    
    // Use the library
    const [env, envJson] = await msghandlerCSR.smartpack(
      "/chat/user/v1/message",
      [["msg", "Hello", "text"]],
      { broker_url: "wss://nats.example.com" }
    );
  </script>
</body>
</html>
```

---

## Documentation

For detailed architecture and implementation information, see:

- [`docs/architecture.md`](docs/architecture.md) - Cross-platform architecture, API parity, platform-specific patterns
- [`docs/requirements.md`](docs/requirements.md) - Business requirements and user stories
- [`docs/spec.md`](docs/spec.md) - Technical specification and contracts
- [`docs/walkthrough.md`](docs/walkthrough.md) - Real-world application building guides

---

## License

MIT License

Copyright (c) 2026 msghandler Contributors

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.