muxu-io-mqtt-connector 1.1.9

A robust MQTT connector for asynchronous MQTT communication

MQTT Connector

A robust MQTT connector for asynchronous MQTT communication.

Features

• Asynchronous API using Python's asyncio
• Async and sync message callbacks - Native support for both sync and async message handlers
• Automatic reconnection handling
• Message throttling to avoid flooding the broker
• Supports both string and JSON message formats
• Thread-safe async callback scheduling from MQTT background threads
• Customizable logging via callback function

Installation

pip install mqtt-connector

Basic Usage

import asyncio
from mqtt_connector import MqttConnector

async def main():
    # Create a connector instance
    connector = MqttConnector(
        mqtt_broker="mqtt.example.com",
        mqtt_port=1883,
        client_id="example_client"
    )

    # Set up message callback (supports both sync and async)
    async def message_handler(topic: str, message: str):
        print(f"Received: {topic} -> {message}")
        # Async operations are supported
        await asyncio.sleep(0.01)

    connector.set_message_callback(message_handler)

    # Connect to the broker
    connected = await connector.connect()

    # Subscribe to a topic
    await connector.subscribe("example/incoming")

    # Publish a message
    await connector.publish(
        topic="example/outgoing",
        message={"status": "online", "timestamp": "2025-08-03T12:00:00Z"}
    )

    # Let messages process for a bit
    await asyncio.sleep(2)

    # Disconnect
    await connector.disconnect()

if __name__ == "__main__":
    asyncio.run(main())

Message Callbacks

The connector supports both synchronous and asynchronous message callbacks:

Async Callback Example

async def async_message_handler(topic: str, message: str):
    """Async handler can perform I/O operations."""
    print(f"Processing: {topic}")

    # Async operations like database writes, file I/O, etc.
    await process_message_async(message)
    await save_to_database(topic, message)

connector.set_message_callback(async_message_handler)

Sync Callback Example

def sync_message_handler(topic: str, message: str):
    """Sync handler for simple processing."""
    print(f"Received: {topic} -> {message}")

    # Synchronous operations only
    process_message_sync(message)

connector.set_message_callback(sync_message_handler)

Thread Safety

The connector automatically handles thread-safe execution of async callbacks using call_soon_threadsafe(), ensuring proper integration with the asyncio event loop even when MQTT messages arrive on background threads.

Advanced Usage

For more advanced usage examples, check the examples directory in the repository.

API Reference

MqttConnector

connector = MqttConnector(
    mqtt_broker="mqtt.example.com",  # Broker address
    mqtt_port=1883,                  # Broker port
    client_id=None,                  # Client ID (auto-generated if None)
    reconnect_interval=5,            # Seconds between reconnection attempts
    max_reconnect_attempts=-1,       # Maximum reconnection attempts (-1 = infinite)
    throttle_interval=0.1            # Minimum seconds between publishes
)

Methods

• await connector.connect(force_reconnect=False) - Connect to the broker
• await connector.disconnect() - Disconnect from the broker
• await connector.publish(topic, message, qos=0, retain=False) - Publish a message
• await connector.subscribe(topic, qos=0) - Subscribe to a topic
• connector.is_connected() - Check connection status
• connector.set_message_callback(callback) - Set message callback (sync or async)
• connector.set_log_callback(callback) - Set logging callback function

Message Callback Signature

# Sync callback
def callback(topic: str, message: str) -> None:
    pass

# Async callback
async def callback(topic: str, message: str) -> None:
    pass

Contributing

We welcome contributions! Please see our Contributing Guide for detailed instructions on:

• Development setup and workflow
• Commit message conventions
• Pre-submission validation checklist
• CI/CD pipeline overview
• Pull request process

License

This project is licensed under the MIT License - see the LICENSE file for details.