Memory Object Server

A high-performance Python object sharing pool with distributed capabilities.

Features

• Support for any Python object: Including system-level objects like sockets
• Asynchronous communication: Built on asyncio for high performance
• Minimal API: Simple and intuitive interface
• Distributed support: Multi-node synchronization and discovery
• Auto-reconnection: Reliable TCP socket communication with automatic reconnection
• Thread-safe: Concurrent operations are fully supported
• Port detection: Smart startup with port availability checking

Installation

pip install memory-obj-server

Quick Start

Server

from memory_obj_pool import Server

server = Server(host='0.0.0.0', port=8765)
server.run()

Client

from memory_obj_pool import SyncClient

client = SyncClient(host='localhost', port=8765)

client.put('key', {'data': 'value'})
result = client.get('key')
print(result)  # {'data': 'value'}

Async Client

import asyncio
from memory_obj_pool import Client

async def main():
    client = Client(host='localhost', port=8765)
    await client.put('key', {'data': 'value'})
    result = await client.get('key')
    print(result)

asyncio.run(main())

API Reference

Server

• Server(host='0.0.0.0', port=8765, max_size=10000): Create a server instance
• server.run(): Start the server (blocking)
• await server.start(): Start the server (async)
• await server.stop(): Stop the server

Client

• Client(host='localhost', port=8765, auto_reconnect=True): Create async client
• await client.connect(): Connect to server
• await client.put(key, obj): Store an object
• await client.get(key): Retrieve an object
• await client.remove(key): Remove an object
• await client.clear(): Clear all objects
• await client.size(): Get pool size
• await client.contains(key): Check if key exists

SyncClient

• SyncClient(host='localhost', port=8765, auto_reconnect=True): Create sync client
• client.put(key, obj): Store an object
• client.get(key): Retrieve an object
• client.remove(key): Remove an object
• client.clear(): Clear all objects
• client.size(): Get pool size
• client.contains(key): Check if key exists

Distributed Usage

from memory_obj_pool import DistributedPool

pool = DistributedPool(
    node_id='node1',
    known_nodes=[('localhost', 8765)]
)
pool.start()

Requirements

• Python 3.7+
• dill
• asyncio

License

Apache License 2.0