maximdanilchenko / radish

Simple in-memory database and its client written in python3/asyncio for python3/asyncio

Radish

Simple in-memory database and its client written in python3/asyncio for python3/asyncio

Supports str, bytes, int, list, tuple data types

Quick start

Server

Run RadishDB Server:

from radish.database import Server

server = Server(host="127.0.0.1", port=7272)
server.run()

After that you will see an awesome output:

_____Serving RadishDB on 127.0.0.1:7272_____
 _  ___  _ ___   ___  ___  ___  _  ___  _ _ 
| ||_ _||// __> | . \| . || . \| |/ __>| | |
| | | |   \__ \ |   /|   || | || |\__ \|   |
|_| |_|   <___/ |_\_\|_|_||___/|_|<___/|_|_|

That means that server is ready for handling connections.

Client

Client with one connection:

from radish.client import Connection


async def run_client():
    con = Connection(host="127.0.0.1", port=7272)
    assert await con.set("my_key", "my_val") == 1
    assert await con.get("my_key") == "my_val"
    assert await con.echo("hello") == "hello"
    assert await con.delete("my_key") == 1
    await con.close()
    # using "async with" statement:
    async with Connection(host="127.0.0.1", port=7272) as con:
        assert await con.mset(k1=1, k2=b"2", k3="3") == "OK"
        assert await con.mget("k1", "k3", "k2") == [1, "3", b"2"]
        assert await con.ping() == "PONG"
        assert await con.exists("k2") == 1
        assert await con.flush() == 3

Client with connection pooling:

import asyncio

from radish.client import ConnectionPool, Connection


async def run_client(pool: ConnectionPool):
    async with pool.acquire() as con:  # type: Connection
        assert await con.ping() == "PONG"


async def run_pool():
    async with ConnectionPool(host="127.0.0.1",
                              port=7272, 
                              min_size=5, 
                              max_size=50) as pool:
        clients = [run_client(pool) for _ in range(1000)]
        await asyncio.gather(*clients)

Find more examples here

Contents:

Files	Description
database	Memory Storage with some most common operations and socket Server with asyncio
client	Client Connection and ConnectionPool implementations
protocol.py	Simplified to pythonic one RESP (REdis Serialization Protocol) realization

Why?

• After I read this article I was inspired to do something like it, but with asyncio and more pythonic, with more commands and with whole client stuff (such as connection pool).
• It was interesting for me to understand how database exchange with client apps is made and how these tools work on client side.

Main parts of this work/studying:

• RESP protocol, which is used for both client and database
• Async socket server based on python3/asyncio
• In-memory database - it is toy-database for now, but interesting for implementation and for benchmarking this whole solution
• Client side implementation with Async Connection Pool. It was most interesting part for me

Pypi package/usage in real tasks

While Radish DB is in development it is not recommended to use it in some real tasks. There are some things I should do before - benchmarks, tests and some fixes and issues are in backlog to implement.

After all it is not a database for production apps - it is redis mini clone, written in Python. But I thing it will be good for prototyping or for small apps because of its simple installation and starting - it is lightweight and has no dependencies. But overall it will be possible only after benchmarks and tests will be done.