duckbridge 0.4.1

A lightweight client-server library fluently extending DuckDB's httpserver setup to Python

DuckBridge

Lightweight Python library enabling fully pythonic interfacing with the DuckDB extension httpserver (extension developed by @quackscience).

What is duckbridge?

DuckDB's httpserver extension turns databases into HTTP OLAP servers; however, setup and access require unix commands and SQL in order to properly set up authentication, install the extension, and communicate.

duckbridge spans this gap, providing a way to natively use httpserver with full pythonic scripting. The DuckDBServer class acts as the bridge operator, connecting to the database, initializing httpserver, and setting authentication values and read/write privileges. DuckDBClient provides a tidy handler which stores client-side authentication and endpoint information, ensuring one-factor authentication by way of knowledge while returning information as a Pandas DataFrame.

How do I use duckbridge?

Duckbridge is split into two main portions: DuckbridgeServer, which handles extension installation/activation, database connection, and client request processing, and DuckbridgeClient, which formats and sends requests to an assumed endpoint, unmarshalling returned data into a convenient format (currently only a Pandas DataFrame).

Examples of how to set up a server or client can be found in more depth at , a repository of tutorials that is constantly being updated.

If you have a use case that you would like to have considered, please and it will be addressed!

Quick start copy-and-paste

####################
### Server setup ###
####################
from duckbridge import *
import os

my_ssh_key_auth : str = "mySuperSecretSSHString"
ssh_userpass_auth : str = "username:password" # !!!The colon is mandatory!!!
my_host : str = "127.0.0.1" # Replace with 0.0.0.0 for public ip access
my_port : int = 8080

bridge : DuckbridgeServer = DuckbridgeServer()
# Assumes readonly permissions and that the database doesn't have `httpserver` installed yet
# If an empty string is provided for auth_info, then no security will be used
bridge.start(os.cwd() + "/path/to/db.db", host=my_host, port=my_port, auth_info=my_ssh_key_auth)


####################
### Client setup ###
####################
from duckbridge import *
import pandas as pd

my_username : str = "username"
my_password : str = "password"

client : DuckbridgeClient = DuckbridgeClient(my_host, my_port, my_username, my_password, my_ssh_key_auth)
data : pd.DataFrame = client.execute("Select * from concept limit 3;", auth="ssh")