Introduction

Kineta is a non-SQL-based relational database which is designed to stay out of your way and make your life easier. For much more information on what Kineta is all about, see the WhyKineta wiki page. To start using Kineta, see GettingStarted.

Reference Documentation is now also available.

Feature Overview

• Written in Python, distributed as a normal Python package.
• "Do what I want" semantics:
• To create a table, insert data.
• To add a column, update data.
• To remove a column, update all its values to 'None'.
• Put any data type in any column.
• Querying a non-existant column returns None. Querying a non-existant table returns 0 rows.
• No more schema migration problems!
• Automatically creates indexes based on past queries, drops indexes when they are no longer in use. Indexes can also be manually created/frozen.
• Robust relational query engine with joins/aggregate functions. Optimized to return large result sets from queries.
• Queries are specified as Python dictionaries/lists or JSON; easily manipulate queries as data structures.
• No reserved words, disallowed characters, or quoting issues. Any unicode string can be a table or column name.
• Can be embedded in Python apps or hosted in any WSGI-compliant web server.
• Communicate via open standards: JSON over HTTP.
• Simple, clean, secure: User-friendly API. No embedded programming languages; keep your data in your database and your code in one place.

Status

Currently Kineta is not ready for general use. The code is coming together well and is very usable, but a couple major things still need to be implemented. Thanks for your interest and check back here for updates.