Datastorage and access from another perspective.

Just starting out.

The aim is to be able to have a simple container that I can shove data into. The implementation persistence layer would be whatever, but the api would allow me to pretend like I am talking to a schemaless+nosql backend.

• No need to define a model, just work with python dicts (think JSON)
• Quick CRUD; very light on the "R" implementation
• Multiple backend implementations - some may expose additional features beyond common
• Simple and flexible API
• Auto schema migrations when data structure changes
• Pipe data from one backend to another

Loosly based on how I expect to tackle the project

• In Memory
• SQLite
• MongoDb
• MySql
• ... more