Morf is a library for cross-platform evolutionary relational database mechanics, database access and database imaging/cloning. It has been developed over many years at Alfa and is used to manage all of our RDBMS-backed applications.

Status

Morf is a brand-new open source project.

It is proven technology and has been in mission-critical production applications for many years as part of our flagship product. However, as an open source project, it is in a very early pre-alpha state, as we go through the process of disentangling it from parts of the rest of our stack and tidying up the code for long-term support. Until then, consider the API unstable.

We are doing this tidying work in the open, because we believe this is the right thing to do. Please get involved if you would like to help steer the project at this early stage.

See the Roadmap for more information.

Getting Started

Documentation is currently extremely sparse, not least because the APIs are not quite ready to document. One of the first items on the roadmap is to agree our new user "hello world" user stories, write them up and make sure the necessary APIs are in place to support them.

As we work on those, we will be expanding the Start here part of the Wiki.

Key features

Pure Java API

• No XML/JSON/YAML etc
• Clean and easy to integrate with existing applications

Evolutionary Database Design

• Implements bidirectional run-time verification of upgrades against a known target schema.
• Bootstraps empty databases with no additional coding
• Manages complex data migration and fixing alongside schema changes
• Targets multiple database platforms without platform-specific scripting and without sacrificing low level control
• Cleanly integrates disordered branches
• Exports SQL upgrade scripts in any SQL dialect for DBA use

Extensible

• Easy to add new database platform support

Cross-platform Snapshots and Cloning

• Clone or dump a full database, including schema and data
• Create a dump of a MySQL database as XML, restore to Oracle
• Copy a live NuoDB database directly into in-memory H2
• Run a JUnit test over an in-memory H2 database using a snapshot from a live Oracle environment
• Automatically run missing upgrades after restoring from a snapshot

Database Access API

• Database access at run-time using the same DSL used in upgrades

Contributing

If you're interested in contributing to the project, please read the contribution guide.