Features Overview

JaVers is designed as a framework for auditing changes in your object-oriented data.

With JaVers you can easily commit changes performed on your object graph to a specialized repository (called JaversRepository). Then you can browse the detailed change history of a given object in two forms — diffs and snapshots.

The data auditing framework is built on top of the object diff engine, which could be used as a standalone object diff tool for ad-hoc comparison of two object graphs.

All JaVers functions are exposed via a single Facade, the Javers instance. As you can see, JaVers API is concise and simple.

Object diff

JaVers object diff is the easiest way to deeply compare two object graphs.

How to use it?

• Create a JaVers instance (see getting started) and use javers.compare() to compare two object graphs.

• As the result, you get list of atomic Changes. There are several types of Changes: ValueChange, ReferenceChange, ListChange and so on (see the inheritance hierarchy of Change class to get the complete list).

• Take a look at diff examples.

JaVers Repository

JaversRepository is the central part of our data auditing engine.

It tracks every change made on your data (both values and relations) so you can easily identify when the change was made, who made it and what was the value before and after.

How to use it?

• Configure and build a JaVers instance (see configuration).

• Integrate JaVers with your system by applying the javers.commit() function in every place where important data (domain objects) are being created and modified by application users.

• You don’t need to commit every object. JaVers navigates through the object graph, starting from the object passed to javers.commit() and deeply compares the whole structure with the previous version stored in JaversRepository. Thanks to this approach, you can commit large structures, like trees, graphs and DDD aggregates with a single commit() call.

• If you are using Spring Data, annotate your Repositories with @JaversSpringDataAuditable and take advantage of the auto-audit aspect.

• Once your domain objects are being managed by JaVers, you can query JaversRepository using powerful JQL — JaVers Query Language.

• JaVers provides three views on objects history: Changes, Shadows and Snapshots. Use javers.find*() methods to browse detailed history of a given class, object or property.

• Take a look at repository examples.

JaversRepository is designed to be easily implemented for any kind of database. At the moment we provide MongoDB implementation and SQL implementation for the folowing dialects: H2, PostgreSQL, MySQL/MariaDB, Oracle and Microsoft SQL Server.
See repository configuratoin.

If you are using another database, for example Cassandra, you are encouraged to implement the JaversRepository interface and contribute it to JaVers project.

JSON serialization

JaVers has a well-designed and customizable JSON serialization and deserialization module, based on GSON and Java reflection. Your data are split into chunks (atomic changes) and persisted in a database as JSON with minimal mapping configuration effort (see custom JSON serialization).