Jumper is a Java library for upgrading database schemas and enables comparing database schemas across different database vendors.

Use Cases

• Schema Evolution: Apply the expected new database structure of your application
• Schema Validation: Ensure your production database schema matches your expected schema definition

Cross-Vendor Migration

Compare schemas when migrating from one database vendor to another.

As such, it is a direct competitor to Liquibase and Flyway. The main difference is that Jumper defines upgrades programmatically via a Java DSL, rather than using XML, JSON, YAML or direct SQL. While Jumper's approach is conceptually closer to Liquibase than to Flyway, it provides greater customization capabilities:

• De facto, running Java code is made easier during the upgrade process (using ORMs, generating dynamic SQL, collecting remote data, etc.)
• Schema comparison can be customized to compare more database objects or check specific properties

Overview

Through its core module, Jumper's main purpose is to upgrade database schemas by executing structures defined via a Java DSL, whose entry point is DDLEase

It also contains the difference module designed to collect and compare database schema structures. It helps developers identify differences in tables, columns, indexes, foreign keys, and other database objects, making it ideal for:

• Schema validation between development, staging, and production environments
• Database migration verification
• Multi-vendor database synchronization analysis
• Schema evolution tracking

Features

Schema upgrade 📈

• ⛓ Type-Safe Java DSL: You define your schema changes (tables, columns, foreign keys, etc.) using fluent Java code and benefit from code completion.
• 🔢 Versioned ChangeSets: Modifications are organized into ChangeSet objects, which are then grouped into a FluentChangeLog. This structure helps track and order evolution steps.
• 🎦 Tracked Applied Modifications: Executed ChangeSets are stored in a database table (by default) to avoid re-applying the same modifications.

Schema comparison 🔂

• 🔍 Comprehensive Schema Collection: Extracts complete database schemas including tables, columns, indexes, foreign keys, and constraints
• 🔁 Cross-Database Comparison: Compare schemas from the same or different database vendors
• 🔧 Customizable Comparison Engine: The comparison engine uses a fluent DSL that lets you create your own SchemaDiffer implementation

Multi-vendor support 🎯

The following database vendors are currently supported:

• Derby
• H2
• HSQLDB
• MariaDB
• MySQL
• PostgreSQL

Getting Started

Maven Dependency

<dependency>
    <groupId>org.codefilarete.jumper</groupId>
    <artifactId>difference-postgresql-adapter</artifactId>
    <version>${jumper.version}</version>
</dependency>

(Replace postgresql with your target database: mysql, mariadb, h2, hsqldb, or derby)

Basic Usage

Upgrading database structures

Several static methods are available from the DDLEase class to create a FluentChangeLog from a list of ChangeSets.

FluentChangeLog countryChangeLog = changeLog(
    changeSet("country_tables",
         createTable("Country")
             .addColumn("id", "BIGINT").primaryKey()
             .addColumn("name", "VARCHAR(255)")
             .addColumn("capitalId", "BIGINT")
             .addColumn("presidentId", "BIGINT"),
         createTable("City")
             .addColumn("id", "BIGINT").primaryKey()
             .addColumn("name", "VARCHAR(255)"),
         createTable("Person")
             .addColumn("id", "BIGINT").primaryKey()
             .addColumn("firstname", "VARCHAR(255)")
             .addColumn("lastname", "VARCHAR(255)"),
         createForeignKey("FK_Country_Capital", "Country", "City"),
         createForeignKey("FK_Country_President", "Country", "Person")
    ),
    changeSet("hibernate_tables",
         createTable("IdGenerator")
             .addColumn("sequence_name", "VARCHAR(255)")
             .addColumn("next_val", "BIGINT"))
);
countryChangeLog.applyTo(dataSource);

Comparing database structures

Let's compare two PostgreSQL databases by leveraging the PostgreSQLSchemaElementCollector and PostgreSQLSchemaDiffer classes:

// Connect to first database
PGSimpleDataSource dataSource1 = new PGSimpleDataSource();
dataSource1.setUrl("jdbc:postgresql://localhost:5432/database1");
dataSource1.setUser("admin");
dataSource1.setPassword("admin");

// Collect schema from first database
DefaultSchemaElementCollector collector1 = new PostgreSQLSchemaElementCollector(dataSource1.getConnection().getMetaData());
collector1
    .withCatalog("postgres")
    .withSchema("schema1")
    .withTableNamePattern("%");

// Connect to second database
PGSimpleDataSource dataSource2 = new PGSimpleDataSource();
dataSource2.setUrl("jdbc:postgresql://localhost:5432/database2");
dataSource2.setUser("admin");
dataSource2.setPassword("admin");

// Collect schema from second database
DefaultSchemaElementCollector collector2 = new PostgreSQLSchemaElementCollector(dataSource2.getConnection().getMetaData());
collector2
    .withCatalog("postgres")
    .withSchema("schema2")
    .withTableNamePattern("%");

// Compare and print differences to the console
PostgreSQLSchemaDiffer schemaDiffer = new PostgreSQLSchemaDiffer();
schemaDiffer.compareAndPrint(collector1.collect(), collector2.collect());

By default, Jumper analyzes and compares:

• Tables: Names, types, and presence
• Columns: Names, data types, nullability, default values, size/precision
• Primary Keys: Column composition and naming
• Foreign Keys: Relationships, referential actions, column mappings
• Indexes: Column composition, uniqueness, ordering (ASC/DESC)
• Constraints: Check constraints, unique constraints

Project Structure

├── core/ # Core schema collection functionality
├── difference/ # Schema comparison engine
└── difference-adapter/ # Database-specific adapters
    ├── difference-derby-adapter/
    ├── difference-h2-adapter/
    ├── difference-hsqldb-adapter/
    ├── difference-mariadb-adapter/
    ├── difference-mysql-adapter/
    └── difference-postgresql-adapter/

Requirements

• Java 8 or higher
• JDBC drivers for your target database(s)