Database Support

Kapper is designed to work with any JDBC-compatible database, it simply extends the JDBC APIs. We test Kapper against some standard DBs, but it has been proven to work against many more.

Supported Databases

Kapper officially supports and is tested against:

• PostgreSQL
• MySQL
• SQLite
• MS SQL Server
• Oracle
• DuckDB

Connection Setup

Kapper simply uses a standard DataSource, which can be provided by any supported library. Here as some examples using the Hikari Connection Pool:

PostgreSQL

val dataSource = HikariDataSource().apply {
    jdbcUrl = "jdbc:postgresql://localhost:5432/mydb"
    username = "user"
    password = "password"
}

MySQL

val dataSource = HikariDataSource().apply {
    jdbcUrl = "jdbc:mysql://localhost:3306/mydb?allowMultiQueries=true"
    username = "user" 
    password = "password"
}

SQLite

val dataSource = HikariDataSource().apply {
    jdbcUrl = "jdbc:sqlite:mydb.db"
}

Type Mapping

Kapper automatically handles type conversion between many SQL and JVM/Kotlin types. When a type is not supprted, it can be supported through custom mappers

Array Support

Array types are supported on a per-database basis:

PostgreSQL and DuckDB:

• Full support for List<T> parameters and result mapping
• Supported element types: Int, Long, Short, Float, Double, Boolean, String, BigDecimal
• PostgreSQL also supports UUID element type
• Arrays can be null, contain null elements, or be empty (on result mapping)
• Example: WHERE id = ANY(:ids) with "ids" to listOf(1, 2, 3)

MySQL, SQLite, Oracle, MSSQL:

• Array types are not supported
• Passing a List/Array parameter will throw KapperUnsupportedOperationException

See Array Types in the mapping guide for details and examples.

Testing

For testing, consider using:

• SQLite for in-memory or file-based tests
• Testcontainers for production-like integration tests

Next Steps

• Installation - Getting started
• Basic Usage - Your first queries
• Performance - Optimization tips