Exposed for Kotlin: Complete Guide with Examples

What is Exposed and why is it needed?

Exposed is a lightweight but powerful ORM library (Object-Relational Mapping) for the Kotlin programming language, developed by JetBrains. It provides two main approaches to working with databases: DSL (Domain Specific Language) and DAO (Data Access Objects).

Why is Exposed needed?

• Type safety: Unlike raw SQL queries, Exposed uses typed Kotlin constructs, allowing errors to be caught at compile time.
• Conciseness: Code in Exposed is significantly shorter and more readable than equivalent SQL queries or Java code with JPA.
• Performance: Exposed does not use reflection or proxy objects (like Hibernate), making it very fast and predictable.
• Flexibility: You can easily switch between DSL (for complex queries) and DAO (for simple CRUD operations) within the same project.
• Support for multiple databases: PostgreSQL, MySQL, MariaDB, SQLite, H2, Oracle, SQL Server.

Installing Exposed

Exposed is installed via the Gradle or Maven build system. Add the following dependencies to your build.gradle.kts:

dependencies {    // Core library    implementation("org.jetbrains.exposed:exposed-core:0.44.1")    // DSL for queries    implementation("org.jetbrains.exposed:exposed-dao:0.44.1")    // JDBC driver for your database (example for PostgreSQL)    implementation("org.postgresql:postgresql:42.7.1")    // HikariCP connection pool (recommended)    implementation("com.zaxxer:HikariCP:5.1.0")    // SQL logging (optional)    implementation("ch.qos.logback:logback-classic:1.4.14")}

For Maven (pom.xml):

<dependency>    <groupId>org.jetbrains.exposed</groupId>    <artifactId>exposed-core</artifactId>    <version>0.44.1</version></dependency>

Quick start: code example

Let's create a simple task management application (ToDo):

import org.jetbrains.exposed.sql.*import org.jetbrains.exposed.sql.transactions.transactionimport org.jetbrains.exposed.sql.SqlExpressionBuilder.eqimport java.time.LocalDateTime

// 1. Define the tableobject Tasks : Table() {    val id = integer("id").autoIncrement()    val title = varchar("title", 255)    val description = text("description").nullable()    val isCompleted = bool("is_completed").default(false)    val createdAt = datetime("created_at").default(LocalDateTime.now())

    override val primaryKey = PrimaryKey(id)}

// 2. Data class for conveniencedata class Task(    val id: Int,    val title: String,    val description: String?,    val isCompleted: Boolean,    val createdAt: LocalDateTime)

fun main() {    // 3. Connect to the database (H2 in-memory for example)    Database.connect("jdbc:h2:mem:test;DB_CLOSE_DELAY=-1", driver = "org.h2.Driver")

    transaction {        // 4. Create the table        SchemaUtils.create(Tasks)

        // 5. Insert data        val insertId = Tasks.insert {            it[title] = "Learn Exposed"            it[description] = "Read the documentation and write an example"        } get Tasks.id

        println("Inserted task with ID: $insertId")

        // 6. Select all tasks        val allTasks = Tasks.selectAll().map { row ->            Task(                id = row[Tasks.id],                title = row[Tasks.title],                description = row[Tasks.description],                isCompleted = row[Tasks.isCompleted],                createdAt = row[Tasks.createdAt]            )        }        println("All tasks: $allTasks")

        // 7. Update a task        Tasks.updat