Gradle

The Gradle plugin configures one task for each named execution.

Replace VERSION with the published version shown on the Gradle plugin release page.

Kotlin DSL

plugins {
    id("dev.banking.asyncapi.generator") version "VERSION"
}

asyncApiGenerator {
    executions {
        register("models") {
            generatorName.set("kotlin")
            inputSpec.set(file("src/main/resources/asyncapi.yaml"))
            modelPackage.set("com.example.model")
        }
    }
}

Groovy DSL

plugins {
    id 'dev.banking.asyncapi.generator' version 'VERSION'
}

asyncApiGenerator {
    executions {
        register('models') {
            generatorName.set('kotlin')
            inputSpec.set(file('src/main/resources/asyncapi.yaml'))
            modelPackage.set('com.example.model')
        }
    }
}

Run every configured execution with the aggregate task:

./gradlew generateAsyncApi

Each execution also has its own task. The models execution above creates generateModelsAsyncApi:

./gradlew generateModelsAsyncApi

By default, that execution writes generated files below build/generated/asyncapi/models.

Configure models

modelConfig belongs to an individual execution:

asyncApiGenerator {
    executions {
        register("models") {
            generatorName.set("java")
            inputSpec.set(file("src/main/resources/asyncapi.yaml"))
            modelPackage.set("com.example.model")
            modelConfig {
                modelType.set("java-record")
            }
        }
    }
}

Generate Spring Kafka clients

clientConfig is shared by client-generating executions. Each execution still defines its own model and client packages:

asyncApiGenerator {
    clientConfig {
        clientType.set("spring-kafka")
        clientContract.set("interface")
        producer {
            enabled.set(true)
            additionalPayloadTypes.set(listOf("byte-array"))
        }
        consumer {
            enabled.set(true)
        }
    }

    executions {
        register("client") {
            generatorName.set("kotlin")
            inputSpec.set(file("src/main/resources/asyncapi.yaml"))
            modelPackage.set("com.example.model")
            clientPackage.set("com.example.client")
        }
    }
}

Generate schemas

Use a schema generator profile and set schemaPackage on the execution. This example generates Avro schema files without runtime models:

asyncApiGenerator {
    executions {
        register("schemas") {
            generatorName.set("avro-schema")
            inputSpec.set(file("src/main/resources/asyncapi.yaml"))
            schemaPackage.set("com.example.schema.avro")
        }
    }
}

The schema-only generator profiles are avro-schema, protobuf-schema, and json-schema.

Execution fields

Field	Description	Default
generatorName	Generator profile (kotlin, java, avro-schema, protobuf-schema, json-schema, asyncapi-yaml, asyncapi-json)	required
inputSpec	AsyncAPI YAML or JSON file	required
outputDirectory	Directory for generated sources and schemas	build/generated/asyncapi/<execution-name>
outputFile	File for bundled document output	none
modelPackage	Package for generated payload models	none
clientPackage	Package for generated client contracts	none
schemaPackage	Package for generated schema artifacts	none
modelConfig	Model implementation and annotation settings for this execution	none

clientConfig is configured once on asyncApiGenerator and is shared by all named executions.