Service descriptors

§Service Descriptors

Lagom services are described by an interface, known as a service descriptor. This interface not only defines how the service is invoked and implemented, it also defines the metadata that describes how the interface is mapped down onto an underlying transport protocol. Generally, the service descriptor, its implementation and consumption should remain agnostic to what transport is being used, whether that’s REST, websockets, or some other transport. Let’s take a look at a simple descriptor:

import com.lightbend.lagom.javadsl.api.*;

import static com.lightbend.lagom.javadsl.api.Service.*;

public interface HelloService extends Service {
  ServiceCall<String, String> sayHello();

  default Descriptor descriptor() {
    return named("hello").withCalls(call(this::sayHello));

This descriptor defines a service with one call, the sayHello call. sayHello() is a method that returns something of type ServiceCall, this is a representation of the call that can be invoked when consuming the service, and implemented by the service itself. This is what the interface looks like:

interface ServiceCall<Request, Response> {
  CompletionStage<Response> invoke(Request request);

An important thing to note here is that invoking the sayHello() method does not actually invoke the call, it simply gets a handle to the call, which can then be invoked using the invoke method.

ServiceCall takes two type parameters, Request and Response. The Request parameter is the type of the incoming request message, and the Response parameter is the type of the outgoing response message. In the example above, these are both String, so our service call just handles simple text messages.

While the sayHello() method describes how the call will be programmatically invoked or implemented, it does not describe how this call gets mapped down onto the transport. This is done by providing a default implementation of the descriptor() call, whose interface is described by Service.

You can see that we’re returning a service named hello, and we’re describing one call, the sayHello call. Because this service is so simple, in this case we don’t need to do anything more than simply passing the Service call sayHello defined above in the example as a method reference to the call method.

§Call identifiers

Each service call needs to have an identifier. An identifier is used to provide routing information to the implementation of the client and the service, so that calls over the wire can be mapped to the appropriate call. Identifiers can be a static name or path, or they can have dynamic components, where dynamic path parameters are extracted from the path and passed to the service call methods.

The simplest type of identifier is a name, and by default, that name is set to be the same name as the name of the method on the interface that implements it. A custom name can also be supplied, by passing it to the namedCall method:

default Descriptor descriptor() {
  return named("hello").withCalls(namedCall("hello", this::sayHello));

In this case, we’ve named it hello, instead of the default of sayHello. When implemented using REST, this will mean this call will have a path of /hello.

§Path based identifiers

The second type of identifier is a path based identifier. This uses a URI path and query string to route calls, and from it dynamic path parameters can optionally be extracted out. They can be configured using the pathCall method.

Dynamic path parameters are extracted from the path by declaring dynamic parts in the path. These are prefixed with a colon, for example, a path of /order/:id has a dynamic part called id. Lagom will extract this parameter from the path, and pass it to the service call method. In order to convert it to the type accepted by the method, Lagom will use a PathParamSerializer. Lagom includes many PathParamSerializer’s out of the box, such as for String, Long, Integer and Boolean. Here’s an example of extracting a long parameter from the path and passing it to a service call:

ServiceCall<NotUsed, Order> getOrder(long id);

default Descriptor descriptor() {
  return named("orders").withCalls(pathCall("/order/:id", this::getOrder));

Multiple parameters can of course be extracted out, these will be passed to your service call method in the order they are extracted from the URL:

ServiceCall<NotUsed, Item> getItem(long orderId, String itemId);

default Descriptor descriptor() {
  return named("orders").withCalls(pathCall("/order/:orderId/item/:itemId", this::getItem));

Query string parameters can also be extracted from the path, using a & separated list after a ? at the end of the path. For example, the following service call uses query string parameters to implement paging:

ServiceCall<NotUsed, PSequence<Item>> getItems(long orderId, int pageNo, int pageSize);

default Descriptor descriptor() {
  return named("orders")
      .withCalls(pathCall("/order/:orderId/items?pageNo&pageSize", this::getItems));

When you use call, namedCall or pathCall, if Lagom maps that down to REST, Lagom will make a best effort attempt to map it down to REST in a semantic fashion, so that means if there is a request message, it will use POST, if there’s none, it will use GET.

§REST identifiers

The final type of identifier is a REST identifier. REST identifiers are designed to be used when creating semantic REST APIs. They use both a path, as with the path based identifier, and a request method, to identify them. They can be configured using the restCall method:

ServiceCall<Item, NotUsed> addItem(long orderId);

ServiceCall<NotUsed, Item> getItem(long orderId, String itemId);

ServiceCall<NotUsed, NotUsed> deleteItem(long orderId, String itemId);

default Descriptor descriptor() {
  return named("orders")
          restCall(Method.POST, "/order/:orderId/item", this::addItem),
          restCall(Method.GET, "/order/:orderId/item/:itemId", this::getItem),
          restCall(Method.DELETE, "/order/:orderId/item/:itemId", this::deleteItem));


Every service call in Lagom has a request message type and a response message type. When the request or response message isn’t used, the akka.NotUsed can be used in their place. Request and response message types fall into two categories, strict and streamed.

§Strict messages

A strict message is a single message that can be represented by a simple Java object. The message will be buffered into memory, and then parsed, for example, as JSON. When both message types are strict, the call is said to be a synchronous call, that is, a request is sent and received, then a response is sent and received. The caller and callee have synchronized in their communication.

So far, all of the service call examples we’ve seen have used strict messages, for example, the order service descriptors above accept and return items and orders. The input value is passed directly to the service call, and returned directly from the service call, and these values are serialized to a JSON buffer in memory before being sent, and read entirely into memory before being deserialized back from JSON.

§Streamed messages

A streamed message is a message of type Source. Source is an Akka streams API that allows asynchronous streaming and handling of messages. Here’s an example streamed service call:

ServiceCall<String, Source<String, ?>> tick(int interval);

default Descriptor descriptor() {
  return named("clock").withCalls(pathCall("/tick/:interval", this::tick));

This service call has a strict request type and a streamed response type. An implementation of this might return a Source that sends the input tick message String at the specified interval.

A bidirectional streamed call might look like this:

ServiceCall<Source<String, ?>, Source<String, ?>> sayHello();

default Descriptor descriptor() {
  return named("hello").withCalls(call(this::sayHello));

In this case, the server might return a Source that converts every message received in the request stream to messages prefixed with Hello.

Lagom will choose an appropriate transport for the stream, typically, this will be WebSockets. The WebSocket protocol supports bidirectional streaming, so is a good general purpose option for streaming. When only one of the request or response message is streamed, Lagom will implement the sending and receiving of the strict message by sending or receiving a single message, and then leaving the WebSocket open until the other direction closes. Otherwise, Lagom will close the WebSocket when either direction closes.

§Message serialization

By default, Lagom will choose an appropriate serializer for request and response serialization and deserialization. Out of the box, Lagom will use JSON for communication, using Jackson to serialize and deserialize messages.

For details on message serializers, including how to write and configure custom message serializers, see Message Serializers.

Found an error in this documentation? The source code for this page can be found here. Please feel free to edit and contribute a pull request.