Programming model

Function Catalog and Flexible Function Signatures

One of the main features of Spring Cloud Function is to adapt and support a range of type signatures for user-defined functions, while providing a consistent execution model. That’s why all user defined functions are transformed into a canonical representation by FunctionCatalog.

While users don’t normally have to care about the FunctionCatalog at all, it is useful to know what kind of functions are supported in user code.

It is also important to understand that Spring Cloud Function provides first class support for reactive API provided by Project Reactor allowing reactive primitives such as Mono and Flux to be used as types in user defined functions providing greater flexibility when choosing programming model for your function implementation. Reactive programming model also enables functional support for features that would be otherwise difficult to impossible to implement using imperative programming style. For more on this please read Function Arity section.

Java 8 function support

Spring Cloud Function embraces and builds on top of the 3 core functional interfaces defined by Java and available to us since Java 8.

  • Supplier<O>

  • Function<I, O>

  • Consumer<I>

To avoid constantly mentioning Supplier, Function and Consumer we’ll refer to them a Functional beans for the rest of this manual where appropriate.

In a nutshell, any bean in your Application Context that is Functional bean will lazily be registered with FunctionCatalog. This means that it could benefit from all of the additional features described in this reference manual.

In a simplest of application all you need to do is to declare @Bean of type Supplier, Function or Consumer in your application configuration. Then you can access FunctionCatalog and lookup a particular function based on its name.

For example:

@Bean
public Function<String, String> uppercase() {
    return value -> value.toUpperCase();
}

. . .

FunctionCatalog catalog = applicationContext.getBean(FunctionCatalog.class);
Function uppercase =  catalog.lookup(“uppercase”);

Important to understand that given that uppercase is a bean, you can certainly get it form the ApplicationContext directly, but all you will get is just your bean as you declared it without any extra features provided by SCF. When you do lookup of a function via FunctionCatalog, the instance you will receive is wrapped (instrumented) with additional features (i.e., type conversion, composition etc.) described in this manual. Also, it is important to understand that a typical user does not use Spring Cloud Function directly. Instead a typical user implements Java Function/Supplier/Consumer with the idea of using it in different execution contexts without additional work. For example the same java function could be represented as REST endpoint or Streaming message handler or AWS Lambda and more via Spring Cloud Function provided adapters as well as other frameworks using Spring Cloud Function as the core programming model (e.g., Spring Cloud Stream) So in summary Spring Cloud Function instruments java functions with additional features to be utilised in variety of execution contexts.

Function definition

While the previous example shows you how to lookup function in FunctionCatalog programmatically, in a typical integration case where Spring Cloud Function used as programming model by another framework (e.fg. Spring Cloud Stream), you declare which functions to use via spring.cloud.function.definition property. Knowing that it is important to understand some default behaviour when it comes to discovering functions in FunctionCatalog. For example, if you only have one Functional bean in your ApplicationContext, the spring.cloud.function.definition property typically will not be required, since a single function in FunctionCatalog can be looked up by an empty name or any name. For example, assuming that uppercase is the only function in your catalog, it can be looked up as catalog.lookup(null), catalog.lookup(“”), catalog.lookup(“foo”) That said, for cases where you are using framework such as Spring Cloud Stream which uses spring.cloud.function.definition it is best practice and recommended to always use spring.cloud.function.definition property.

For example,

spring.cloud.function.definition=uppercase

Filtering ineligible functions

A typical Application Context may include beans that are valid java functions, but not intended to be candidates to be registered with FunctionCatalog. Such beans could be auto-configurations from other projects or any other beans that qualify to be Java functions. The framework provides default filtering of known beans that should not be candidates for registration with function catalog. You can also add to this list additional beans by providing coma delimited list of bean definition names using spring.cloud.function.ineligible-definitions property.

For example,

spring.cloud.function.ineligible-definitions=foo,bar

Supplier

Supplier can be reactive - Supplier<Flux<T>> or imperative - Supplier<T>. From the invocation standpoint this should make no difference to the implementor of such Supplier. However, when used within frameworks (e.g., Spring Cloud Stream), Suppliers, especially reactive, often used to represent the source of the stream, therefore they are invoked once to get the stream (e.g., Flux) to which consumers can subscribe to. In other words such suppliers represent an equivalent of an infinite stream. However, the same reactive suppliers can also represent finite stream(s) (e.g., result set on the polled JDBC data). In those cases such reactive suppliers must be hooked up to some polling mechanism of the underlying framework.

To assist with that Spring Cloud Function provides a marker annotation org.springframework.cloud.function.context.PollableBean to signal that such supplier produces a finite stream and may need to be polled again. That said, it is important to understand that Spring Cloud Function itself provides no behavior for this annotation.

In addition PollableBean annotation exposes a splittable attribute to signal that produced stream needs to be split (see Splitter EIP)

Here is the example:

@PollableBean(splittable = true)
public Supplier<Flux<String>> someSupplier() {
	return () -> {
		String v1 = String.valueOf(System.nanoTime());
		String v2 = String.valueOf(System.nanoTime());
		String v3 = String.valueOf(System.nanoTime());
		return Flux.just(v1, v2, v3);
	};
}

Function

Function can also be written in imperative or reactive way, yet unlike Supplier and Consumer there are no special considerations for the implementor other then understanding that when used within frameworks such as Spring Cloud Stream and others, reactive function is invoked only once to pass a reference to the stream (Flux or Mono) and imperative is invoked once per event.

Consumer

Consumer is a little bit special because it has a void return type, which implies blocking, at least potentially. Most likely you will not need to write Consumer<Flux<?>>, but if you do need to do that, remember to subscribe to the input flux.

Function Composition

Function Composition is a feature that allows one to compose several functions into one. The core support is based on function composition feature available with Function.andThen(..) support available since Java 8. However on top of it, we provide few additional features.

Declarative Function Composition

This feature allows you to provide composition instruction in a declarative way using | (pipe) or , (comma) delimiter when providing spring.cloud.function.definition property.

For example

--spring.cloud.function.definition=uppercase|reverse

Here we effectively provided a definition of a single function which itself is a composition of function uppercase and function reverse. In fact that is one of the reasons why the property name is definition and not name, since the definition of a function can be a composition of several named functions. And as mentioned you can use , instead of pipe (such as …​definition=uppercase,reverse).

Composing non-Functions

Spring Cloud Function also supports composing Supplier with Consumer or Function as well as Function with Consumer. What’s important here is to understand the end product of such definitions. Composing Supplier with Function still results in Supplier while composing Supplier with Consumer will effectively render Runnable. Following the same logic composing Function with Consumer will result in Consumer.

And of course you can’t compose uncomposable such as Consumer and Function, Consumer and Supplier etc.

Function Routing and Filtering

Since version 2.2 Spring Cloud Function provides routing feature allowing you to invoke a single function which acts as a router to an actual function you wish to invoke This feature is very useful in certain FAAS environments where maintaining configurations for several functions could be cumbersome or exposing more than one function is not possible.

The RoutingFunction is registered in FunctionCatalog under the name functionRouter. For simplicity and consistency you can also refer to RoutingFunction.FUNCTION_NAME constant.

This function has the following signature:

public class RoutingFunction implements Function<Object, Object> {
. . .
}

The routing instructions could be communicated in several ways. We support providing instructions via Message headers, System properties as well as pluggable strategy. So let’s look at some of the details

MessageRoutingCallback

The MessageRoutingCallback is a strategy to assist with determining the name of the route-to function definition.

public interface MessageRoutingCallback {
	FunctionRoutingResult routingResult(Message<?> message);
	. . .
}

All you need to do is implement and register it as a bean to be picked up by the RoutingFunction. For example:

@Bean
public MessageRoutingCallback customRouter() {
	return new MessageRoutingCallback() {
		@Override
		public FunctionRoutingResult routingResult(Message<?> message) {
			return new FunctionRoutingResult((String) message.getHeaders().get("func_name"));
		}
	};
}

In the preceding example you can see a very simple implementation of MessageRoutingCallback which determines the function definition from func_name Message header of the incoming Message and returns the instance of FunctionRoutingResult containing the definition of function to invoke.

Message Headers

If the input argument is of type Message<?>, you can communicate routing instruction by setting one of spring.cloud.function.definition or spring.cloud.function.routing-expression Message headers. As the name of the property suggests spring.cloud.function.routing-expression relies on Spring Expression Language (SpEL). For more static cases you can use spring.cloud.function.definition header which allows you to provide the name of a single function (e.g., …​definition=foo) or a composition instruction (e.g., …​definition=foo|bar|baz). For more dynamic cases you can use spring.cloud.function.routing-expression header and provide SpEL expression that should resolve into definition of a function (as described above).

SpEL evaluation context’s root object is the actual input argument, so in the case of Message<?> you can construct expression that has access to both payload and headers (e.g., spring.cloud.function.routing-expression=headers.function_name).
SpEL allows user to provide string representation of Java code to be executed. Given that the spring.cloud.function.routing-expression could be provided via Message headers means that ability to set such expression could be exposed to the end user (i.e., HTTP Headers when using web module) which could result in some problems (e.g., malicious code). To manage that, all expressions coming via Message headers will only be evaluated against SimpleEvaluationContext which has limited functionality and designed to only evaluate the context object (Message in our case). On the other hand, all expressions that are set via property or system variable are evaluated against StandardEvaluationContext, which allows for full flexibility of Java language. While setting expression via system/application property or environment variable is generally considered to be secure as it is not exposed to the end user in normal cases, there are cases where visibility as well as capability to update system, application and environment variables are indeed exposed to the end user via Spring Boot Actuator endpoints provided either by some of the Spring projects or third parties or custom implementation by the end user. Such endpoints must be secured using industry standard web security practices. Spring Cloud Function does not expose any of such endpoints.

In specific execution environments/models the adapters are responsible to translate and communicate spring.cloud.function.definition and/or spring.cloud.function.routing-expression via Message header. For example, when using spring-cloud-function-web you can provide spring.cloud.function.definition as an HTTP header and the framework will propagate it as well as other HTTP headers as Message headers.

Application Properties

Routing instruction can also be communicated via spring.cloud.function.definition or spring.cloud.function.routing-expression as application properties. The rules described in the previous section apply here as well. The only difference is you provide these instructions as application properties (e.g., --spring.cloud.function.definition=foo).

It is important to understand that providing spring.cloud.function.definition or spring.cloud.function.routing-expression as Message headers will only work for imperative functions (e.g., Function<Foo, Bar>). That is to say that we can only route per-message with imperative functions. With reactive functions we can not route per-message. Therefore you can only provide your routing instructions as Application Properties. It’s all about unit-of-work. In imperative function unit of work is Message so we can route based on such unit-of-work. With reactive function unit-of-work is the entire stream, so we’ll act only on the instruction provided via application properties and route the entire stream.

Order of priority for routing instructions

Given that we have several mechanisms of providing routing instructions it is important to understand the priorities for conflict resolutions in the event multiple mechanisms are used at the same time, so here is the order:

  1. MessageRoutingCallback (If function is imperative will take over regardless if anything else is defined)

  2. Message Headers (If function is imperative and no MessageRoutingCallback provided)

  3. Application Properties (Any function)

Unroutable Messages

In the event route-to function is not available in catalog you will get an exception stating that.

There are cases when such behavior is not desired and you may want to have some "catch-all" type function which can handle such messages. To accomplish that, framework provides org.springframework.cloud.function.context.DefaultMessageRoutingHandler strategy. All you need to do is register it as a bean. Its default implementation will simply log the fact that the message is un-routable, but will allow message flow to proceed without the exception, effectively dropping the un-routable message. If you want something more sophisticated all you need to do is provide your own implementation of this strategy and register it as a bean.

@Bean
public DefaultMessageRoutingHandler defaultRoutingHandler() {
	return new DefaultMessageRoutingHandler() {
		@Override
		public void accept(Message<?> message) {
			// do something really cool
		}
	};
}

Function Filtering

Filtering is the type of routing where there are only two paths - 'go' or 'discard'. In terms of functions it mean you only want to invoke a certain function if some condition returns 'true', otherwise you want to discard input. However, when it comes to discarding input there are many interpretation of what it could mean in the context of your application. For example, you may want to log it, or you may want to maintain the counter of discarded messages. you may also want to do nothing at all. Because of these different paths, we do not provide a general configuration option for how to deal with discarded messages. Instead we simply recommend to define a simple Consumer which would signify the 'discard' path:

@Bean
public Consumer<?> devNull() {
   // log, count or whatever
}

Now you can have routing expression that really only has two paths effectively becoming a filter. For example:

--spring.cloud.function.routing-expression=headers.contentType.toString().equals('text/plain') ? 'echo' : 'devNull'

Every message that does not fit criteria to go to 'echo' function will go to 'devNull' where you can simply do nothing with it. The signature Consumer<?> will also ensure that no type conversion will be attempted resulting in almost no execution overhead.

When dealing with reactive inputs (e.g., Publisher), routing instructions must only be provided via Function properties. This is due to the nature of the reactive functions which are invoked only once to pass a Publisher and the rest is handled by the reactor, hence we can not access and/or rely on the routing instructions communicated via individual values (e.g., Message).

Multiple Routers

By default the framework will always have a single routing function configured as described in previous sections. However, there are times when you may need more than one routing function. In that case you can create your own instance of the RoutingFunction bean in addition to the existing one as long as you give it a name other than functionRouter.

You can pass spring.cloud.function.routing-expression or spring.cloud.function.definition to RoutinFunction as key/value pairs in the map.

Here is a simple example

@Configuration
protected static class MultipleRouterConfiguration {

	@Bean
	RoutingFunction mySpecialRouter(FunctionCatalog functionCatalog, BeanFactory beanFactory, @Nullable MessageRoutingCallback routingCallback) {
		Map<String, String> propertiesMap = new HashMap<>();
		propertiesMap.put(FunctionProperties.PREFIX + ".routing-expression", "'reverse'");
		return new RoutingFunction(functionCatalog, propertiesMap, new BeanFactoryResolver(beanFactory), routingCallback);
	}

	@Bean
	public Function<String, String> reverse() {
		return v -> new StringBuilder(v).reverse().toString();
	}

	@Bean
	public Function<String, String> uppercase() {
		return String::toUpperCase;
	}
}

and a test that demonstrates how it works

`

@Test
public void testMultipleRouters() {
	System.setProperty(FunctionProperties.PREFIX + ".routing-expression", "'uppercase'");
	FunctionCatalog functionCatalog = this.configureCatalog(MultipleRouterConfiguration.class);
	Function function = functionCatalog.lookup(RoutingFunction.FUNCTION_NAME);
	assertThat(function).isNotNull();
	Message<String> message = MessageBuilder.withPayload("hello").build();
	assertThat(function.apply(message)).isEqualTo("HELLO");

	function = functionCatalog.lookup("mySpecialRouter");
	assertThat(function).isNotNull();
	message = MessageBuilder.withPayload("hello").build();
	assertThat(function.apply(message)).isEqualTo("olleh");
}

[[input/output-enrichment]] == Input/Output Enrichment

There are often times when you need to modify or refine an incoming or outgoing Message and to keep your code clean of non-functional concerns. You don’t want to do it inside of your business logic.

You can always accomplish it via Function Composition. Such approach provides several benefits:

  • It allows you to isolate this non-functional concern into a separate function which you can compose with the business function as function definition.

  • It provides you with complete freedom (and danger) as to what you can modify before incoming message reaches the actual business function.

@Bean
public Function<Message<?>, Message<?>> enrich() {
    return message -> MessageBuilder.fromMessage(message).setHeader("foo", "bar").build();
}

@Bean
public Function<Message<?>, Message<?>> myBusinessFunction() {
    // do whatever
}

And then compose your function by providing the following function definition enrich|myBusinessFunction.

While the described approach is the most flexible, it is also the most involved as it requires you to write some code, make it a bean or manually register it as a function before you can compose it with the business function as you can see from the preceding example.

But what if modifications (enrichments) you are trying to make are trivial as they are in the preceding example? Is there a simpler and more dynamic and configurable mechanism to accomplish the same?

Since version 3.1.3, the framework allows you to provide SpEL expression to enrich individual message headers for both input going into function and and output coming out of it. Let’s look at one of the tests as the example.

@Test
public void testMixedInputOutputHeaderMapping() throws Exception {
	try (ConfigurableApplicationContext context = new SpringApplicationBuilder(
			SampleFunctionConfiguration.class).web(WebApplicationType.NONE).run(
					"--logging.level.org.springframework.cloud.function=DEBUG",
					"--spring.main.lazy-initialization=true",
					"--spring.cloud.function.configuration.split.output-header-mapping-expression.keyOut1='hello1'",
					"--spring.cloud.function.configuration.split.output-header-mapping-expression.keyOut2=headers.contentType",
					"--spring.cloud.function.configuration.split.input-header-mapping-expression.key1=headers.path.split('/')[0]",
					"--spring.cloud.function.configuration.split.input-header-mapping-expression.key2=headers.path.split('/')[1]",
					"--spring.cloud.function.configuration.split.input-header-mapping-expression.key3=headers.path")) {

		FunctionCatalog functionCatalog = context.getBean(FunctionCatalog.class);
		FunctionInvocationWrapper function = functionCatalog.lookup("split");
		Message<byte[]> result = (Message<byte[]>) function.apply(MessageBuilder.withPayload("helo")
				.setHeader(MessageHeaders.CONTENT_TYPE, "application/json")
				.setHeader("path", "foo/bar/baz").build());
		assertThat(result.getHeaders().containsKey("keyOut1")).isTrue();
		assertThat(result.getHeaders().get("keyOut1")).isEqualTo("hello1");
		assertThat(result.getHeaders().containsKey("keyOut2")).isTrue();
		assertThat(result.getHeaders().get("keyOut2")).isEqualTo("application/json");
	}
}

Here you see a properties called input-header-mapping-expression and output-header-mapping-expression preceded by the name of the function (i.e., split) and followed by the name of the message header key you want to set and the value as SpEL expression. The first expression (for 'keyOut1') is literal SpEL expressions enclosed in single quotes, effectively setting 'keyOut1' to value hello1. The keyOut2 is set to the value of existing 'contentType' header.

You can also observe some interesting features in the input header mapping where we actually splitting a value of the existing header 'path', setting individual values of key1 and key2 to the values of split elements based on the index.

if for whatever reason the provided expression evaluation fails, the execution of the function will proceed as if nothing ever happen. However you will see the WARN message in your logs informing you about it 
o.s.c.f.context.catalog.InputEnricher    : Failed while evaluating expression "hello1"  on incoming message. . .

In the event you are dealing with functions that have multiple inputs (next section), you can use index immediately after input-header-mapping-expression

--spring.cloud.function.configuration.echo.input-header-mapping-expression[0].key1=‘hello1'
--spring.cloud.function.configuration.echo.input-header-mapping-expression[1].key2='hello2'

Function Arity

There are times when a stream of data needs to be categorized and organized. For example, consider a classic big-data use case of dealing with unorganized data containing, let’s say, ‘orders’ and ‘invoices’, and you want each to go into a separate data store. This is where function arity (functions with multiple inputs and outputs) support comes to play.

Let’s look at an example of such a function (full implementation details are available here),

@Bean
public Function<Flux<Integer>, Tuple2<Flux<String>, Flux<String>>> organise() {
	return flux -> ...;
}

Given that Project Reactor is a core dependency of SCF, we are using its Tuple library. Tuples give us a unique advantage by communicating to us both cardinality and type information. Both are extremely important in the context of SCSt. Cardinality lets us know how many input and output bindings need to be created and bound to the corresponding inputs and outputs of a function. Awareness of the type information ensures proper type conversion.

Also, this is where the ‘index’ part of the naming convention for binding names comes into play, since, in this function, the two output binding names are organise-out-0 and organise-out-1.

IMPORTANT: At the moment, function arity is only supported for reactive functions (Function<TupleN<Flux<?>…​>, TupleN<Flux<?>…​>>) centered on Complex event processing where evaluation and computation on confluence of events typically requires view into a stream of events rather than single event.

Input Header propagation

In a typical scenario input Message headers are not propagated to output and rightfully so, since the output of a function may be an input to something else requiring it’s own set of Message headers. However, there are times when such propagation may be necessary so Spring Cloud Function provides several mechanisms to accomplish this.

First you can always copy headers manually. For example, if you have a Function with the signature that takes Message and returns Message (i.e., Function<Message, Message>), you can simply and selectively copy headers yourselves. Remember, if your function returns Message, the framework will not do anything to it other then properly converting its payload. However, such approach may prove to be a bit tedious, especially in cases when you simply want to copy all headers. To assist with cases like this we provide a simple property that would allow you to set a boolean flag on a function where you want input headers to be propagated. The property is copy-input-headers.

For example, let’s assume you have the following configuration:

@EnableAutoConfiguration
@Configuration
protected static class InputHeaderPropagationConfiguration {

	@Bean
	public Function<String, String> uppercase() {
		return x -> x.toUpperCase();
	}
}

As you know you can still invoke this function by sending a Message to it (framework will take care of type conversion and payload extraction)

By simply setting spring.cloud.function.configuration.uppercase.copy-input-headers to true, the following assertion will be true as well

Function<Message<String>, Message<byte[]>> uppercase = catalog.lookup("uppercase", "application/json");
Message<byte[]> result = uppercase.apply(MessageBuilder.withPayload("bob").setHeader("foo", "bar").build());
assertThat(result.getHeaders()).containsKey("foo");

Type conversion (Content-Type negotiation)

Content-Type negotiation is one of the core features of Spring Cloud Function as it allows to not only transform the incoming data to the types declared by the function signature, but to do the same transformation during function composition making otherwise un-composable (by type) functions composable.

To better understand the mechanics and the necessity behind content-type negotiation, we take a look at a very simple use case by using the following function as an example:

@Bean
public Function<Person, String> personFunction {..}

The function shown in the preceding example expects a Person object as an argument and produces a String type as an output. If such function is invoked with the type Person, than all works fine. But typically function plays a role of a handler for the incoming data which most often comes in the raw format such as byte[], JSON String etc. In order for the framework to succeed in passing the incoming data as an argument to this function, it has to somehow transform the incoming data to a Person type.

Spring Cloud Function relies on two native to Spring mechanisms to accomplish that.

  1. MessageConverter - to convert from incoming Message data to a type declared by the function.

  2. ConversionService - to convert from incoming non-Message data to a type declared by the function.

This means that depending on the type of the raw data (Message or non-Message) Spring Cloud Function will apply one or the other mechanisms.

For most cases when dealing with functions that are invoked as part of some other request (e.g., HTTP, Messaging etc) the framework relies on MessageConverters, since such requests already converted to Spring Message. In other words, the framework locates and applies the appropriate MessageConverter. To accomplish that, the framework needs some instructions from the user. One of these instructions is already provided by the signature of the function itself (Person type). Consequently, in theory, that should be (and, in some cases, is) enough. However, for the majority of use cases, in order to select the appropriate MessageConverter, the framework needs an additional piece of information. That missing piece is contentType header.

Such header usually comes as part of the Message where it is injected by the corresponding adapter that created such Message in the first place. For example, HTTP POST request will have its content-type HTTP header copied to contentType header of the Message.

For cases when such header does not exist framework relies on the default content type as application/json.

Content Type versus Argument Type

As mentioned earlier, for the framework to select the appropriate MessageConverter, it requires argument type and, optionally, content type information. The logic for selecting the appropriate MessageConverter resides with the argument resolvers which trigger right before the invocation of the user-defined function (which is when the actual argument type is known to the framework). If the argument type does not match the type of the current payload, the framework delegates to the stack of the pre-configured MessageConverters to see if any one of them can convert the payload.

The combination of contentType and argument type is the mechanism by which framework determines if message can be converted to a target type by locating the appropriate MessageConverter. If no appropriate MessageConverter is found, an exception is thrown, which you can handle by adding a custom MessageConverter (see User-defined Message Converters).

Do not expect Message to be converted into some other type based only on the contentType. Remember that the contentType is complementary to the target type. It is a hint, which MessageConverter may or may not take into consideration.

Message Converters

MessageConverters define two methods:

Object fromMessage(Message<?> message, Class<?> targetClass);

Message<?> toMessage(Object payload, @Nullable MessageHeaders headers);

It is important to understand the contract of these methods and their usage, specifically in the context of Spring Cloud Stream.

The fromMessage method converts an incoming Message to an argument type. The payload of the Message could be any type, and it is up to the actual implementation of the MessageConverter to support multiple types.

Provided MessageConverters

As mentioned earlier, the framework already provides a stack of MessageConverters to handle most common use cases. The following list describes the provided MessageConverters, in order of precedence (the first MessageConverter that works is used):

  1. JsonMessageConverter: Supports conversion of the payload of the Message to/from POJO for cases when contentType is application/json using Jackson (DEFAULT) or Gson libraries. This message converter also aware of type parameter (e.g., application/json;type=foo.bar.Person). This is useful for cases where types may not be known at the time when function is developed, hence function signature may look like Function<?, ?> or Function or Function<Object, Object>. In other words for type conversion we typically derive type from function signature. Having, mime-type parameter allows you to communicate type in a more dynamic way.

  2. ByteArrayMessageConverter: Supports conversion of the payload of the Message from byte[] to byte[] for cases when contentType is application/octet-stream. It is essentially a pass through and exists primarily for backward compatibility.

  3. StringMessageConverter: Supports conversion of any type to a String when contentType is text/plain.

When no appropriate converter is found, the framework throws an exception. When that happens, you should check your code and configuration and ensure you did not miss anything (that is, ensure that you provided a contentType by using a binding or a header). However, most likely, you found some uncommon case (such as a custom contentType perhaps) and the current stack of provided MessageConverters does not know how to convert. If that is the case, you can add custom MessageConverter. See User-defined Message Converters.

User-defined Message Converters

Spring Cloud Function exposes a mechanism to define and register additional MessageConverters. To use it, implement org.springframework.messaging.converter.MessageConverter, configure it as a @Bean. It is then appended to the existing stack of `MessageConverter`s.

It is important to understand that custom MessageConverter implementations are added to the head of the existing stack. Consequently, custom MessageConverter implementations take precedence over the existing ones, which lets you override as well as add to the existing converters.

The following example shows how to create a message converter bean to support a new content type called application/bar:

@SpringBootApplication
public static class SinkApplication {

    ...

    @Bean
    public MessageConverter customMessageConverter() {
        return new MyCustomMessageConverter();
    }
}

public class MyCustomMessageConverter extends AbstractMessageConverter {

    public MyCustomMessageConverter() {
        super(new MimeType("application", "bar"));
    }

    @Override
    protected boolean supports(Class<?> clazz) {
        return (Bar.class.equals(clazz));
    }

    @Override
    protected Object convertFromInternal(Message<?> message, Class<?> targetClass, Object conversionHint) {
        Object payload = message.getPayload();
        return (payload instanceof Bar ? payload : new Bar((byte[]) payload));
    }
}

Note on JSON options

In Spring Cloud Function we support Jackson and Gson mechanisms to deal with JSON. And for your benefit have abstracted it under org.springframework.cloud.function.json.JsonMapper which itself is aware of two mechanisms and will use the one selected by you or following the default rule. The default rules are as follows:

  • Whichever library is on the classpath that is the mechanism that is going to be used. So if you have com.fasterxml.jackson.* to the classpath, Jackson is going to be used and if you have com.google.code.gson, then Gson will be used.

  • If you have both, then Gson will be the default, or you can set spring.cloud.function.preferred-json-mapper property with either of two values: gson or jackson.

That said, the type conversion is usually transparent to the developer, however given that org.springframework.cloud.function.json.JsonMapper is also registered as a bean you can easily inject it into your code if needed.

Kotlin Lambda support

We also provide support for Kotlin lambdas (since v2.0). Consider the following:

@Bean
open fun kotlinSupplier(): () -> String {
    return  { "Hello from Kotlin" }
}

@Bean
open fun kotlinFunction(): (String) -> String {
    return  { it.toUpperCase() }
}

@Bean
open fun kotlinConsumer(): (String) -> Unit {
    return  { println(it) }
}

The above represents Kotlin lambdas configured as Spring beans. The signature of each maps to a Java equivalent of Supplier, Function and Consumer, and thus supported/recognized signatures by the framework. While mechanics of Kotlin-to-Java mapping are outside of the scope of this documentation, it is important to understand that the same rules for signature transformation outlined in "Java 8 function support" section are applied here as well.

To enable Kotlin support all you need is to add Kotlin SDK libraries on the classpath which will trigger appropriate autoconfiguration and supporting classes.

Function Component Scan

Spring Cloud Function will scan for implementations of Function, Consumer and Supplier in a package called functions if it exists. Using this feature you can write functions that have no dependencies on Spring - not even the @Component annotation is needed. If you want to use a different package, you can set spring.cloud.function.scan.packages. You can also use spring.cloud.function.scan.enabled=false to switch off the scan completely.