Skip to content
Permalink
Branch: master
Find file Copy path
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
436 lines (325 sloc) 19.6 KB

Spring Cloud Function

Mark Fisher, Dave Syer, Oleg Zhurakousky, Anshul Mehra

{spring-cloud-function-version}


Getting Started

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, using primitives defined by the Project Reactor (i.e., Flux<T> and Mono<T>). Users can supply a bean of type Function<String,String>, for instance, and the FunctionCatalog will wrap it into a Function<Flux<String>,Flux<String>>.

Using Reactor based primitives not only helps with the canonical representation of user defined functions, but it also facilitates a more robust and flexible(reactive) execution model.

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.

Java 8 function support

Generally speaking users can expect that if they write a function for a plain old Java type (or primitive wrapper), then the function catalog will wrap it to a Flux of the same type. If the user writes a function using Message (from spring-messaging) it will receive and transmit headers from any adapter that supports key-value metadata (e.g. HTTP headers). Here are the details.

User Function Catalog Registration

Function<S,T>

Function<Flux<S>, Flux<T>>

Function<Message<S>,Message<T>>

Function<Flux<Message<S>>, Flux<Message<T>>>

Function<Flux<S>, Flux<T>>

Function<Flux<S>, Flux<T>> (pass through)

Supplier<T>

Supplier<Flux<T>>

Supplier<Flux<T>>

Supplier<Flux<T>>

Consumer<T>

Function<Flux<T>, Mono<Void>>

Consumer<Message<T>>

Function<Flux<Message<T>>, Mono<Void>>

Consumer<Flux<T>>

Consumer<Flux<T>>

Supplier

As you can see from the table above 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.PollableSupplier 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 PollableSupplier annotation exposes a splittable attribute to signal that produced stream needs to be split (see Splitter EIP)

Here is the example:

@PollableSupplier(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. If you declare a Consumer of a non publisher type (which is normal), it will be converted to a function that returns a publisher, so that it can be subscribed to in a controlled way.

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.

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

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 then 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;

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. 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 which allows you to use Spring Expression Language (SpEL) and provide SpEL expression that should resolve into definition of a function (as described above).

Note
SpEL evaluation context’s root object is the actual input argument, so in he 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).

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).

Important
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).

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 spring-cloud-function-kotlin module to your classpath which contains the appropriate autoconfiguration and supporting classes.

Standalone Web Applications

The spring-cloud-function-web module has autoconfiguration that activates when it is included in a Spring Boot web application (with MVC support). There is also a spring-cloud-starter-function-web to collect all the optional dependencies in case you just want a simple getting started experience.

With the web configurations activated your app will have an MVC endpoint (on "/" by default, but configurable with spring.cloud.function.web.path) that can be used to access the functions in the application context. The supported content types are plain text and JSON.

Method Path Request Response Status

GET

/{supplier}

-

Items from the named supplier

200 OK

POST

/{consumer}

JSON object or text

Mirrors input and pushes request body into consumer

202 Accepted

POST

/{consumer}

JSON array or text with new lines

Mirrors input and pushes body into consumer one by one

202 Accepted

POST

/{function}

JSON object or text

The result of applying the named function

200 OK

POST

/{function}

JSON array or text with new lines

The result of applying the named function

200 OK

GET

/{function}/{item}

-

Convert the item into an object and return the result of applying the function

200 OK

As the table above shows the behaviour of the endpoint depends on the method and also the type of incoming request data. When the incoming data is single valued, and the target function is declared as obviously single valued (i.e. not returning a collection or Flux), then the response will also contain a single value. For multi-valued responses the client can ask for a server-sent event stream by sending `Accept: text/event-stream".

If there is only a single function (consumer etc.) in the catalog, the name in the path is optional. Composite functions can be addressed using pipes or commas to separate function names (pipes are legal in URL paths, but a bit awkward to type on the command line).

For cases where there is more then a single function in catalog and you want to map a specific function to the root path (e.g., "/"), or you want to compose several functions and then map to the root path you can do so by providing spring.cloud.function.definition property which essentially used by spring-=cloud-function-web module to provide default mapping for cases where there is some type of a conflict (e.g., more then one function available etc).

For example,

--spring.cloud.function.definition=foo|bar

The above property will compose 'foo' and 'bar' function and map the composed function to the "/" path.

Functions and consumers that are declared with input and output in Message<?> will see the request headers on the input messages, and the output message headers will be converted to HTTP headers.

When POSTing text the response format might be different with Spring Boot 2.0 and older versions, depending on the content negotiation (provide content type and accpt headers for the best results).

Standalone Streaming Applications

To send or receive messages from a broker (such as RabbitMQ or Kafka) you can leverage spring-cloud-stream project and it’s integration with Spring Cloud Function. Please refer to Spring Cloud Function section of the Spring Cloud Stream reference manual for more details and examples.

Deploying a Packaged Function

Spring Cloud Function provides a "deployer" library that allows you to launch a jar file (or exploded archive, or set of jar files) with an isolated class loader and expose the functions defined in it. This is quite a powerful tool that would allow you to, for instance, adapt a function to a range of different input-output adapters without changing the target jar file. Serverless platforms often have this kind of feature built in, so you could see it as a building block for a function invoker in such a platform (indeed the Riff Java function invoker uses this library).

The standard entry point is to add spring-cloud-function-deployer to the classpath, the deployer kicks in and looks for some configuration to tell it where to find the function jar.

<dependency>
	<groupId>org.springframework.cloud</groupId>
	<artifactId>spring-cloud-function-deployer</artifactId>
	<version>${spring.cloud.function.version}</version>
</dependency>

At a minimum the user has to provide a spring.cloud.function.location which is a URL or resource location for the archive containing the functions. It can optionally use a maven: prefix to locate the artifact via a dependency lookup (see FunctionProperties for complete details). A Spring Boot application is bootstrapped from the jar file, using the MANIFEST.MF to locate a start class, so that a standard Spring Boot fat jar works well, for example. If the target jar can be launched successfully then the result is a function registered in the main application’s FunctionCatalog. The registered function can be applied by code in the main application, even though it was created in an isolated class loader (by deault).

Here is the example of deploying a JAR which contains an 'uppercase' function and invoking it .

@SpringBootApplication
public class DeployFunctionDemo {

	public static void main(String[] args) {
		ApplicationContext context = SpringApplication.run(DeployFunctionDemo.class,
				"--spring.cloud.function.location=..../target/uppercase-0.0.1-SNAPSHOT.jar",
				"--spring.cloud.function.function-name=uppercase");

		FunctionCatalog catalog = context.getBean(FunctionCatalog.class);
		Function<String, String> function = catalog.lookup("uppercase");
		System.out.println(function.apply("hello"));
	}
}

Functional Bean Definitions

Dynamic Compilation

There is a sample app that uses the function compiler to create a function from a configuration property. The vanilla "function-sample" also has that feature. And there are some scripts that you can run to see the compilation happening at run time. To run these examples, change into the scripts directory:

cd scripts

Also, start a RabbitMQ server locally (e.g. execute rabbitmq-server).

Start the Function Registry Service:

./function-registry.sh

Register a Function:

./registerFunction.sh -n uppercase -f "f->f.map(s->s.toString().toUpperCase())"

Run a REST Microservice using that Function:

./web.sh -f uppercase -p 9000
curl -H "Content-Type: text/plain" -H "Accept: text/plain" localhost:9000/uppercase -d foo

Register a Supplier:

./registerSupplier.sh -n words -f "()->Flux.just(\"foo\",\"bar\")"

Run a REST Microservice using that Supplier:

./web.sh -s words -p 9001
curl -H "Accept: application/json" localhost:9001/words

Register a Consumer:

./registerConsumer.sh -n print -t String -f "System.out::println"

Run a REST Microservice using that Consumer:

./web.sh -c print -p 9002
curl -X POST -H "Content-Type: text/plain" -d foo localhost:9002/print

Run Stream Processing Microservices:

First register a streaming words supplier:

./registerSupplier.sh -n wordstream -f "()->Flux.interval(Duration.ofMillis(1000)).map(i->\"message-\"+i)"

Then start the source (supplier), processor (function), and sink (consumer) apps (in reverse order):

./stream.sh -p 9103 -i uppercaseWords -c print
./stream.sh -p 9102 -i words -f uppercase -o uppercaseWords
./stream.sh -p 9101 -s wordstream -o words

The output will appear in the console of the sink app (one message per second, converted to uppercase):

MESSAGE-0
MESSAGE-1
MESSAGE-2
MESSAGE-3
MESSAGE-4
MESSAGE-5
MESSAGE-6
MESSAGE-7
MESSAGE-8
MESSAGE-9
...

Serverless Platform Adapters

As well as being able to run as a standalone process, a Spring Cloud Function application can be adapted to run one of the existing serverless platforms. In the project there are adapters for AWS Lambda, Azure, and Apache OpenWhisk. The Oracle Fn platform has its own Spring Cloud Function adapter. And Riff supports Java functions and its Java Function Invoker acts natively is an adapter for Spring Cloud Function jars.

You can’t perform that action at this time.