Vert.x 4.5.0 released!

[Julien Viet]

Hi everyone,

we have just released Eclipse Vert.x 4.5.0 which comes with a great
set of new features, including virtual thread support!

Here is the blog announcement https://vertx.io/blog/eclipse-vert-x-4-5-0/
and most importantly a recap of new important things you can find
https://vertx.io/blog/whats-new-in-vert-x-4-5/

Kudos to all contributors for the great work 🎉

Note : some features announced were already available in 4.4.x but not
properly announced, so we thought it was best to recap all the
important changes since the 4.4.0 announce.

Enjoy

Julien

[Yeikel]

Thank you for the release, many exiting features!  

I just wanted to drop by to mention, that the following code : 

private <T> TestObserver<T> futureToTestObserver(Future<T> future) {
final TestObserver<T> testObserver = new TestObserver<>();
SingleHelper.toSingle(future::onComplete).subscribe(testObserver);
return testObserver;
}

No longer compiles with Vertx 4.5.0 due to : Reference to 'onComplete' is ambiguous, both 'onComplete(Handler<AsyncResult<T>>)' and 'onComplete(Handler<T>, Handler<Throwable>)' match

I believe that this was introduced with https://github.com/eclipse-vertx/vert.x/pull/4830, and it is a minor breaking change that I do not know if it was intentional  given that I could not find any reference to the release notes about the same

Thank you! 

[Julien Viet]

Hi,

can you the toSingle method signature with which it creates an issue ?

Julien

> --
> You received this message because you are subscribed to the Google Groups "vert.x" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to vertx+un...@googlegroups.com.
> To view this discussion on the web, visit https://groups.google.com/d/msgid/vertx/a6b25e08-f27b-471c-8920-bec050e3c7b4n%40googlegroups.com.

[Yeikel]

Hi, 

Your message was cut-off but If you meant to ask what's the signature of the toSingle method, here it is : 

package io.vertx.reactivex;

// imports omitted for brevity 
public class SingleHelper {

/**
* Returns a {@link Single} that, when subscribed, uses the provided {@code handler} to adapt a callback-based asynchronous method.
* <p>
* For example:
* <pre> {@code
* io.vertx.core.Vertx vertx = Vertx.vertx();
* Single<String> deploymentId = SingleHelper.toSingle(handler -> vertx.deployVerticle("org.acme.MyVerticle", handler));
* }</pre>
* <p>
* This is useful when using RxJava without the Vert.x Rxified API or your own asynchronous methods.
* <p>
* The asynchronous method result <strong>must not</strong> be {@code null}, as an RxJava 2 {@link Single} does not allow {@code null} values.
*
* @param handler the code executed when the returned {@link Single} is subscribed
*/
public static <T> Single<T> toSingle(Consumer<Handler<AsyncResult<T>>> handler) {
return AsyncResultSingle.toSingle(handler);
}

[ariffi...@gmail.com]

Hi Julien,

I appreciate the constant improvements to new versions.

I am upgrading to 4.5.0.... I changed the DeploymentOptions to reflect the new way to specify a worker verticle:

// Deploy Cassandra verticle as a worker verticle
DeploymentOptions dbOptions = new DeploymentOptions().setThreadingModel(ThreadingModel.WORKER).setInstances(verticleInstanceCounts * 4).setHa(true);
vertx.deployVerticle(verticleFactory.prefix() + ":" + BeaconDbApiVerticle.class.getName(), dbOptions, results -> {
if (results.succeeded()) {
LOGGER.info("Cassandra DB Deployment id is: " + results.result());
} else {
LOGGER.error("Cassandra DB Deployment failed! " + results + " Cause :" + results.cause().toString());
System.exit(1);
}
});

But I am getting a deprecated warning below:

=======================================================

ariffin@Ariffins-iMac-2020 beaconApi % java -version

openjdk version "17.0.9" 2023-10-17 LTS

OpenJDK Runtime Environment Corretto-17.0.9.8.1 (build 17.0.9+8-LTS)

OpenJDK 64-Bit Server VM Corretto-17.0.9.8.1 (build 17.0.9+8-LTS, mixed mode, sharing)

ariffin@Ariffins-iMac-2020 beaconApi % ./gradlew clean build 

Starting a Gradle Daemon (subsequent builds will be faster)

> Task :compileJava

/Volumes/Ariffin EXT/Dev/git/beaconApi/src/main/java/com/analytics/BeaconApiApplication.java:102: warning: [deprecation] setThreadingModel(ThreadingModel) in DeploymentOptions has been deprecated

        DeploymentOptions dbOptions = new DeploymentOptions().setThreadingModel(ThreadingModel.WORKER).setInstances(verticleInstanceCounts * 4).setHa(true);

                                                             ^

1 warning

BUILD SUCCESSFUL in 9s

6 actionable tasks: 6 executed

=======================================================

Can you tell me what I am doing wrong?

Thank you.

-a

[ariffi...@gmail.com]

Bump.

-a

[Julien Viet]

Hi,

that is a mistake that only appears in Java 11

if you compile with Java 8 that will not happen.

it will be fixed in 4.5.1

thanks for bringing this valuable issue to my eyes.

Julien

> --
> You received this message because you are subscribed to the Google Groups "vert.x" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to vertx+un...@googlegroups.com.

> To view this discussion on the web, visit https://groups.google.com/d/msgid/vertx/f4cb90c7-147a-406d-908d-164aa32d75f2n%40googlegroups.com.

[Yeikel]

How you'd refactor the following snippet after the changes to the Future class? 

The alternatives I am thinking about seem overly complicated

return SingleHelper.toSingle(resultFuture::onComplete).toObservable().onErrorResumeNext(err -> {
if (err instanceof DocumentNotFoundException) {
return Observable.empty();
} else {
return Observable.error(err);
}
});

[Julien Viet]

Hi,

you should get rid of single and use await on resultFuture

try {
return Collections.singletonList(await(resultFuture));
} catch (DocumentNotFoundException) {
return Collections.emptyList();

> To view this discussion on the web, visit https://groups.google.com/d/msgid/vertx/37f29ce8-9cb8-4393-8a6a-9cff6325b650n%40googlegroups.com.

[Yeikel]

Thank you for the suggestion but in this case it is a library and that would mean cha ging the return type and a larger refactoring excercise

Also, from what I understand, I'd need Java 21 to use await, no? We're still using Java 17 and migrating will take some time and effort

To view this discussion on the web, visit https://groups.google.com/d/msgid/vertx/CA%2BH-X4_wQ2HqbAzJDDE0fieyw-bmz2TbYMUwVDtOGAESC2qhug%40mail.gmail.com.

[Julien Viet]

I see.

await is actually very new in the Java landscape and we don't expect
an adoption before a few years.

> To view this discussion on the web, visit https://groups.google.com/d/msgid/vertx/CAMMhyA3hhqGL3aBAPvxG4jaMVRKcSF1fFa5rZiJGuDRnQLdELA%40mail.gmail.com.

[Yeikel]

Yeah, that's understandable

How do you suggest I refactor the code I mentioned? The earlier code was "clean" and now it seems that after the Future class changes, I need to do some overly complicated alternative. Maybe I am overthinking it? 

Thanks!