Provides Optional support inside Hibernate with Panache and MongoDB with Panache

[loicmathieu]

Fixes #1368

This is a draft PR as when #5583 is merged we will want to consider also having overload for with LockModeType for findByIdOptional.

cc @FroMage and @emmanuelbernard

[FroMage]

LGTM

[FroMage]

Ah wait… it's missing tests.

[FroMage]

Missing tests

[FroMage]

CI failed due to formatting errors.

[loicmathieu]

@FroMage I add tests and correct a bunch of bugs :)

I added an overload on findByIdOptional with a LockModeType

Two questions:

• I don't very like the implementatiion of PanacheQueryImpl.singleResultOptional that reads the database with a limit of 2 to be sure that there is 0 or 1 items. But I don't comes up with something better. If you have an idea ?
• Generics implementation of PanacheEntityBase is not very easily usable and we need to use a not very known construct Person.<Person>findByIdOptional(personId); . I don't know if we have a better option here.

[FroMage]

I don't very like the implementatiion of PanacheQueryImpl.singleResultOptional that reads the database with a limit of 2 to be sure that there is 0 or 1 items. But I don't comes up with something better. If you have an idea ?

No, I think that's fine.

Generics implementation of PanacheEntityBase is not very easily usable and we need to use a not very known construct Person.<Person>findByIdOptional(personId); . I don't know if we have a better option here.

The Person.<Person>findByIdOptional(personId); issue is the same you'll get with the RX mongodb API. We have to tell users to use intermediate variables:

Optional<Person> res = Person.findByIdOptional(personId);
res...

[loicmathieu]

@FroMage I updated the documenation to use an interlediate variable.
The PR is ready now :)

[FroMage]

Keycloak tests failed due to:

Caused by: java.lang.ClassNotFoundException: io.quarkus.jackson.ObjectMapperProducer

Panache tests failed due to:

org.opentest4j.AssertionFailedError: expected: <{"id":null,"name":"max","uniqueName":null,"address":null,"status":null,"dogs":[],"serialisationTrick":1}> but was: <{"id":null,"name":"max","uniqueName":null,"address":null,"status":null,"dogs":[],"serialisationTrick":1,"persistent":false}>
	at org.junit.jupiter.api.AssertionUtils.fail(AssertionUtils.java:55)
	at org.junit.jupiter.api.AssertionUtils.failNotEqual(AssertionUtils.java:62)
	at org.junit.jupiter.api.AssertEquals.assertEquals(AssertEquals.java:182)
	at org.junit.jupiter.api.AssertEquals.assertEquals(AssertEquals.java:177)
	at org.junit.jupiter.api.Assertions.assertEquals(Assertions.java:1124)
	at io.quarkus.it.panache.PanacheFunctionalityTest.jacksonDeserilazationIgnoresPersitantAttribute(PanacheFunctionalityTest.java:71)

Except the last one is one of those tests that can't possibly fail. It's a test that's reusing a module which doesn't have the fix.

[loicmathieu]

@FroMage I just launched the Hibernate with Panache native integration test on the branch locally and it pass. Can you re-launch the CI ? We saw multiple CI intermitent failures lately ....

[FroMage]

Re-running them, but last time, once they started failing, they kept failing. The CI was corrupted. Let's see how this goes.

[machi1990]

@loicmathieu you can rebase with master an force push to have a fresh CI run

[loicmathieu]

@machi1990 done 🤞

[machi1990]

Thanks @loicmathieu let’s see what CI has to say.

[loicmathieu]

@machi1990 still failing seems CI related :(

[FroMage]

That's what I thought: once it's bust, it's dead. Perhaps open a new PR?

[FroMage]

@machi1990 you were on that PR which had the same issues last week, right? How did it get solved there?

[machi1990]

@machi1990 still failing seems CI related :(

@loicmathieu :-( shame, hope we get this sorted soon.

@machi1990 you were on that PR which had the same issues last week, right? How did it get solved there?

Yeah, I have seen CI related issues in some PR lately and many (including mine) were solved by rebasing on top of master. But there was this #5306 PR that you manually had to connect to Azure and clear the cache or something. @n1hility helped with that.

[geoand]

Rebase onto master is usually the way to go. We need to look into this more as part of #5520

[loicmathieu]

Rebased again on master, if it still fails I will close it and open a new one ... 🤞

[loicmathieu]

@FroMage @machi1990 @geoand it works !!! 🎉

[FroMage]

LGTM except minor points.

[loicmathieu]

@FroMage I think I adress all your points. Can you approve thsi PR ?

[FroMage]

Done, thanks.