Upgrade to Hibernate ORM 5.6.15.Final

[yrodiere]

Fixes #30234

Release notes for ORM 5.6.15.Final: https://in.relation.to/2023/02/06/hibernate-orm-5615-final/

Note this patch should apply cleanly to 2.16, at least for the Java part, as it does not rely on JPA classes directly.

[quarkus-bot]

/cc @Sanne (hibernate-orm), @gsmet (hibernate-orm)

[gsmet]

Don't merge this yet. Something went wrong with the CI.

[gsmet]

@famod I think something went wrong with GIB here. And the only culprit I could see would be the set-output change as AFAICS GIB lists the modules properly.

[yrodiere]

@gsmet It just occurred to me the tests won't work on 2.13/2.16 because they rely on a feature introduced after 2.16.0: #30640

I can either submit a different PR to 2.16 that relies on quarkus.hibernate-orm.unsupported-properties instead, or do the change directly in this PR. Any preference?

[gsmet]

Or we also backport #30640 to 2.16?

(That won't solve the 2.13 issue though)

[gsmet]

Given we want this in 2.13, maybe a PR targeting 2.16 that I could apply to 2.13 too would be better.

[yrodiere]

Given we want this in 2.13, maybe a PR targeting 2.16 that I could apply to 2.13 too would be better.

Done: #30956

The labels on that PR might not be what you need; sorry about that, I'm not sure what to do when actually backporting since you always do it for me :)

[quarkus-bot]

Failing Jobs - Building b189ead

Status	Name	Step	Failures	Logs	Raw logs
✔️	JVM Tests - JDK 11
✖	JVM Tests - JDK 11 Windows	Build	Failures	Logs	Raw logs
✔️	JVM Tests - JDK 17
✔️	JVM Tests - JDK 18

Full information is available in the Build summary check run.

Failures

⚙️ JVM Tests - JDK 11 Windows #

- Failing: integration-tests/mongodb-rest-data-panache 

📦 integration-tests/mongodb-rest-data-panache

✖ io.quarkus.it.mongodb.rest.data.panache.MongoDbRestDataPanacheTest.shouldNotCreateOrDeleteAuthor - More details - Source on GitHub

java.lang.RuntimeException: 
java.lang.RuntimeException: io.quarkus.builder.BuildException: Build failure: Build failed due to errors
	[error]: Build step io.quarkus.mongodb.deployment.DevServicesMongoProcessor#startMongo threw an exception: java.lang.RuntimeException: java.lang.IllegalStateException: Previous attempts to find a Docker environment failed. Will not retry. Please see logs and check configuration

[famod]

@famod I think something went wrong with GIB here. And the only culprit I could see would be the set-output change as AFAICS GIB lists the modules properly.

@gsmet do you have a link to a worflow run that failed? I can't seem to find one here...

[gsmet]

@famod it didn't fail per se but it didn't run the JVM tests nor the native tests: https://github.com/quarkusio/quarkus/actions/runs/4113233527

So something is off.

[famod]

@gsmet the delimiter differs:

• https://github.com/quarkusio/quarkus/actions/runs/4113233527/jobs/7099579657#step:5:27 %0A (html linefeed?)
• https://github.com/quarkusio/quarkus/actions/runs/4121765793/jobs/7118084939#step:5:27 "proper" linefeed

[gsmet]

We are doing something related to that here: https://github.com/quarkusio/quarkus/pull/30867/files#diff-d8e36253822cffbdf21e333bcba467345b46e6095927dda98399fe7f21f4679aL168-R168

But I don't understand why it would change (except if GitHub is doing some magic).

[famod]

Ah, I vaguely remember that one. Since we don't convert %0A back to linefeed, I think the problem makes sense.
::set-output was probably not "symmetric" in that sense.

[gsmet]

So do you think we could just push the line feeds to the file without replacing them at all?

Or do we need to change them back when we read the output?

[famod]

I'll try to find some time this evening to test this in a sample repo.

[famod]

@gsmet #31019 contains the fix