Skip to content

Conversation

Postremus
Copy link
Member

The resteasy reactive param converter, which gets automatically generated, calls the fromString/valueOf/constructor of the wrapper class directly. This fails if the paramconveter and the wrapper class is housed in different class loader. We can ensure that the TCCL is the same CL as the wrapper class, by not considering the wrapper an application class anymore if it is in an external jar.

related to #47471

This reverts #39691 (but also adds tests)
Follows up on #39823

The resteasy reactive param converter, which gets automatically generated, calls the fromString/valueOf/constructor of the wrapper class directly.
This fails if the paramconveter and the wrapper class is housed in different class loader. We can ensure that the TCCL is the same CL as the wrapper class, by not considering the wrapper an application class anymore if it is in an external jar.
Copy link

quarkus-bot bot commented Apr 27, 2025

Status for workflow Quarkus CI

This is the status report for running Quarkus CI on commit cea28c7.

✅ The latest workflow run for the pull request has completed successfully.

It should be safe to merge provided you have a look at the other checks in the summary.

You can consult the Develocity build scans.

@gsmet gsmet merged commit ae9f7bc into quarkusio:main Apr 28, 2025
50 checks passed
@quarkus-bot quarkus-bot bot added this to the 3.23 - main milestone Apr 28, 2025
@Postremus Postremus deleted the issues/47471-rr-app-class-again branch April 28, 2025 08:38
@gsmet gsmet modified the milestones: 3.23 - main, 3.22.1 Apr 29, 2025
@jmartisk jmartisk modified the milestones: 3.22.1, 3.20.1 May 6, 2025
gsmet added a commit to gsmet/quarkus that referenced this pull request May 7, 2025
gsmet added a commit to gsmet/quarkus that referenced this pull request May 7, 2025
@gsmet
Copy link
Member

gsmet commented May 7, 2025

@Postremus I would like to clarify something about this PR, does it make sense to backport it to 3.20? Or should it be merged only in branches with the class loader changes from Holly?

I backported it to 3.20 and tests pass but I'm not sure it's the right thing to do.

gsmet added a commit to gsmet/quarkus that referenced this pull request May 13, 2025
gsmet added a commit to gsmet/quarkus that referenced this pull request Jun 25, 2025
@gsmet gsmet modified the milestones: 3.20.1, 3.20.2 Jun 25, 2025
@gsmet
Copy link
Member

gsmet commented Aug 13, 2025

@jmartisk I see this one in 3.20 AFAICS. I don't think we need to backport it again. Can you check?

@jmartisk
Copy link
Contributor

Right, the related product JIRA was in the wrong state hence I thought this still wasn't in, but this is already in 3.20.2... Thanks for spotting it.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants