Michael Edgar

Yes, I can do some more experimentation with test metaspace allocations. What is your thinking on the issue with rest-assured potentially leaking metaspace memory from json-path evaluations? It seems as...

@gsmet I have a PR open [1] in rest-assured that seems to make a significant difference in the metaspace issue. Regarding Jayway, I was still using RestAssured for making HTTP...

While looking at the re-base of https://github.com/rest-assured/rest-assured/pull/1844 and why Groovy 5 causes new leaks, I've found there are several locations in Groovy that start-up "cleaner" threads [1]. Forcing the Groovy...

Is the loading/reloading an issue outside of the `QuarkusUnitTest` use case?

> I think we should create a PR with this and then ask the people at https://github.com/quarkiverse/quarkus-groovy for feedback. Is anyone planning on this? If not, I'll open the PR...

Adding the three Groovy artifacts in `core/runtime/pom.xml` definitely helps with the thread leaking issue. It does not work with rest-assured 5.5.6 on Quarkus `main` because it breaks the internal `ThreadLocal`...

@lloydmeta this is a common issue. The correct way to indicate that a class should be used as the `items` of an array is to specify `implementation = Greeting[].class` rather...

If the previous suggestion doesn't work, you can try adding `@Schema` to the `Greeting` class also.

This should get things closer to want you expect. Note that `ImmutableList` should likely extend/implement `List` or `Iterable` for it to behave how it would using some collection library. ```java...

This does seem like something that probably should be supported. This might take some care, with the scoping/enclosure functionality of the `@NullMarked` and `@NullUnmarked` annotations. I don't believe we have...