I’m looking to gain a better understanding of the upcoming project isolation features. I’ve recently upgraded to gradle 8.8, and am trying to wrap my head around IsolatedProject and gradle.lifecycle.beforeProject/afterProject. I mostly understand the gist of it, but I am confused why the lambda parameter of gradle.lifecycle.beforeProject/afterProject is a Project and not an IsolatedProject:
public interface GradleLifecycle {
void beforeProject(IsolatedAction<? super Project> action);
void afterProject(IsolatedAction<? super Project> action);
}
The code above is the current implementation. What I would expect to see, however, would be:
public interface GradleLifecycle {
void beforeProject(IsolatedAction<? super IsolatedProject> action);
void afterProject(IsolatedAction<? super IsolatedProject> action);
}
If I understand correctly, an IsolatedAction is a serializable configuration operation which cannot modify global state, and thus is parallel-safe. And also if I understand correctly, an IsolatedProject is a constrained view of a Project which cannot access other projects, making it more parallel safe. Therefore, why would a Project be passed into the IsolatedAction instead of an IsolatedProject? It seems to me that whatever extra features Project has that IsolatedProject does not have wouldn’t be safe to use in an IsolatedAction.