In more complex rendering hierarchies child-nodes might be unmounted, before the parents get unmounted. Thereforce we have to make sure that registered beforeUnmount-actions are only executed, if their job is still active, as it may be called by its own MountPoint or by any MountPoint of its parent.
This PR adds the necessary safety belts and unit tests.
If some beforeUnmount-action throws, the exception is caught and logged to the console in order to keep the application alive.
In more complex rendering hierarchies child-nodes might be unmounted, before the parents get unmounted. Thereforce we have to make sure that registered
beforeUnmount
-actions are only executed, if their job is still active, as it may be called by its ownMountPoint
or by anyMountPoint
of its parent.This PR adds the necessary safety belts and unit tests.
If some
beforeUnmount
-action throws, the exception is caught and logged to the console in order to keep the application alive.