#1406 silently skipped the call to moveDirectory in case newBuildDir.isDirectory(), rather than throwing an IOE as renameTo does.
Also deleting apparently useless call to forceMkdir.

[JENKINS-25691] Might have prevented the need for: https://github.com/jenkinsci/maven-plugin/pull/33

Allow Projects to take the Queue.Item's actions into consideration when determining their default authentication

- Also tidy up some holes in locking strategy that can cause builds to be scheduled on
  executors that are about to be interrupted or nodes in the process of being removed

jenkins.model.Jenkins::getInstance() method is marked as @CheckForNull, but the most of Jenkins code does not actually check the return value. It leads to tons of static analysis errors, hence it is hard to analyze the code.

The change also adds annotations to old methods like Hudson::getInstance()
Signed-off-by: NOleg Nenashev <o.v.nenashev@gmail.com>

[JENKINS-2786] Makes no sense for Computer.currentComputer to return MasterComputer when not on an executor thread.

[FIXED JENKINS-25400] Rework fix of JENKINS-22769 (c04cdcd9) to put the burden on each listener to impersonate ACL.SYSTEM if it needs to.

The issue is related to JENKINS-18351.
Signed-off-by: NOleg Nenashev <o.v.nenashev@gmail.com>

Adjust my prior change to better utilize structured form submission to accomodate a plugins that are not using p:config-assignedLabel

Adjust my prior change to better utilize structured form submission to accomodate a few plugins that are improperly submitting data through this processing logic

fromExternalizableId should just return null, not throw IAE, if the job has since been renamed/move/deleted.
(It could already return null if the run has since been deleted; #1264 failed to mark this properly.)
Reserving IAE for inputs that getExternalizableId could not actually have produced,
and making sure to also throw this in case of non-numeric text after the hash.
Adding Javadoc for both methods.

The original fix was made in 0c3d6709.
In this fix, I'm also moving the code that creates LeftItem.

It makes no sense that there's Item.cancel(Queue) and Queue.cancel(Item)
and they do different things!

Observed in a test against 1.480:
… hudson.model.Executor run
SEVERE: Unexpected executor death
java.lang.NullPointerException
	at hudson.model.Computer.getNode(Computer.java:473)
	at hudson.slaves.SlaveComputer.getNode(SlaveComputer.java:177)
	at hudson.slaves.SlaveComputer.getRetentionStrategy(SlaveComputer.java:609)
	at hudson.slaves.SlaveComputer.taskCompleted(SlaveComputer.java:285)
	at hudson.model.queue.WorkUnitContext.synchronizeEnd(WorkUnitContext.java:140)
	at hudson.model.Executor.run(Executor.java:253)

Signed-off-by: NOleg Nenashev <o.v.nenashev@gmail.com>

The fix also includes comments in #1424 regarding Nonnull usage
Signed-off-by: NOleg Nenashev <o.v.nenashev@gmail.com>

Signed-off-by: NOleg Nenashev <o.v.nenashev@gmail.com>

This changes prevents the improper usage of method without the preliminary supportsMakeDisabled() check.
As an example, Subversion plugin behaves in such way (JENKINS-24341)
Signed-off-by: NOleg Nenashev <o.v.nenashev@gmail.com>