So what would be the problem with having a check in execute to see if p1 is
null? �If so, call init. �Possible problems? �We only need p1 to exist
before execute so the createProperty method works. �Same check in
createProperty? �Lots of seemingly redundant calls, but it does get us what
we need - on demand use of a project.
Glenn McAllister
TID - Software Developer - VisualAge for Java
IBM Toronto Lab, (416) 448-3805
"An approximate answer to the right question is better than the
right answer to the wrong question." - John W. Tukey
Please respond to [EMAIL PROTECTED]
To: � � � [EMAIL PROTECTED]
cc:
Subject: � � � �Re: [PATCH] Less memory consumption for subbuilds
>>>>> "gm" == glennm �<[EMAIL PROTECTED]> writes:
�gm> (1) sounds good to me, but how do we do it? I didn't think there
gm> was an easy way to tell if there was a reference to a class
gm> instance. �Mind you, I'm happy to be wrong. :-)
I wasn't talking about a Java language reference but rather about
Project.getReferences. Should have been clearer.
On second thought it doesn't work. Each task can be executed multiple
times via
((Target) project.getTargets().get("nameOfSurroundingTarget")).execute()
so we can never be sure a task won't be called a second time. Each
target can be addressed that way.
Stefan
