Re: [aspectwerkz-dev] : new version

the_mindstorm <[email protected]> Thu, 07 Oct 2004 20:08:55 +0300
Newsgroups gmane.comp.java.aspectwerkz.devel
Message-ID <[email protected]>
Quote [Alexandre Vasseur]:

> it does not make sense to have it.
> load time weaving is use to start up an app server and stuff like that.
> 
> well, could be limited to something like the ant <java task, but then
> with Java 5 or JRockit you don't need to write a task at all, you just
> add a VM option.
> 

absolutely (i hadn't any intention in creating custom tasks for all types of load time weaving - it 
would make no sense). the whole thinking about load time weaving was about development. I always use 
the ant builds to start my applications during development - this way I assure myself I work on the 
latest modifications. am i completely wrong?

./the_mindstorm

> Alex
> 
> 
> On Thu, 07 Oct 2004 19:46:28 +0300, the_mindstorm
> <[email protected]> wrote:
>> Quote [Jonas Bonér]:
>> > Hey.
>> >
>> > On Thu, 07 Oct 2004 16:59:52 +0300, the_mindstorm
>> > <[email protected]> wrote:
>> >> Quote [Jonas Bonér]:
>> >> > Hi.
>> >> >
>> >> > thanks a lot for your contributions, I like it. committed to head.
>> >>
>> >> 10x
>> >>
>> >> > I was just wondering if it would be more intuitive to use 'classpath'
>> >> > instead of 'cls' or 'clsdir' and to use 'srcdir' and not 'src'.
>> >>
>> >> +1 for classpath.
>> >> As for src vs srcdir I have used the notations according to Ant provided tasks (see javac). I think
>> >> for the moment we can keep the attribute name srcdir and inner element src.
>> >>
>> >
>> > Fine with me.
>> >
>> >> >
>> >> > Ps. We are in high need of an ant task for the offline compiler
>> >> > (AspectWerkzC) as well,
>> 
>> Why are you talking only about offline mode? This task cannot be used also for classload weaving?
>> 
>> ./the_mindstorm
>>
>