It is possible your second piece of advice there "call(*.new(..))" may
be matching the call to create the aspect instance - and that could
lead to a NoAspectBoundException. To investigate constructors I'd
maybe use execution.
after(): execution(*.new(..)) && !within(TheAspectType) {
System.out.println("*2* ctor: " +
thisJoinPointStaticPart.getSignature());
}
That should tell you which ctors are running.
I can't immediately see what is wrong with your initial piece of advice though.
Andy
On 25 March 2011 11:33, Grey, Lee <lee.g...@ncr.com> wrote:
> Andy,
>
> I took your advice (#2) below, and it's been working great in development for
> a month. However, it turns out that there's something different in
> production (not sure what) that's preventing my harmless weaving from
> triggering, so my aspect is never instantiated.
>
> So, I'm trying to add something to the aspect to dump out all constructors,
> so I can choose something else as my eager-weaving trigger. But, so far, I
> either get no output, or I get org.aspectj.lang.NoAspectBoundException.
>
> I don't know what class I want to trigger on, but I am 99% certain it's in a
> package somewhere beneath com.sonicsw.*. Can you see what's wrong with this?
>
> pointcut init(Object o):
> this(o) &&
> within(com.sonicsw..*) &&
> initialization(*.new());
>
> after(Object o) returning: init(o) {
> System.out.println("*1* ctor: " +
> thisJoinPointStaticPart.getSignature());
> }
>
> after() returning(Object o): call(*.new(..)) {
> System.out.println("*2* ctor: " +
> thisJoinPointStaticPart.getSignature());
> }
>
> Thanks,
> Lee
>
>
> -----Original Message-----
> From: aspectj-users-boun...@eclipse.org
> [mailto:aspectj-users-boun...@eclipse.org] On Behalf Of Andy Clement
> Sent: Tuesday, February 22, 2011 12:22 PM
> To: aspectj-users@eclipse.org
> Subject: Re: [aspectj-users] Lazy weaving vs eager weaving?
>
> Hi,
>
> There is no direct control for this. You kind of have two options:
>
> (1) touch the aspect yourself much earlier, to cause it to be loaded up.
> Just refer to something in it from one of your earlier loaded classes that
> will cause it to be loaded.
> (2) Use it to 'harmlessly' weave something that is loaded much earlier, which
> kind of causes (1) to happen.
>
> cheers
> Andy
>
> On 22 February 2011 08:57, Grey, Lee <lee.g...@ncr.com> wrote:
>> Is there any way to control when an aspect gets loaded and initialized?
>>
>> Thanks,
>> Lee
>> ________________________________
>> From: aspectj-users-boun...@eclipse.org
>> [mailto:aspectj-users-boun...@eclipse.org] On Behalf Of Grey, Lee
>> Sent: Monday, February 21, 2011 12:58 PM
>> To: aspectj-users@eclipse.org
>> Subject: [aspectj-users] Lazy weaving vs eager weaving?
>>
>> I'm still working on the weaving of Sonic ESB container services, and
>> it's working perfectly with one exception.
>>
>> I'm instantiating a thread object in a static block of my aspect.
>> Unfortunately, the static block doesn't execute until the first time a
>> message is received, even though the container JVM had started long ago.
>> Aside from the fact that the thread doesn't start when it should,
>> there is also a very long hit for all the class loading and weaving on
>> receipt of that first message.
>>
>> Is there any way to have the weaving occur immediately, which I'm
>> assuming would be the way to have the static block execute prior to
>> receiving a message?
>>
>> Thanks,
>> Lee Grey
>>
>> _______________________________________________
>> aspectj-users mailing list
>> aspectj-users@eclipse.org
>> https://dev.eclipse.org/mailman/listinfo/aspectj-users
>>
>>
> _______________________________________________
> aspectj-users mailing list
> aspectj-users@eclipse.org
> https://dev.eclipse.org/mailman/listinfo/aspectj-users
> _______________________________________________
> aspectj-users mailing list
> aspectj-users@eclipse.org
> https://dev.eclipse.org/mailman/listinfo/aspectj-users
>
_______________________________________________
aspectj-users mailing list
aspectj-users@eclipse.org
https://dev.eclipse.org/mailman/listinfo/aspectj-users
