[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
Re: [aspectj-dev] RC1 coming soon?
|
Tons of thanks Alex. I will put some thinking on the @declare part, maybe I can come out with something.
About the declare parents, I think that the next development build will contain it and ajdt too, so
it is enough for me for the moment :-).
./alex
--
.w( the_mindstorm )p.
#: Alexandre Vasseur changed the world a bit at a time by saying on 10/26/2005 2:41 PM :#
Yes we skip declare annotation in @AspectJ. Here is why:
It must be possible to compile @AspectJ with only javac.
Assume we want to add @Foo on some constructor. Assume the @Foo
annotation has its target set to constructor (@Target(...))
To do that we would write something like:
@Aspect class Some {
@Foo(someofmyvalueIwant)
/// here I need a constructor
}
so we could then use a fake Some constructor like in:
@Aspect class Some {
@DeclareAnnotation(....)
@Foo(someofmyvalueIwant)
Some() {}
}
but this is bad as aspect ' constructor is a non sense in the
assymetric AOP field.
instead we could use some nested class like
@Aspect class Some {
static class SomeHost() {
@DeclareAnnotation(....)
@Foo(someofmyvalueIwant)
SomeHost() { }
}
}
but this is really weird to introduce so much new OO structures just
to add annotations in some places.
May be you have ideas for a better design that would still preserve
type safety that Java 5 brings to annotation (ie no
@DeclareAnnotation(annotation="string representation of @Foo you want
to introduce", target=" constructor pattern etc"))
As per the @DeclareParents it did not make it for M4 but it is now in
CVS and thus will be in RC1.
Alex
On 10/26/05, Alexandru Popescu <the.mindstorm.mailinglist@xxxxxxxxx> wrote:
Alex can you please detail a little bit on the aboves:
#: Alexandre Vasseur changed the world a bit at a time by saying on 10/26/2005 9:47 AM :#
>
> Olivier, I hope you noticed we changed some of the details for
> @AspectJ mainly declare annotation for this style will not be
> supported in 1.5.0,
this means that AJ1.5 will not have support through annotation for doing declare @<kind>?
and declare parent is looking slightly different.
according to the doc the syntax @DeclareParents("TypePattern") is supported starting with M4. Is
this wrong or are you saying that it will change till the final version?
thanks in advance,
./alex
--
.w( the_mindstorm )p.
> This will surface in the next published docs, but get it from the CVS
> if you need to fix your book. I am sure you will be willing to wait
> some for up to date and accurate content for your book rather than
> hurry up and ending up with both bugs in the shipment and features
> deltas in your book.
>
> Alex
>
>
> On 10/25/05, Oliver Böhm <boehm@xxxxxxxxxx> wrote:
>> Hello,
>>
>> I guess the plan on http://www.eclipse.org/aspectj/plans.php is a
>> little bit outdated (or I missed something;-) When I extrapolate the
>> dates correct RC1 will come soon and we can count on the final version
>> about midth of November. Is this realistic?
>>
>> I ask because first I want to finalize my book and secondly I manage
>> the AspectJ Winter Camp at the beginning of December (8.-10., see
>> http://www.jugs.org/ajwc/) where I want to use both: the book and
>> AspectJ 5.
>>
>> regards
>> Oliver
>> --
>> Oliver Böhm
>> http://www.javatux.de
>> _______________________________________________
>> aspectj-dev mailing list
>> aspectj-dev@xxxxxxxxxxx
>> https://dev.eclipse.org/mailman/listinfo/aspectj-dev
>>
> _______________________________________________
> aspectj-dev mailing list
> aspectj-dev@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/aspectj-dev
>
_______________________________________________
aspectj-dev mailing list
aspectj-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/aspectj-dev
_______________________________________________
aspectj-dev mailing list
aspectj-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/aspectj-dev