Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [aspectj-users] declare parents semantics clarification withannotations on interface mixin

I have raised a new bug to track what we need to do to fix things or alter documentation on this - if I don't then it will just get forgotten - like that question you mentioned in your post.

Bug is https://bugs.eclipse.org/bugs/show_bug.cgi?id=243203

I can't answer your question specifically right now, I'd have to spend an hour or so immersing myself in @DeclareParents and remind myself how it behaves, but I'm up to my neck in generic ITDs right now.  If you have suggestions for changes that would improve it, please do append them to that bug report.  I'd appreciate any help I can get to improve that area of the code in terms of updates to documentation or improving it by changing a couple of things about how it works.

cheers,
Andy.

2008/8/5 Larry Chu <LChu@xxxxxxx>

To follow up, I found these references in the archive about the same topic.

 

http://dev.eclipse.org/mhonarc/lists/aspectj-users/msg07722.html

http://dev.eclipse.org/mhonarc/lists/aspectj-users/msg05884.html

 

 

I found a question posted here that interested me but was never answered:

 

http://dev.eclipse.org/mhonarc/lists/aspectj-users/msg05884.html

 

> The short question is:
> When using @DeclareParents to define a mixin, how do I gain access to
> the instrumented object from within the mixin object?

 

I was hoping someone might be able to answer this previously asked question.

 

Regards,

Larry

 


From: Andy Clement [mailto:andrew.clement@xxxxxxxxx]
Sent: Saturday, July 26, 2008 8:34 PM
To: aspectj-users@xxxxxxxxxxx
Subject: Re: [aspectj-users] RE: declare parents semantics clarification withannotations on interface mixin

 

2008/7/25 Larry Chu <LChu@xxxxxxx>


This excerpt from the AspectJ 5 Developer's Notebook, "It's very
important to preserve the same semantics between the code style and the
annotation style," which is implying that the semantics are the same is
just misleading, because the semantics are indeed quite different.


I agree that is misleading.  Although we want to have identical semantics there are places where it is just not possible and mixins were added as a kind of best-can-do for AspectJ intertype declarations.  There may still be places where mixins can be tweaked to bring them more inline with what is possible for ITDs (they haven't had a lot of work done on them) but at the very least the doc needs updating to detail the difference in semantics in this area, I just haven't had the time.

cheers,

Andy.


_______________________________________________
aspectj-users mailing list
aspectj-users@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/aspectj-users



Back to the top