Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [aspectj-dev] AspectJ Decompling

I understand you completely, this was just the
information I was looking for. 

In context, I am attempting to justify a development
that effectively parses an application and attempts to
pull out join points (at this stage only 'call' type,
although I realise this is not black and white) and
rebuild the aspect.aj file. This is carried out as
part of a dissertation that is concerned with AOP. 

Thank you for your time, it is most appreciated.

wag81


--- Andy Clement <andrew.clement@xxxxxxxxx> wrote:

> On 07/08/06, wag81uk <wag81uk@xxxxxxxxxxx> wrote:
> > Am I correct in believing that it is not possible
> to determine what join
> > points are defined by a pointcut from the
> Aspect.class file ?
> 
> For a given build of a system (aspects+classes) you
> can create a
> symbol/relationships file that contains the
> information about which
> join points were affected by advice (and from the
> advice determine the
> pointcut) - we don't make it easy to parse this
> symbol file at the
> moment.  A form of this relationship structure is
> used by AJDT to show
> crosscutting in the UI.
> 
> The information on what was matched is not currently
> captured in the
> aspect .class file - because you may reuse that
> aspect over and over
> in different configurations and each time it could
> affect other types
> in different ways, we don't want to rewrite the
> aspect .class over and
> over each time when the aspect hasn't actually
> changed.
> 
> > Rather, that in order to achieve this it would
> effectively be necessary to
> > trawl thought the Aspect.class and Java.class
> files matching join points to
> > their pointcuts?
> 
> We have thought about making it easier to identify
> in a particular
> class how it was affected by aspects, that was an
> intended capability
> of the tool I mentioned in my previous post - it
> would tell you how
> code was affected without you having to interpret
> strange bytecode
> sequences in the javap output.
> _______________________________________________
> aspectj-dev mailing list
> aspectj-dev@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/aspectj-dev
> 



		
___________________________________________________________ 
Yahoo! Messenger - with free PC-PC calling and photo sharing. http://uk.messenger.yahoo.com


Back to the top