Subject: RE: HW2 Question
From: Karl Lieberherr (lieber@ccs.neu.edu)
Date: Thu Oct 10 2002 - 10:09:06 EDT
I agree with what Paul said. I also view a call to a static
method like Main.getClassGraph() a global method call and we should
allow those.
What Sergei asks: "should you also detect local calls in static methods"
is an interesting question which we want to implement eventually
but it does not have to be in this assignment.
-- Karl
> -----Original Message-----
> From: Nicole and Paul Freeman [mailto:freeberm@earthlink.net]
> Sent: Thursday, October 10, 2002 8:09 AM
> To: Kojarski Sergei; Dr. Karl J. Lieberherr
> Cc: com3205@ccs.neu.edu
> Subject: Re: HW2 Question
>
>
> Hi Sergei,
> (Karl - Sergei had a question regarding local calls and static methods.
> I was wondering if you would comment on my answer.)
>
> First off, I'm not sure which part of homework 2 this question applies
> to, but if it is part 2 we are not concerned with local calls to static
> methods, only local calls to immediate part objects, i.e. instance
> variables.
>
> If you are asking how static methods are treated in general by the LoD:
> I agree with what you wrote below that a call in method M1 (non static
> method = M) of object O to method M2 of object O should definately be
> treated as a local call. But where static methods (static method = SM)
> are concerned, I think there are a couple of cases:
> call from M1 in object O to SM1 in object O // seems to be ok
> cal from SM1 in object O to M1 in object O // seems to be ok
> call from SM1 in object O to SM2 in object O //again would seem to be
> ok since it is in the same object (or class if using class form)
>
> I think the question rises when we try to determine how we treat calls
> from SM1 in object O1 to calls in SM3 of object O2. I would argue that
> this is acceptable on the basis that static methods could be considered
> "global" methods, as there is no need to instantiate an object to have
> access to them. Since the spirit of the law is that we restrict our
> knowledge of the structure of classes we don't need to directly know
> about, I don't see a violation occurring when we are calling a method
> that is designed to be globally known.
>
> Karl, would you comment on this?
>
> Thanks,
> Paul
>
> Kojarski Sergei wrote:
>
> > Hi Paul,
> >
> > I just want to feel sure about "local call".
> > I believe it is only a call from an
> > object method to the method of the same object and NOT
> > call of a static method from other static method.
> > Is it true?
> >
> > Thanks,
> > Sergei
>
This archive was generated by hypermail 2b28 : Thu Oct 10 2002 - 10:05:58 EDT