Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!tut.cis.ohio-state.edu!gem.mps.ohio-state.edu!apple!agate!garnet.berkeley.edu!fjlim
From: fjlim@garnet.berkeley.edu
Newsgroups: comp.sys.mac.programmer
Subject: Re: OOPS/TCL question
Keywords: Beginner,Questions,OOPS,TCL
Message-ID: <1989Oct25.052253.15855@agate.berkeley.edu>
Date: 25 Oct 89 05:22:53 GMT
References: <2849@murtoa.cs.mu.oz.au>
Sender: usenet@agate.berkeley.edu (USENET Administrator;;;;ZU44)
Organization: University of California, Berkeley
Lines: 29

In article <2849@murtoa.cs.mu.oz.au> rob@cs.mu.oz.au (Robert Wallen) writes:
>Am I missing something or is the 'A::' not necessary in the call to x() because
>B has inherited it and thus 'this->x()' will do?  Therefore 'x()' will do too.
>
>struct A : direct { void x(void); };
>struct B : A { void y(void); };
>...
>B::y(void)
>{
>	A::x();
>}
>
>I know its not necessary because I tried compiling it and it works fine.
>Why does the Think Class Library seem to specify these things when it doesnt
>need to??

I believe that the only places where such unnecessary specifications occur
in the TCL are in initialization methods. For example,
	CBureaucrat::IBureaucrat(aSupervisor);
appears in the CDirector::IDirector() method.

CDirector is a subclass of CBureaucrat, so the IBureaucrat() method is
inherited and it is unnecessary to explicitly specify the class name.
I included the specification as a self-documenting feature which reinforces
the concept that the superclass initialization method is being called.
It's a matter of personal style, similar to the choice of whether or not
to prefix instance variables and methods with "this->".

Greg Dow, author of the TCL