Newsgroups: comp.lang.pascal
Path: utzoo!utgpu!watserv1!maytag!watstat.waterloo.edu!dmurdoch
From: dmurdoch@watstat.waterloo.edu (Duncan Murdoch)
Subject: Re: TP5.5: Can't store method in procedure variable
Message-ID: <1990Dec10.135543.13134@maytag.waterloo.edu>
Keywords: Method Object Procedure TP5.5
Sender: daemon@maytag.waterloo.edu (Admin)
Organization: University of Waterloo
References: <1990Dec3.061013.22258@ux1.cso.uiuc.edu> <1990Dec4.232438.927@ux1.cso.uiuc.edu> <1990Dec6.143527.21330@watserv1.waterloo.edu> <1990Dec7.153019.23716@csn.org>
Distribution: na
Date: Mon, 10 Dec 90 13:55:43 GMT
Lines: 22

In article <1990Dec7.153019.23716@csn.org> streich@tigger.Colorado.EDU (Mark Streich) writes:
>
>This looks quite suspicious because you're taking the address of a class,
>and not the address of an *instance* of the class.  Two instances may 
>(obviously?) have different virtual methods.  

It wasn't the address of the class, rather tthe address of the entry
point of a method defined 
within the class.  (Please excuse my bouncy keyboardd!  Nortthgatte has 
promised me a replacement any day now; when it arrives, I should be able to
type again.)  But your commment aboutt virtual methods is right on:  there's
no easy way to find the appropriate virtual method for an instance, other than
just calling it.  

In fact, it says somewhere in the manual that the "@" 
operattor applied to routines is useless as long as you stay within TP; it's
just provided for completeness, in case you want to pass the address to 
some assembly code.


Duncan Murdoch