Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!cs.utexas.edu!uunet!kddlab!titcca!sragwa!wsgw!socslgw!diamond!diamond
From: diamond@diamond.csl.sony.junet (Norman Diamond)
Newsgroups: comp.lang.c++
Subject: Re: ?-tion on sizeof classes
Keywords: sizeof structure boundaries
Message-ID: <10391@socslgw.csl.sony.JUNET>
Date: 19 Jun 89 06:11:27 GMT
References: <2425@blake.acs.washington.edu>
Sender: news@csl.sony.JUNET
Reply-To: diamond@csl.sony.junet (Norman Diamond)
Organization: Sony Computer Science Laboratory Inc., Tokyo, Japan
Lines: 29

In article <2425@blake.acs.washington.edu> keffer@ocean.washington.edu writes:

>Aside from ..., is there any other reason why the size of an
>object might not be the sum of its members?

There can be hidden information, sometimes necessary for current
features (e.g. vtable) or maybe for eventual assistance from friendly
implementors (garbage collection, debugging, etc.).  If you want (need)
to know the size of a class, use "sizeof".  That is the entire purpose
of "sizeof".

>The reason I ask is that it would sure be nice if I could assume
>	sizeof(Complex) == 2*sizeof(double)
>because this is what a LOT of FORTRAN routines assume.  FFT
>routines, for example, routinely treat an N point complex vector as
>a 2N real vector.  

A lot of FORTRAN routines make assumptions about C++ implementations?
I'll bet C++ wasn't even a gleam in BS's eye when most of those FORTRAN
programs were written.  I'll bet that those FORTRAN programs conformed
to the FORTRAN standard of their day though.  When you write C++
programs, conform to C++ implementations and try to guess what
existing practices (:-I irony) will eventually be standardized.

--
Norman Diamond, Sony Computer Science Lab (diamond%csl.sony.jp@relay.cs.net)
 The above opinions are claimed by your machine's init process (pid 1), after
 being disowned and orphaned.  However, if you see this at Waterloo, Stanford,
 or Anterior, then their administrators must have approved of these opinions.