Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ames!ncar!unmvax!tut.cis.ohio-state.edu!rutgers!att!westmark!mole-end!mat
From: mat@mole-end.UUCP (Mark A Terribile)
Newsgroups: comp.lang.c++
Subject: Re: cfront feature or bug?
Summary: Don't forget ?:
Message-ID: <175@mole-end.UUCP>
Date: 7 May 89 08:29:10 GMT
References: <NEATH.89May4105353@solar-1.stars.flab.Fujitsu.JUNET>
Distribution: comp.lang.c++
Organization: mole-end--private system.  admin: mole-end!newtnews
Lines: 51

> I have just spent an hour trying to find out why my Vector class
> implementation was core dumping.  ...
   
_>         class X {
_>           int i;
_>         public:
_> 	  X() {};
_>           X(int n) { i = n; }
_>           void times_two () {i *= 2;}
_>         };
 
_>         void dummy (int k) {
_>           int i;
_>           if (k > 3)
_> 	    X* t = new X[10];
_>           else
_> 	    X* t = new X[20];
_> 	  t.times_two();
_>         }
 
> At a first glance, this   code  seems to be  OK.

Eh?  Try

	X* t = ( k > 3 ) ? new X[ 10 ]
			 : new X[ 20 ] ;

> ... An even worse problem ...

_>         void dummy (int k) {
_>           int i;
_>           if (k > 3) {
_>             ::: // some code here
_> 	    X* t = new X[10];		// <== Wrong scope, fershure ...
_>             ::: // more code here
_>           }
_>           else
_> 	    X* t = new X[20];
_> 	  t.times_two();
_>         }
 
> My questions are: 1) Is this a bug in cfront?
>                   2) If it is a bug, is it fixed in 2.0?

More likely, it's a weakness in the language.  This is one of those places
where you really ought to declare the variable and initialize it later, I
think.
-- 

(This man's opinions are his own.)
From mole-end				Mark Terribile