Newsgroups: comp.os.msdos.djgpp,comp.lang.c++
Subject: Re: 2d Matrix class with conventional c/c++ element access semantics
From: frenchc AT cadvision DOT com (Calvin French)
References: <34C36C45 DOT 24966424 AT pentek DOT com>
MIME-Version: 1.0
Content-Type: Text/Plain; charset=US-ASCII
NNTP-Posting-Host: ts7ip236.cadvision.com
Message-ID: <34c3f1ff.0@news.cadvision.com>
Date: 20 Jan 98 00:38:23 GMT
Organization: CADVision Development Corporation (http://www.cadvision.com/)
Lines: 23
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Precedence: bulk

The reason your code is crashing is probably that the vector<T>'s inside 
vector< vector<T> > aren't getting initialized (but I don't know). If I were 
you, i would do it much differently. Realize that by putting a vector inside 
a vector you are allowing functionality which isn't really normal for a 
matrix. For instance, to have one row longer than another. If I were you, I 
would do it this way:

template< class T > 
class matrix
{
   vector<T> v; // internal data storage.
   unsigned int w, h;    // dimensions of matrix.
public:
   matrix( unsigned int W, unsigned int H ) : w(W), h(H), v(W*H) {}; 
	// hoping this will init v correctly :)

   T& operator( unsigned int X, unsigned int Y ) { return v[ Y*w + X ]; };
};

That ought to give you the general idea. Just hope it works :)

- Calvin -