Xref: news-dnh.mv.net comp.os.msdos.djgpp:1173
Path: news-dnh.mv.net!mv!news.sprintlink.net!hookup!news.kei.com!news.mathworks.com!gatech!news.uoregon.edu!news.rediris.es!diable.upc.es!diable!frankie
From: frankie AT si DOT upc DOT es (Compte per el frankie)
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Help with classes!
Date: 26 Jul 1995 11:52:23 GMT
Organization: Serveis Informatics -- Universitat Politecnica de Catalunya
Lines: 45
Distribution: world
Reply-To: frankie AT si DOT upc DOT es
Nntp-Posting-Host: diable.upc.es
To: djgpp AT sun DOT soe DOT clarkson DOT edu
Dj-Gateway: from newsgroup comp.os.msdos.djgpp

In article 100000 AT athene, Ian Tester <94024831 AT athene DOT mit DOT csu DOT edu DOT au> () writes:
> hi,
> 	Sorry about this semi-djgpp content, but the c++ newsgroup is 
> choka full of over 1500 messages, and I thought that my problem might be 
> specific to djgpp.
> 	Anyway, I'm writting this 3D math library with classes for 
> vectors, quaternions, and matrices. The vectors and quaternions work OK, 
> but the matrix class doesn't. I'm pretty sure it's to do with the fact 
> that the vec. and quat. classes don't have to allocate memory in their 
> constructors. You see, in my matrix class, I have two ints which hold the 
> number of rows and columns, and a 'float *' which will hold the array 
> of elements. But to use it I have a constructor which takes the number of 
> rows and columns to initialize it to, sets the two ints and allocates the 
> memory with 'elements = new float[rows * cols]'. Is this right?

Yeah, this allocates memory for such a quantity of floats

> 	Anyway, in a little test program, I declare a matrix with 
> 'Matrix M;', and then latter initialize it with 'M = Matrix(3,3);'.
> Is this right? Should I declare it as a pointer ?
> 
> i.e 'Matrix *M;' and
>     'M = new Matrix(3,3);'

New returns a pointer for the new object it allocates memory for, so M gotta
be a pointer to Matrix, so declare it 
	Matrix *M;


> 	In the test program, it doesn't seem to set the two ints with the 
> new size, so all other multiplications and stuff fail because its the 
> wrong size.

Are you sure it doesn't ? And what's the answer of sizeof(elements) after
the new command has been done ? 
 
> 	 I would really like some help with this one. puh-lease!

I'm doing a try.

____________________________________________________________________________
Francesc Guasch Ortiz			Universitat Politecnica de Catalunya
frankie AT si DOT upc DOT es					 Serveis Informatics
http://www.upc.es/~frankie