Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!thunder.mcrcim.mcgill.edu!mouse
From: mouse@thunder.mcrcim.mcgill.edu (der Mouse)
Newsgroups: comp.lang.c
Subject: Re: complex data manipulation
Message-ID: <1991Jun1.213156.28507@thunder.mcrcim.mcgill.edu>
Date: 1 Jun 91 21:31:56 GMT
References: <1991May29.211117.29303@mailer.cc.fsu.edu>
Organization: McGill Research Centre for Intelligent Machines
Lines: 34

In article <1991May29.211117.29303@mailer.cc.fsu.edu>, wjbrown@evax0.eng.fsu.edu (WILLIAM J. BROWN) writes:

> I'm writing a program in C (of course) which requires the inversion
> of a rather large matrix.  This by itself is no problem except that
> some of the elements in the array are complex.  Is there an easy way
> to handle complex data in C?

Not especially.  You can define a small struct and then have functions
that operate on these things, but C provides no way to make the
"natural" way of writing things with the usual arithmetic operators
work.  For that you need a language that allows the programmer to
extend operator polymorphism, like C++.

> I have a book titled *C TOOLS for Scientists and Engineers* where the
> author uses the header file #include <complex.h> and uses functions
> such as CADD(),CSUB(),CMULT() etc. for complex arithmetic.

Not hard to do.  Something like

complex.h:
	typedef struct { double r; double i; } complex;
	complex CADD(complex,complex);
[prototypes for other functions omitted]

complex.c:
	complex CADD(complex a, complex b)
	{ complex sum; sum.r = a.r + b.r; sum.i = a.i + b.i; return(sum); }
[code for other functions omitted]

					der Mouse

			old: mcgill-vision!mouse
			new: mouse@larry.mcrcim.mcgill.edu