Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!gatech!hubcap!beede
From: beede@hubcap.UUCP (Mike Beede)
Newsgroups: comp.lang.c
Subject: Re: Two dimensional arrays in C
Message-ID: <238@hubcap.UUCP>
Date: Mon, 29-Jun-87 19:10:46 EDT
Article-I.D.: hubcap.238
Posted: Mon Jun 29 19:10:46 1987
Date-Received: Wed, 1-Jul-87 01:59:15 EDT
References: <2215@mmintl.UUCP>
Organization: Clemson University, Clemson, SC
Lines: 33

in article <2215@mmintl.UUCP>, franka@mmintl.UUCP (Frank Adams) says:
] 
] In article <733@cod.UUCP> murphys@cod.UUCP (Steven P. Murphy) writes:
]>The following example is what I would like to be able to do. Is it possible ?
]>...
]>matrix_stuff(a, b, array, x, y);
]>
]>int a, b, x, y;
]>double array[x][y];
] 
] No.  Array bounds in C have to be constants.  To do this, you would have to
] do the array arithmetic explicitly.  Sorry.

However, you CAN do the arithmetic explicitly if you like, using macros
to maintain fair readability:  something like

#define matrix_el(array,i,j) (array[y*sizeof(double)+j])

which produces an l-valued expression.  At the same time, you can insert
temporary bounds-checking until you're sure your code is correct (or
forever, whichever is smaller :-) ).  Now, of course, your declaration
for array is
 double array[];
and you can have things like matrix_el(A,i,j) = matrix_el(A,i+1,j) , etc.

Of course, since this depends on the value of y, it is still
sloppy, but you can include y as a discriptor bolted onto the array or
something similar, if you like (which the macro would reference).
-- 
Mike Beede                      UUCP: . . . !gatech!hubcap!beede
Computer Science Dept.          ARPA: BEEDE@TECNET-CLEMSON.ARPA
Clemson University              INET: beede@hubcap.clemson.edu
Clemson SC 29631-1906           PH: (803)656-{2845,3444}