two-dimensional array

[my.correo.basura]

Hi,
I'm having some problems working with two-dimensional arrays. Here's
the story.
I got two 50x50 matrixes (is that the plural of matrix?) defined at
compile time.
I want to work with them inside a for loop, but depending on the
iteration, one is the real one and the other is dummy. At the end of
the iteration I swap them.
I'm working with pointers but it is'nt working as assignments in lines
15 and 16 are from different types. I'm pretty sure I've been through
this before but can't remember how I solved it. Any help? Here's the
code I've got:
-----------------
#include <stdio.h>

#define ROWS 50
#define COLS 50
#define MAXITERS 100

int main(void){
int matrix1[ROWS][COLS];
int matrix2[ROWS][COLS];
int **realMatrix;
int **dummyMatrix;
int **swapMatrix;
int i;

realMatrix=matr ix1; /*Line 15*/
dummyMatrix=mat rix2; /*Line 16*/

for (i=0; i < MAXITERS; i++){
/*Work with Matrix*/
/*...*/
/*Swap Matrix*/
swapMatrix=real Matrix;
realMatrix=dumm yMatrix;
dummyMatrix=swa pMatrix;
}

return 0;
}

[lawrence.jones]

my************* *@gmail.com wrote:

#include <stdio.h>

#define ROWS 50
#define COLS 50
#define MAXITERS 100

int main(void){
int matrix1[ROWS][COLS];
int matrix2[ROWS][COLS];
int **realMatrix;
int **dummyMatrix;
int **swapMatrix;
int i;

realMatrix=matr ix1; /*Line 15*/
dummyMatrix=mat rix2; /*Line 16*/

test.c:15: warning: assignment from incompatible pointer type
test.c:16: warning: assignment from incompatible pointer type

The correct declarations are:

int (*realMatrix)[COLS];
int (*dummyMatrix)[COLS];
int (*swapMatrix)[COLS];

-Larry Jones

Hmm... That might not be politic. -- Calvin

[jhartzell42]

The declaration for the pointers should be:
int (*realMatrix)[COLS];
Remember, although an array is automatically converted to a pointer to
the first element in expressions, it is fundamentally different from a
pointer.
IIRC, this is one of the innovations made to turn the older "B"
programming language into "C".
For example, the declaration "int a[5];" creates room for five
integers, but does not store a pointer to the first integer anywhere in
memory. Therefore, you can't point to an array with a pointer to a
pointer ("int **p; p=&a;"), because a pointer to a pointer must point
to a pointer physically present in memory, but "int a[5];" does not
create a pointer physically present in memory.
Thus, you must use a pointer to an array ("int (*p)[5]; p=&a;"), which
recognizes that you are not pointing to a pointer, but rather to an
array. You are physically pointing to the first element, and when you
dereference it, you get an array, which is automatically converted to a
pointer to the first element. Therefore, ironically (void *)p==(void
*)*p
Note that this does not invalidate "int a[5]; int *p; p=a;", as this
expression derives a pointer to the first element of a automatically,
and assigns it to p.
I hope this wasn't too confusing to you, because I barely understand it
and wish someone more familiar with ANSI C could explain it.
Good luck!
Jimmy Hartzell

[Micah Cowan]

my************* *@gmail.com writes:

Hi,
I'm having some problems working with two-dimensional arrays. Here's
the story.
I got two 50x50 matrixes (is that the plural of matrix?) defined at
compile time.
No, the correct spelling would be matrices.
I want to work with them inside a for loop, but depending on the
iteration, one is the real one and the other is dummy. At the end of
the iteration I swap them.
I'm working with pointers but it is'nt working as assignments in lines
15 and 16 are from different types. I'm pretty sure I've been through
this before but can't remember how I solved it. Any help? Here's the
code I've got:
-----------------
#include <stdio.h>

#define ROWS 50
#define COLS 50
#define MAXITERS 100

int main(void){
int matrix1[ROWS][COLS];
int matrix2[ROWS][COLS];
int **realMatrix;
int **dummyMatrix;
int **swapMatrix;
These pointers are all the wrong type.

matrix1 and matrix2 are both {arrays of /ROWS/ array of /COLS/ ints}.
The type of matrix1[0] is {array of /COLS/ ints}. If the size of an
int is 4 bytes, then the size of matrix1[10] is COLS x 4 == 200 bytes.

realMatrix, etc., are all of type {pointer to pointer to int}.
The type of realMatrix[0] is {pointer to int}, which has whatever size
a pointer to int might have (*very* unlikely to be 200 bytes).
int i;

realMatrix=matr ix1; /*Line 15*/
dummyMatrix=mat rix2; /*Line 16*/
And here you run into trouble. First of all, no C implementation will
let you do this without at least complaining about it. You'd need to
cast it (but don't). After making these assignments, then trying to
access something using realMatrix[n] or realMatrix[n][m] will give you
undefined behavior. Because realMatrix[n] thinks it's dereferencing a
pointer to a pointer to int, when in reality it's dereferencing a
converted pointer to /array/ of int.

If you declare realMatrix, etc, as:

int (*realMatrix)[COLS]; /* pointer to array of COLS ints. */

things are much more likely to behave the way you expect them to,
although, since you didn't post the actual code that works with these
matrices, I can't know that for certain.

for (i=0; i < MAXITERS; i++){
/*Work with Matrix*/
/*...*/
/*Swap Matrix*/
swapMatrix=real Matrix;
realMatrix=dumm yMatrix;
dummyMatrix=swa pMatrix;
}

return 0;
}

[Keith Thompson]

my************* *@gmail.com writes:

I'm having some problems working with two-dimensional arrays. [snip] int matrix1[ROWS][COLS]; [snip] int **realMatrix; [snip] realMatrix=matr ix1; /*Line 15*/

Arrays are not pointers, and pointers-to-pointers are not useful in
dealing with true 2-dimensional arrays. (Multidimension al arrays can
also be implemented, or perhaps I should say emulated, using pointers
to pointers, but that's not what you're doing here.)

Read section 6 of the comp.lang.c FAQ, <http://www.c-faq.com/>.
(Consider reading the whole thing while you're at it.) If you're
still confused after that, feel free to come back with more questions.

--
Keith Thompson (The_Other_Keit h) ks***@mib.org <http://www.ghoti.net/~kst>
San Diego Supercomputer Center <*> <http://users.sdsc.edu/~kst>
We must do something. This is something. Therefore, we must do this.

[Default User]

Micah Cowan wrote:

my************* *@gmail.com writes:

Hi,
I'm having some problems working with two-dimensional arrays. Here's
the story.
I got two 50x50 matrixes (is that the plural of matrix?) defined at
compile time.

No, the correct spelling would be matrices.

From Merriam-Webster:

Main Entry: ma·trix
Pronunciation: 'mA-triks
Function: noun
Inflected Form(s): plural ma·tri·ces /'mA-tr&-"sEz, 'ma-/; or
ma·trix·es /'mA-trik-s&z/

Brian

[Mark McIntyre]

On 5 Mar 2006 00:24:50 GMT, in comp.lang.c , "Default User"
<de***********@ yahoo.com> wrote:

Micah Cowan wrote:

No, the correct spelling would be matrices.

From Merriam-Webster:

ma·trix·es /'mA-trik-s&z/

We can't be held accountable for mistakes in Merriam-Webster's
dictionaries. I recommend writing to the authors to point it out.

Mark McIntyre
--
"Debugging is twice as hard as writing the code in the first place.
Therefore, if you write the code as cleverly as possible, you are,
by definition, not smart enough to debug it."
--Brian Kernighan

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

[Richard Heathfield]

Mark McIntyre said:

On 5 Mar 2006 00:24:50 GMT, in comp.lang.c , "Default User"
<de***********@ yahoo.com> wrote:

Micah Cowan wrote:

No, the correct spelling would be matrices.

From Merriam-Webster:

ma·trix·es /'mA-trik-s&z/

We can't be held accountable for mistakes in Merriam-Webster's
dictionaries. I recommend writing to the authors to point it out.

You'd better get on to Chambers, too, then. They list -ices and -ixes.

--
Richard Heathfield
"Usenet is a strange place" - dmr 29/7/1999
http://www.cpax.org.uk
email: rjh at above domain (but drop the www, obviously)

[Ben Bacarisse]

On Sun, 05 Mar 2006 11:58:54 +0000, Mark McIntyre wrote:

On 5 Mar 2006 00:24:50 GMT, in comp.lang.c , "Default User"
<de***********@ yahoo.com> wrote:

Micah Cowan wrote:

No, the correct spelling would be matrices.
From Merriam-Webster:
ma·trix·es /'mA-trik-s&z/

We can't be held accountable for mistakes in Merriam-Webster's
dictionaries. I recommend writing to the authors to point it out.

And no hint of irony! If you do write, cc both Collins and the OED
as they have also perpetuated this error (the shorter OED has the audacity
to list the wrong spelling first). ;-)

Mark McIntyre

--
Ben.

[Mark McIntyre]

On Sun, 5 Mar 2006 13:39:15 +0000 (UTC), in comp.lang.c , Richard
Heathfield <in*****@invali d.invalid> wrote:

Mark McIntyre said:

On 5 Mar 2006 00:24:50 GMT, in comp.lang.c , "Default User"
<de***********@ yahoo.com> wrote:

Micah Cowan wrote:

No, the correct spelling would be matrices.

From Merriam-Webster:

ma·trix·es /'mA-trik-s&z/

We can't be held accountable for mistakes in Merriam-Webster's
dictionaries. I recommend writing to the authors to point it out.

You'd better get on to Chambers, too, then. They list -ices and -ixes.

I never said which one was right :-)

(FWIW I prefer the european -ices over the atlantic -ixes, but the
latter is probably acceptable as long as one is consistent).
Mark McIntyre
--
"Debugging is twice as hard as writing the code in the first place.
Therefore, if you write the code as cleverly as possible, you are,
by definition, not smart enough to debug it."
--Brian Kernighan

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----