How to write a small graceful gcd function?

[lovecreatesbeauty]

My small function works, but I have some questions. And I want to
listen to you on How it is implemented?

1. The function does not check if parameter x is larger or smaller than
parameter y.

2. Is it better to use unsigned int or unsigned long as the type of
parameters x and y? This change may remove the if statement.

More comments are still welcome.

/*The Greatest Common Divisor, GCD for short, of two positive integers
can be computed with Euclid's division algorithm. Let the given numbers
be a and b, a >= b. Euclid's division algorithm has the following
steps:
1. Compute the remainder c of dividing a by b.
2. If the remainder c is zero, b is the greatest common divisor.
3. If c is not zero, replace a with b and b with the remainder c. Go
back to step (1).

http://www.cs.mtu.edu/~shene/COURSES...hap04/gcd.html */

/*computes the GCD (Greatest Common Divisor) of positive integers x and
y with Euclid's algorithm. return the GCD, or -1 for failure. - jhl,
Jul 2006*/

int gcd(int x, int y){
if (x 0 && y 0){
while (x % y){
y = x + y;
x = y - x;
y = (y - x) % x;
}
} else {
y = -1; /*input invalid*/
}
return y;
}

lovecreatesbeau ty

[Don]

Keith Thompson wrote:

"Julian V. Noble" <jv*@virginia.e duwrites:
[...]

int gcd( int m, int n ) // recursive
{
if(n==0) { return m; }
else
{ int answer = gcd(n,m%n);
return answer; }
}

The temporary is unnecessary, and the formatting is just odd. Here's
how I'd write it:

int gcd(int m, int n)
{
if (n == 0) {
return m;
}
else {
return gcd(n, m % n);
}
}

int gcd(int m, int n)
{
return n ? gcd(n, m%n) : m;
}

Don

[Keith Thompson]

"Don" <do*******@gmai l.comwrites:

Keith Thompson wrote:

>"Julian V. Noble" <jv*@virginia.e duwrites:
[...]

int gcd( int m, int n ) // recursive
{
if(n==0) { return m; }
else
{ int answer = gcd(n,m%n);
return answer; }
}

The temporary is unnecessary, and the formatting is just odd. Here's
how I'd write it:

int gcd(int m, int n)
{
if (n == 0) {
return m;
}
else {
return gcd(n, m % n);
}
}

int gcd(int m, int n)
{
return n ? gcd(n, m%n) : m;
}

Sure, that's also valid, but I prefer my version. Personally, I
dislike treating an integer as a condition rather than explicitly
comparing it to 0 *unless* it's actually being used as a boolean. I
also tend to prefer if/else to ?: in most cases. For something this
small, it's not a big deal, but it can quickly grow into IOCCC
material.

<OT>
To the person who e-mailed me regarding this thread: anonymous e-mails
are unwelcome and will be ignored. If you want to comment on
something I've written here, do so here. If you must contact me by
e-mail, have the courtesy to identify yourself.
</OT>

--
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.

[lovecreatesbeauty]

Keith Thompson wrote:

"Don" <do*******@gmai l.comwrites:

int gcd(int m, int n)
{
return n ? gcd(n, m%n) : m;
}

Sure, that's also valid, but I prefer my version. Personally, I
dislike treating an integer as a condition rather than explicitly
comparing it to 0 *unless* it's actually being used as a boolean.

I remember that you said you dislike the _Bool keyword and bool, true,
false macros in a post before.

And, in K&R2 sec3.2, it says that the shortcut is preferred when the
expression tests its value even the expression is a numeric value. It
says, ` the most obvious is writing if (expression) instead of if
(expression ! = 0) '.

In sec5.5, it says, a comparison against 0 ('\0' equals 0, right?) is
redundant. And it writes the final abbreviation version of strcpy
containing this single statement: while (*s++ = *t++) ;

[Keith Thompson]

"lovecreatesbea uty" <lo************ ***@gmail.comwr ites:

Keith Thompson wrote:

>"Don" <do*******@gmai l.comwrites:

int gcd(int m, int n)
{
return n ? gcd(n, m%n) : m;
}

Sure, that's also valid, but I prefer my version. Personally, I
dislike treating an integer as a condition rather than explicitly
comparing it to 0 *unless* it's actually being used as a boolean.

I remember that you said you dislike the _Bool keyword and bool, true,
false macros in a post before.

No, I don't believe I did. In fact, I wish that bool, true, and false
had been in the language from the very beginning <OT>as they were in
C++</OT>. Of course we can't yet assume that they're supported by all
compilers -- but it's easy enough to define them yourself if you need
them. (I find the keyword _Bool a bit ugly, but I understand the need
to avoid stepping on existing code.)

And, in K&R2 sec3.2, it says that the shortcut is preferred when the
expression tests its value even the expression is a numeric value. It
says, ` the most obvious is writing if (expression) instead of if
(expression ! = 0) '.

Ok, so I disagree with K&R2 on this point. I don't have my copy of
K&R2 handy; I'll look up the exact wording later, and perhaps comment
further.

In sec5.5, it says, a comparison against 0 ('\0' equals 0, right?) is
redundant. And it writes the final abbreviation version of strcpy
containing this single statement: while (*s++ = *t++) ;

That particular form is idiomatic enough that it doesn't bother me too
much, but in general I prefer to be more explicit, even at the expense
of a few extra keystrokes. I understand that lots of very smart
people don't share this opinion, and we all need to be able to read
code in a variety of styles even if we're more restrictive in what we
write.

--
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.

[ozbear]

On Sun, 23 Jul 2006 06:47:12 GMT, Keith Thompson <ks***@mib.or g>
wrote:

>"lovecreatesbe auty" <lo************ ***@gmail.comwr ites:

>Keith Thompson wrote:

>>"Don" <do*******@gmai l.comwrites:
int gcd(int m, int n)
{
return n ? gcd(n, m%n) : m;
}
Sure, that's also valid, but I prefer my version. Personally, I
dislike treating an integer as a condition rather than explicitly
comparing it to 0 *unless* it's actually being used as a boolean.

I remember that you said you dislike the _Bool keyword and bool, true,
false macros in a post before.

No, I don't believe I did. In fact, I wish that bool, true, and false
had been in the language from the very beginning <OT>as they were in
C++</OT>. Of course we can't yet assume that they're supported by all

<snip>

There is no such language as C++< / OT>

Haven't you been reading the rants about this sort of joining?

:O)

Oz
--
A: Because it fouls the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?