WITHIN was meant to replace BETWEEN but it hasn't always
been a comfortable fit.

A characteristic of WITHIN is that the upper limit is not included
in the range. This can lead to obfuscated code e.g.

CHAR a CHAR z 1+ WITHIN

Thus the functionality of BETWEEN is still being used, if not
the word itself. When defined today, it is usually as:

: BETWEEN ( n1|u1 n2|u2 n3|u3 -- flag ) 1+ WITHIN ;

But it is not without problem e.g. it fails when u1 = u3 = MAX-U .
Perhaps not so significant for 32-bit systems as for 16-bit, it is
nevertheless an inconsistency which can catch a programmer
unawares.

Playing with the algorithm for WITHIN (A.6.2.2440) one gets:

: BETWEEN ( n1|u1 n2|u2 n3|u3 -- flag ) OVER - -ROT - U< 0= ;

It seems to work. If anyone finds a problem or knows something
better, let us know. In the meantime, here are several code versions.

Ed wrote:

The other way around. WITHIN came first. It may be as nice, but its code
is so snazzy that I find it more esthetically pleasing.

Clearly, WITHIN came first in this case.

It is the comparison operator in the definition that "fails".

"Jerry Avins" ieee.org> wrote in message news:rIadnSizg9q1YUjanZ2dnUVZ_gWdnZ2d@rcn.net...

In the version I show, the algorithms are very similar. WITHIN uses
U< whereas BETWEEN uses U<= .

AFAIK, BETWEEN can't be made from WITHIN and vice-versa so
it suggests they're equally primitive. Both seem to have their
applications.

Op Wed, 12 Mar 2008 10:59:55 +1100 schreef Ed:

Of course:

: BETWEEN 1+ WITHIN ;
: WITHIN 1- BETWEEN ;

On Mar 11, 9:02 pm, Coos Haak wrote:

Ed wrote:

?? You show below how to do it with 1+ .

Ed wrote:

WITHIN came first *historically*. It handles its arguments as it does
to be consistent with the fact that 0 DO ... LOOP will run n-1 times
(i.e. exclusive at the top end), which made that behavior easy to teach
and remember.

On Mar 11, 10:57 pm, Elizabeth D. Rather forth.com> wrote:

An off-by-one error. It will run n times.

On Mar 10, 5:05 pm, "Ed" invalid.com> wrote:

Stuff like this leads one to believe that Forth code is
unreadable, cryptic, unsharable, and unmaintainable.
It's worse than some assembly language I've seen.
Forth needs to become a higher-level language.

William James yahoo.com> wrote:

What are your standards for readability?

I see six commands, and I understand every one of them. No control
structures. It doesn't get much clearer than that.

The name tells me what it's supposed to do, surrounded by the comments
in the surrounding topic.

What's missing is the test cases. Give us test cases with say the Hayes
convention, and it would be perfect.

n3-n2 n1-n2 U< 0=