From: Josh Cheek <josh.cheek@...>
Date: 2010-05-15T11:01:16+09:00
Subject: Re: Range on strings.

--00163683351ec00dd00486985db3
Content-Type: text/plain; charset=ISO-8859-1

On Fri, May 14, 2010 at 4:40 AM, Vikrant Chaudhary <nasa42@gmail.com> wrote:

> Hi,
> If I do -
>
> ('A'..'Z').include?('AA')
>
> It returns "true", while
>
> ('A'..'Z').to_a.include?('AA')
>
> (of course) returns "false". Is it intentional or possibly a bug?
> I'm using ruby 1.8.7 (2010-01-10 patchlevel 249) [x86_64-linux] on
> Ubuntu 10.04 x64
>
>
I think the problem is that on 1.8 it checks include by seeing if the value
is greater than or equal to the beginning value, and less then or equal to
the end value, since string comparison would have "A" < "AA" < "Z", it
returns true.

(http://svn.ruby-lang.org/repos/ruby/branches/ruby_1_8/range.c)

/*
 *  call-seq:
 *     rng === obj       =>  true or false
 *     rng.member?(val)  =>  true or false
 *     rng.include?(val) =>  true or false
 *
 *  Returns <code>true</code> if <i>obj</i> is an element of
 *  <i>rng</i>, <code>false</code> otherwise. Conveniently,
 *  <code>===</code> is the comparison operator used by
 *  <code>case</code> statements.
 *
 *     case 79
 *     when 1..50   then   print "low\n"
 *     when 51..75  then   print "medium\n"
 *     when 76..100 then   print "high\n"
 *     end
 *
 *  <em>produces:</em>
 *
 *     high
 */

static VALUE
range_include(range, val)
    VALUE range, val;
{
    VALUE beg, end;

    beg = rb_ivar_get(range, id_beg);
    end = rb_ivar_get(range, id_end);
    if (r_le(beg, val)) {
	if (EXCL(range)) {
	    if (r_lt(val, end)) return Qtrue;
	}
	else {
	    if (r_le(val, end)) return Qtrue;
	}
    }
    return Qfalse;
}





However, when it converts to an array, that is a method from Enumerable. I'm
not initiated enough to figure out how it's C code works, but my
understanding is that it uses #succ (I got that from this thread
http://www.ruby-forum.com/topic/192758#new, but please correct me if I
misunderstood) to iterate from the beginning element to the end element, to
create the array. Here is it's code, if you understand it, feel free to
explain.
(http://svn.ruby-lang.org/repos/ruby/branches/ruby_1_8/enum.c)

/*
 *  call-seq:
 *     enum.to_a      =>    array
 *     enum.entries   =>    array
 *
 *  Returns an array containing the items in <i>enum</i>.
 *
 *     (1..7).to_a                       #=> [1, 2, 3, 4, 5, 6, 7]
 *     { 'a'=>1, 'b'=>2, 'c'=>3 }.to_a   #=> [["a", 1], ["b", 2], ["c", 3]]
 */
static VALUE
enum_to_a(argc, argv, obj)
    int argc;
    VALUE *argv;
    VALUE obj;
{
    VALUE ary = rb_ary_new();

    rb_block_call(obj, id_each, argc, argv, collect_all, ary);

    return ary;
}





Then Array#include iterates through each of it's elements to check if it
contains the desired element
(http://svn.ruby-lang.org/repos/ruby/branches/ruby_1_8/array.c)

/*
 *  call-seq:
 *     array.include?(obj)   -> true or false
 *
 *  Returns <code>true</code> if the given object is present in
 *  <i>self</i> (that is, if any object <code>==</code> <i>anObject</i>),
 *  <code>false</code> otherwise.
 *
 *     a = [ "a", "b", "c" ]
 *     a.include?("b")   #=> true
 *     a.include?("z")   #=> false
 */

VALUE
rb_ary_includes(ary, item)
    VALUE ary;
    VALUE item;
{
    long i;

    for (i=0; i<RARRAY(ary)->len; i++) {
	if (rb_equal(RARRAY(ary)->ptr[i], item)) {
	    return Qtrue;
	}
    }
    return Qfalse;
}





The elements the array contains are
$ ruby -e "p ('A'..'Z').to_a"
["A", "B", "C", "D", "E", "F", "G", "H", "I", "J", "K", "L", "M", "N", "O",
"P", "Q", "R", "S", "T", "U", "V", "W", "X", "Y", "Z"]
So "AA" is not found, and thus it returns false.




I honestly don't know about 1.9, I tried looking at it's code, but I don't
understand it.
(http://svn.ruby-lang.org/repos/ruby/branches/ruby_1_9_1/range.c)

--00163683351ec00dd00486985db3--