From: brian ford <brixen@...>
Date: 2009-06-11T03:21:39+09:00
Subject: [ruby-core:23789] Re: Ducktyping interface

Hi,

On Sat, Jun 6, 2009 at 5:30 PM, Yehuda Katz<wycats@gmail.com> wrote:
> I should add that further examination reveals that people who make proxy
> objects (Drb, Rails Association Proxy) expect that their respond_to? will
> work correctly. If implementations were allowed to do internal checks only,
> proxies would need to explicitly define all coercion methods, something they
> do not currently do.
> -- Yehuda

To illustrate a bit more why I believe calling #respond_to? is an
implementation detail, consider the following:

$ irb
>> fake = Object.new
=> #<Object:0x5f12fc>
>> fake == 1
=> false
>> fake <=> 1
NoMethodError: undefined method `<=>' for #<Object:0x5f12fc>
        from (irb):3
>> fake.extend Comparable
=> #<Object:0x5f12fc>
>> fake == 1
=> nil
>> def fake.<=>(other)
>> 0 if other == 1
>> end
=> nil
>> fake == 1
=> true
>> fake == 2
=> nil

And this is the MRI C code that performs the magic (compar.c):

static VALUE
cmp_eq(a)
    VALUE *a;
{
    VALUE c = rb_funcall(a[0], cmp, 1, a[1]);

    if (NIL_P(c)) return Qnil;
    if (rb_cmpint(c, a[0], a[1]) == 0) return Qtrue;
    return Qfalse;
}

static VALUE
cmp_failed()
{
    return Qnil;
}

/*
 *  call-seq:
 *     obj == other    => true or false
 *
 *  Compares two objects based on the receiver's <code><=></code>
 *  method, returning true if it returns 0. Also returns true if
 *  _obj_ and _other_ are the same object.
 */

static VALUE
cmp_equal(x, y)
    VALUE x, y;
{
    VALUE a[2];

    if (x == y) return Qtrue;

    a[0] = x; a[1] = y;
    return rb_rescue(cmp_eq, (VALUE)a, cmp_failed, 0);
}

In this case, MRI does not call #respond_to?. And I would argue that a
more valid implementation in general is to dispatch and rescue if
necessary, as this MRI code demonstrates.

Thanks,
Brian

>
> On Sat, Jun 6, 2009 at 5:12 PM, Yehuda Katz <wycats@gmail.com> wrote:
>>
>> Matz,
>> I've been working on adding some specs to RubySpec related to the
>> ducktyping interface. I have observed the following behavior and want to
>> know whether or not it's intentional. Let's use the #to_s coercion as an
>> example:
>> 1. If the object has a method #to_s, it's called.
>> 2. If #to_s returns something that is not a String, a TypeError is
>> raised.
>> In MRI, the mechanism for determining if an object has a method #to_s is
>> to call respond_to?. This means that even if the method does not exist,
>> #to_s will get called, and fall through to #method_missing, if the
>> #respond_to method returns true for :to_s. As a result, when using MRI, it
>> is possible to implement the duck-typing interface without defining #to_s,
>> but instead defining a combination of #respond_to? and #method_missing that
>> returns a String.
>> Is this intentional? Would it be ok for an alternative implementation to
>> look up the #to_s method using an internal check instead of calling the
>> user-defined #respond_to?, as MRI does?
>> In other words, is the only "correct" way to enlist in Ruby coercion to
>> define to_s explicitly, or is another correct way to define respond_to? and
>> method_missing to return an appropriate response?
>> --
>> Yehuda Katz
>> Developer | Engine Yard
>> (ph) 718.877.1325
>
>
>
> --
> Yehuda Katz
> Developer | Engine Yard
> (ph) 718.877.1325
>