From: David Alan Black <dblack@...>
Date: 2001-12-10T23:43:24+09:00
Subject: [ruby-talk:28087] Re: How does puts decide how to print a given object?

Hello --

On Mon, 10 Dec 2001, Yukihiro Matsumoto wrote:

> Hi,
>
> In message "[ruby-talk:28052] How does puts decide how to print a given object?"
>     on 01/12/10, "Harry Ohlsen" <harryo@zip.com.au> writes:
>
> |This got me to thinking how puts does its conversion to a string.  I ask
> |because (a) I'm surprised it doesn't just use to_s and because someone
> |reading the article is bound to ask :-).
>
> "puts a" does
>
>   if a is an array
>     puts each element of it
>   else
>     a = a.to_s
>     print a
>     print "\n" unless a[-1] == ?\n
>   end

I can understand that for Array... but the thing that I found un-POLS-like
is:

  class Thing < Array
    def to_s
    # ...
    end
  end

and then Thing objects still get puts'd according to the rules of how
Array is handled.

In other words, not only is Array a special case, but overriding the
puts behavior/results for Array subclasses is a special case.  The
former makes sense to me; the latter doesn't.

Is this negotiable? :-)


David

-- 
David Alan Black
home: dblack@candle.superlink.net
work: blackdav@shu.edu
Web:  http://pirate.shu.edu/~blackdav