From: Aaron Patterson <tenderlove@...>
Date: 2012-03-03T01:41:37+09:00
Subject: [ruby-core:43055] Re: return values of methods on String subclasses


--GvXjxJ+pjyke8COw
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Fri, Mar 02, 2012 at 10:58:43PM +0900, Nobuyoshi Nakada wrote:
> Hi,
>=20
> (12/03/02 13:38), Aaron Patterson wrote:
> >> Seems that patch would break other subclasses of String, whose
> >> initialize_copy doesn't expect whole original string.
> >=20
> > When would this happen?
>=20
> With your patch:
>=20
>   $ ./ruby -e 'class S < String; end; p S.new("abcdef")[0, 2]'
>   "abcdef"

Ah right.  I suppose we could check if the subclass actually has
`initialize_copy` in it's method table.  But I think if the subclass
calls `super` it would still break. :(

Maybe we do need a new hook method.

--=20
Aaron Patterson
http://tenderlovemaking.com/

--GvXjxJ+pjyke8COw
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (Darwin)

iQEcBAEBAgAGBQJPUPg8AAoJEJUxcLy0/6/GFDsIAIJIolUXO3JGbAg3pm3W0Iyz
NO3sXmyqDyrSBMp6SxnSrPDP40ezWt5PmqhhInXcucLarFVEBN5hfIW3hW/XcBc7
qQDt0kk7tUhYZrib8oLIBYMqZ49Z7rDWQMBJvUDGMdNwsb7Ta++3L9W/Ibt6L8tY
40BCIudZ/CDG/ek1cPoEf0/Z0qIK3qpf8EsOOFsQbWoPCedDePkLo9f4aYgrC88k
HL59s63CbP2eUsaDvcLWbwgqv8Seet0kjJe+4CFt0LVNVzyq6iWBggYyUIrklhuu
3vWAeNFDJy41BcCqoZmVEf/uXDe+yrBOZRaSjoOnpSzGMKo3Cywd2A5j09d+sow=
=V290
-----END PGP SIGNATURE-----

--GvXjxJ+pjyke8COw--