--FCuugMFkClbJLl1L
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Tue, Jul 13, 2010 at 04:06:26AM +0900, Yukihiro Matsumoto wrote:
> Hi,
> 
> In message "Re: [ruby-core:31217] [Bug #3562] regression in respond_to?"
>     on Tue, 13 Jul 2010 03:45:55 +0900, Aaron Patterson <redmine / ruby-lang.org> writes:
> 
> |This code works in 1.8, 1.9.2, but not trunk.  This change to respond_to?() breaks existing ruby code[1].  Is that intended?  Could we compromise by including private and protected methods when an object inflects upon itself?
> 
> Hmm, that check is costly for respond_to?()  Could you replace
> 
>   respond_to?(:"do_#{name}")
> 
> by
> 
>   respond_to?(:"do_#{name}",true)
> 
> ?

That seems fine, but someone writing code like my example against 1.9.2
will find it broken in 1.9.3.

Could we backport this to 1.9.2 before release?  Or find some other
solution?

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

--FCuugMFkClbJLl1L
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (Darwin)

iQEcBAEBAgAGBQJMO4llAAoJEJUxcLy0/6/GlSQH/16i7PzSyC/E9zILM6QKu7Gz
AOp22yGyjCF8sxIK+/gE4st53BM9X5ZQvTjlU88ODwF+SyJ6JevE4V3S1oJZCOaZ
Ex2OkgdoH7F7dV0Ctc5S0nYjWDoi2Tl7PrGazQYJB7iNZvMsSRGPgmJm18qkOuwb
XPim/HZWNjJQzgkAcPyPpNP2Xv1Br0eXXydC0qDq7cKiFbAE3LYZcOY5v43vp3vm
ss0ZXU6Cueeg8H7IqeLc5hbWxy+Be1Q7q0GqG9OZCEsQkt7K0IZcOeBKSFLF0i9K
71BRXJXlAcpzo4FLaKTJGkHsx/ZAWN9WbVwRS0oxOAR+bobqVs0SqZG+E2cLFnUWs
-----END PGP SIGNATURE-----

--FCuugMFkClbJLl1L--