[Zope-Checkins] CVS: Zope27/lib/python/TAL - TALInterpreter.py:1.68.26.2

[Matt Behrens]

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

On Mon, Sep 09, 2002 at 08:16:22AM -0400, Guido van Rossum wrote:

> > > What on earth is "monkey-patching"?  From the patch, it looks like you
> > > mean "overriding a method."
> >=20
> > Well, that's pretty much it -- replacing a function or class or
> > something in an already-loaded module.  Most (all?) hotfixes use
> > this mechanism.
>=20
> Um, that's not the same.  The word "overriding" implies "in a
> subclass".  You're talking about patching a method in an *existing*
> class.  The difference is that the latter affects all users of the
> class.  For urgent bugfixes I can understand this technique.  But
> Florent's change suggests that this is a *feature* in this case.  I
> find that hard to swallow as a general mechanism, and because his
> checkin comments gave not additional motivation or a pointer to an
> accepted decision, I question the validity of his checkin, *unless* he
> meant overriding the method in a subclass (which is goodness).

Right, it is not overriding in the OO sense.  Sorry,  I am still
waking up. :-)  But monkey-patching is still generally accepted as
how I described it.  Maybe FG means something different by it?

--=20
Matt Behrens <matt@zigg.com> <URL:http://zigg.com/>

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

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (OpenBSD)

iD8DBQE9fJOi+xq4JbgNGlMRAqt2AJ4n2j59xCpVTCslpF5Ux7tnjbT1oACcD9ze
3vI6POrnqOip9BIdqJCODbo=
=2op0
-----END PGP SIGNATURE-----

--FCuugMFkClbJLl1L--