Why no number methods?
Michael Hudson
mwh at python.net
Tue May 15 02:13:55 EDT 2001
"Magnus Lie Hetland" <mlh at idi.ntnu.no> writes:
> It's online at O'Reilly's website. (No URL handy...)
>
> He says that Python 3000 is a *long* way off. Whether that means that
> the type/class gap will stay open far into the future is another
> matter :)
Ah, right. Yes Py3K probably is a long way off - but some of the
things that you might expect to be in Py3K might be in 2.2. Or not.
> > Well, assume that we have magically healed the type/class split so
> > that all objects behave like instances. Then imagine:
> >
> > class C:
> > def __repr__(self):
> > return "bongle"
> >
> > What's C.__repr__? Quick!
>
> <scratches head> An unbound method? (I *must* be missing something here ;)
But if all objects behave like instances, somewhere there is a class
"class" (probably written in C) that looks a bit like this:
class class:
...
def __repr__(self):
return "<class %s.%s at %d>"%(self.__module__,
self.__class__,
id(self))
...
So C.__repr__ might be expected to return a bound version of *this*
method (and in some situations will have to, as otherwise
print C
would blow up with an exception).
Cheers,
M.
--
Hmmm... its Sunday afternoon: I could do my work, or I could do a
Fourier analysis of my computer's fan noise.
-- Amit Muthu, ucam.chat (from Owen Dunn's summary of the year)
More information about the Python-list
mailing list