[Python-Dev] Re: the "3*x works w/o __rmul__" bug

[Alex Martelli]

On Tuesday 28 October 2003 07:28 pm, Guido van Rossum wrote:
   ...
> You're making a mountain of a molehill here, Alex.  I know that in

You have a point: when one is in love (and I still _am_ madly in love
with Python!-), it's hard to admit of imperfections in the loved one:-).
Even a tiny defect...:-).

> group theory there are non-Abelian groups (for which AB != BA), but
> I've never encountered one myself in programming; more typical such
> non-commutative operations are modeled as __add__ rather than as
> __mul__.

I don't remember ever coding a __mul__ that I WANTED to be
non-commutative, right.

> Anyway, the real issue AFAICT is not that people depend on __rmul__'s
> absence to raise a TypeError, but that people learn by example and
> find __rmul__ isn't necessary by experimenting with integers.

Or more seriously: they write what LOOK like perfectly adequate unit
tests, but the numbers they try in "number * x" happen to be ints;
so the unit tests pass -- but their code is broken because they forgot
the __rmul__ and the unittests-with-ints didn't catch that.

> The reason why it works at all for integers without __rmul__ is
> complicated; it has to do with very tricky issues in trying to
> implement multiplication of a sequence with an integer.  That code has

Yes, I think I understand some of that -- I included the analysis of the
bug in my bugreport on SF.

> gone through a number of iterators, and every time someone eventually
> found a bug in it, so I'd rather leave the __rmul__ blemish than
> uproot it again.  If you can come up with a fix that doesn't break
> sequence repetition I'd be open to accepting it (for 2.4 only, in 2.3
> there may be too much code depending on the bug) but only after
> serious review -- and not by me, because I'm not all that familiar
> with all the subtleties of that code any more. :-(

I do have one weird idea that might help here (for 2.4), but I'd better
post that separately because I suspect it's going to fuel its own long
discussion thread.  As things are, without an ability to distinguish a
sequence from a number securely, and IF sequences must work w/o
__rmul__ (but they didn't in classic classes...? and the docs don't
indicate that...?) then I'm stumped.  Who'd be the right people to
review such proposed changes, btw?


Alex