re bad leadership
tjreedy at home.com
Wed Jul 25 00:38:50 EDT 2001
"David Bolen" <db3l at fitlinxx.com> wrote in message
news:u4rs1khuu.fsf at ctwd0143.fitlinxx.com...
> "Terry Reedy" <tjreedy at home.com> writes:
> > > So why not swap the proposed meanings of // and /? I realize
> > is the
> > > third time I've said this but I think it would silence a lot of
> > objections
> > > and no one seems to have noticed the other two times.
> > This would break as much or more code as the current proposal.
> > the PEP.
> Are you sure? I can't really see anything in the PEP that would
> to imply this.
PEP238 proposes that / be the inverse of *, so that '(a/b) * b = a
(more or less)'
and that 'a // b == floor(a/b)'. If we 'swap the proposed meanings of
// and /', then we would have a/b == floor(a/b), which would break
almost all existing floating point code.
Terry J. Reedy
More information about the Python-list