Some syntactic sugar proposals

[Mark Dickinson]

On Nov 15, 12:46 pm, Tim Chase <python.l... at tim.thechases.com> wrote:
> On 11/15/2010 12:39 AM, Dmitry Groshev wrote:
>
> > x in range optimisation
>
> I've often thought this would make a nice O(1)-test lookup on an
> xrange() generator.

An O(1) test for 'x in <range_object>' is implemented in Python 3.2,
at least provided that x has type 'int', 'long' or 'bool'.  (If x is
an instance of a subclass of int or long, then there's a risk that the
semantics of the membership test have been changed by an explicitly
overridden __eq__, so Python wimps out and falls back to the O(n)
algorithm in that case.)

Python 3.2a4+ (py3k:86635:86636M, Nov 21 2010, 19:22:18)
[GCC 4.2.1 (Apple Inc. build 5664)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> -1 in range(10**9)
False
>>> 5 in range(0, 10**100, 2)
False
>>> 10**99 in range(0, 10**100, 2)
True

IIRC, there wasn't sufficient interest to get it backported to Python
2.7 in time for its release.  Though as a pure optimization, one could
argue that it would still be possible to get this into Python 2.7.2.

Mark