[Python-3000] sets in P3K?

[Nick Coghlan]

Edward Loper wrote:
> One way around this conflict might be to define a new factory function, 
> that generates a set from an iterable.  I.e., something like:
> 
>    s = set.from_iter(my_list)
>    s = set.from_iter(enumerate(my_list))
>    s = set.from_iter(x for x in collection if x>10)
> 
> (But with a better name than 'from_iter' :) )

Some might object to this as being too magical, but one option is a metaclass 
that overrides __getitem__ as an alternative constructor to the usual 
__call__. Numpy uses this trick to good effect for building matrices.

A single metaclass could cover most builtin and standard library containers:

   def __getitem__(self, args):
       # This just defines the semantics. A real implementation would
       # avoid the temporary objects wherever possible.
       if not isinstance(args, tuple):
           args = (args,)
       arglist = []
       for item in args:
           if isinstance(item, slice):
               if item.step is None:
                   item = (item.start, item.stop)
               else:
                   item = (item.start, item.stop, item.step)
           arglist.append(item)
       return self(*arglist)

Then you could have:

   s = set[1, 2, 3]         # == set((1, 2, 3))
   f = frozenset[1, 2, 3]   # == frozenset((1, 2, 3))
   d = dict[1:'x', 2:'y']   # == {1:'x', 2:'y'} == dict(((1,'x'), (2,'y')))
   l = list[1, 2, 3]        # == [1, 2, 3] == list((1, 2, 3))
   t = tuple[1, 2, 3]       # == (1, 2, 3) == tuple((1, 2, 3))

This can also be spelt more explicitly, but doing so pretty much misses the 
point of this discussion (we're only talking about avoiding a couple of 
parentheses, after all).

Cheers,
Nick.

-- 
Nick Coghlan   |   ncoghlan at gmail.com   |   Brisbane, Australia
---------------------------------------------------------------
             http://www.boredomandlaziness.org