Rationale behind the deprecation of __getslice__?

[Steven Bethard]

Fernando Perez wrote:
> I guess that conceptually it just felt natural to me to keep separate methods
> for dealing with a slice (get many elements out) and other types of indexing,
> which I tend to think of as 'scalar' indexing.

Yeah, I can see that a bit.

Ignoring dicts for the moment (and concerning ourselves only with 
"sequences"), you're probably right in thinking that that slice objects 
are the second most common thing to get in __getitem__ (second to ints 
of course).  But there is heavy use of other objects in various other 
modules, most notably tuples in numarray:

 >>> import numarray as na
 >>> a = na.arange(24, shape=(2, 3, 4))
 >>> a
array([[[ 0,  1,  2,  3],
         [ 4,  5,  6,  7],
         [ 8,  9, 10, 11]],

        [[12, 13, 14, 15],
         [16, 17, 18, 19],
         [20, 21, 22, 23]]])
 >>> a[0,1,2]
6
 >>> a[0,1]
array([4, 5, 6, 7])
 >>> a[...,3]
array([[ 3,  7, 11],
        [15, 19, 23]])
 >>> a[1,:,0]
array([12, 16, 20])

Presumably the numarray code has to do quite a bit of type checking to 
perform all these slicings right (and I didn't even show you what 
happens when you use another array as an "index").  I'm not necessarily 
saying that all this type checking is a good thing, but because people 
will always find new things that they want to index by, adding 
__getxxx__ methods for each of the index types is probably not the right 
road to go down...

Steve