Re: Slice implementation bug

[Andrea Crotti]

Tambet <qtv...@gmail.com> writes:

> Hello!
>
> Let's say slice is multidimensional now - how to interpret it?
>

But who said that slice is multidimensional in the first place?

Since python is not matlab I don't think it will ever be done, so what's
the purpose of talking about a bug in something that doesn't exist?

PS. probably in numpy you can find what you look for

[Robert Kern]

On 5/12/11 6:06 AM, Tambet wrote:
> Hello!
>
> Let's say slice is multidimensional now - how to interpret it?
>

> I excpect these to work:
>
> * m[0, 3] - get one element from matrix
> * m[0:2, 0:2] - get four elements from matrix, iterate over them (I have
> actually an rtree if it doesn't make sense to you)
>
> But it won't, because if m[0, 3] returns something, then m[0:2, 0:2] cannot
> yield anymore.

Let me try to rephrase, since you seem to be leaving out a lot of assumptions.
You are trying to say that you want m[0:2,0:2] to return an iterator and that if
you define __getitem__() such that m[0,3] returns a value, then you cannot
implement __getitem__() to be a generator using a yield statement.

Okay. Fine.

But there is no reason that __getitem__() needs to be a generator function for
you to return an iterator. You can simply return another generator. For example:

def __getitem__(self, key):
if isinstance(key, tuple):
if any(isinstance(x, slice) for x in key):
return self._generate_from_slice(key)
# The default, boring case.
return self._get_value(key)

def _generate_from_slice(self, key):
yield foo
yield bar
yield etc

> Ofcourse I could return an iterator, but this would not be so simple.

Really, it is.

--
Robert Kern

"I have come to believe that the whole world is an enigma, a harmless enigma
that is made terrible by our own mad attempt to interpret it as though it had
an underlying truth."
-- Umberto Eco