On Jan 11, 9:41 am, Maarten <maarten.sn...@knmi.nl> wrote:
> On Jan 8, 10:16 pm, Paul van Delst <paul.vande...@noaa.gov> wrote:
>
>
>
>> Maarten wrote:
>>> On Jan 7, 6:56 pm, mgalloy <mgal...@gmail.com> wrote:
>>>> I think we are agreeing here, but just to be sure: Python and IDL would
>>>> be specifying the endpoints of the range in the same way, it's just that
>>>> Python always includes the start index and excludes the end index (even
>>>> if not using negative indices):
>
>>>> >>> a = [1, 2, 3, 4]
>>>> >>> a[1:3]
>>>> [2, 3]
>
>>> Yes. Although this is a fundamental difference that is the result of a
>>> choice both language developers made. Thinking about it a bit longer,
>>> I don't think the two can be made to act the same: IDL always includes
>>> the end index of the range, while Python always excludes it. Some
>>> emphasis on this in the documentation may be needed, as Python
>>> probably is the most widespread programming language that offers the
>>> facility of negative indices.
>
>> Well, since they're mucking about with operators in general, maybe ITTVIS could go the
>> ruby route and introduce the ".." and "..." range operators. The former is an inclusive
>> range (same functionality as ":") and the latter is a range that excludes the higher
>> value. So,
>
>> $ irb
>> irb(main)> a = [1,2,3,4,5,6]
>> => [1, 2, 3, 4, 5, 6]
>
>> irb(main)> a[1..3]
>> => [2, 3, 4]
>
>> irb(main)> a[1...3]
>> => [2, 3]
>
>> irb(main)> a[1..-1]
>> => [2, 3, 4, 5, 6]
>
>> irb(main)> a[1...-1]
>> => [2, 3, 4, 5]
>
> That is one option. Of course, python doesn't stop at a[1:-1], it can
> also do a[-1:1:-1], resulting in [6, 5, 4, 3] (with a as above). That
> is, it includes a stride (including negative stride) in its array
> indexing.
Oh, to add to the fun: python uses ... for a different operation:
specifying all elements for all non-explicitly mentioned dimensions.
This allows you to write code the can handle an arbitrary number of
dimensions of your array.
import numpy as np
a = np.arange(0,3*4*5*6,1)
a = a.reshape((3,4,5,6))
b = a[...,2,:]
c = a[1,...,0]
print(c.shape)
(4, 5)
print(c)
array([[120, 126, 132, 138, 144],
[150, 156, 162, 168, 174],
[180, 186, 192, 198, 204],
[210, 216, 222, 228, 234]])
print(b.shape)
(3, 4, 6)
(won't print here, too large).
Best,
Maarten - Each language will reinvent similar things in different ways
- Sneep
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
![Maarten[1] is currently offline](theme/default/images/xoffline.png.pagespeed.ic.XRkd1fkXye.png "Maarten[1] is currently offline")
