[Numpy-discussion] New patch for new mgrid / ogrid functionality

[Robert Kern]

On Mon, Sep 22, 2008 at 03:53, Gael Varoquaux
<gael.varoquaux at normalesup.org> wrote:
> On Mon, Sep 22, 2008 at 09:40:42AM +0200, Stéfan van der Walt wrote:
>> 2008/9/19 David M. Kaplan <David.Kaplan at ird.fr>:
>> > My 2 cents - I personally think the version that always returns a list
>> > will ultimately be more transparent and cause fewer problems than the
>> > newer version.  In either case, the plan should be to eventually have it
>> > always return a list as that is the only fully consistent option, the
>> > question is just when that switch should be made and by who.  If it is
>> > done at the next major release, someone else will have to remember to ax
>> > the additional code and correct the documentation....
>
>> I think this change is worth making, because:
>
>> - It provides a consistent API for gridding.
>> - It optimises the most frequenty used situation:
>
>> x, y, z = ogrid[...]
>
>> - Someone already took the trouble of writing the patch.
>
>> If the (minor) API breakage is considered a problem, could we go
>> through the standard deprecation process and have this included?
>
> I must admit I don't really like the API breakage. People will find out
> that some code that used to work doesn't work anymore.
>
> In addition there is some real usecase behind the mgrid behavior:
>
>
> X = mgrid[0:4, 0:4, 0:4]
> X = X.reshape((3, -1))
>
> scatter(X[0, :], X[1, :])
>
> R = matrix(((cos(0.5), sin(0.5), 0), (-sin(0.5), cos(0.5), 0), (0, 0, 1)))
>
> Y = array(R*X)
>
> scatter(Y[0, :], Y[1, :], color='r')

Agreed. Please don't change this behavior even through a deprecation.

-- 
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