Re: Behavior of linspace with vector inputs

[Rik]

On 08/27/2015 01:49 PM, John W. Eaton wrote:
> On 08/27/2015 04:13 PM, rik wrote:
>> 8/27/15
>>
>> Thanks everyone for the input.  It seems like more trouble than it's worth
>> so I left in place the current behavior and just fixed the compatibility
>> issue for N < 1.  See changeset
>> http://hg.savannah.gnu.org/hgweb/octave/rev/16b9ec39ff46.
>
> It appears to me that the following changesets introduced the vector
> behavior of linspace:
>
>   http://hg.savannah.gnu.org/hgweb/octave/rev/e087d7c77ff9
>   http://hg.savannah.gnu.org/hgweb/octave/rev/3429c956de6f
>
> I was unable to find any discussion of this change.  I'd bet that
> Jaroslav just thought it was a good thing to do and made the changes
> without really consulting anyone else.  I wonder if there is much code
> that depends on this feature.  Do we need it for Octave?  Is it worth
> keeping?
>

I think it's kind of cool to be a superset of Matlab and offer more
functionality.  I doubt it gets used that often, although Carne reported
using it so at least someone has chanced upon the code.

--Rik