semantics of sum, prod, cumsum etc.

[adler]

Here is a question I asked previously,
but it was burried in the answer to another,
so it seems to have been overlooked.

I would like to be able to use sum(x,1) to
force column behaviour, even when ncols==1.

I'd like to submit a patch to implement this
behaviour - would it be accepted, or is the
current semantics accepted as correct?

> For column vectors, the direction parameter of sum
> allows control of the direction
>
>    octave-2.1.34:37> sum( [1 2 3]' )
>    ans = 6
>    octave-2.1.34:38> sum( [1 2 3]',1 )
>    ans = 6
>    octave-2.1.34:39> sum( [1 2 3]',2 )
>    ans =
>      1
>      2
>      3
>
> However, this is not the case for row vectors.
>
>    octave-2.1.34:34> sum( [1 2 3] )
>    ans = 6
>    octave-2.1.34:35> sum( [1 2 3],1 )
>    ans = 6
>    octave-2.1.34:36> sum( [1 2 3],2 )
>    ans = 6
>
> In other words sum(x) == sum(x,1).
> There is no way to force column behaviour.
>
> Is this a bug?

_______________________________________
Andy Adler,                address@hidden



-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.octave.org
How to fund new projects:  http://www.octave.org/funding.html
Subscription information:  http://www.octave.org/archive.html
-------------------------------------------------------------