From maintainers-request at octave dot org Fri Jan 21 03:28:08 2005
Subject: Re: cell behavior
From: David Bateman <David dot Bateman at motorola dot com>
To: Daniel J Sebald <daniel dot sebald at ieee dot org>
Cc: octave maintainers mailing list <maintainers at octave dot org>
Date: Fri, 21 Jan 2005 10:29:07 +0100

Daniel J Sebald wrote:

> Is the following proper behavior of cell assignments?  (2.1.64)
>
> octave:124> C = cell(2,1)
> C =
>
> {
>  [1,1] = [](0x0)
>  [2,1] = [](0x0)
> }
>
> octave:125> C(2) = []
> C =
>
> {
>  [1,1] = [](0x0)
> }
>
> It could be interpretted two ways, but my intuition would be that C 
> should remain just as it already is.
>
> I have an iterative search that some times comes up empty.  (I realize 
> it is possible to fix this with a conditional "isempty".)
>
> Dan


Dan,

This is intentional. The equivalent for a matrix might be

octave:1> M = rand(2,2)

M =

  0.402375  0.721572
  0.164786  0.040139

octave:2> M(:,2) = []

M =

  0.40238
  0.16479

Assigning a part of a matrix or cell array to the empty matrix has the 
effect of deleting that part of the matrix. Matlab v7sp1 gives the same 
result


 >> C = cell(2,1)

C =

     []
     []

 >> C(2) = []

C =

    {[]}


Octave and matlab also agree for the version "C{2} = []" which assigns 
an empty matrix rather than deletes the row. Maybe this is what you 
wanted to write..

Regards
David

-- 
David Bateman                                David dot Bateman at motorola dot com
Motorola CRM                                 +33 1 69 35 48 04 (Ph) 
Parc Les Algorithmes, Commune de St Aubin    +33 1 69 35 77 01 (Fax) 
91193 Gif-Sur-Yvette FRANCE

The information contained in this communication has been classified as: 

[x] General Business Information 
[ ] Motorola Internal Use Only 
[ ] Motorola Confidential Proprietary