[scilab-Users] Weird behaviour dealing with matrices

[Eduardo Torrecillas]

Hi Adrien,

Thanks for your reply, but that's not exactly what I meant.
I understand in your example we're dealing with passing variables through
different functions. In my case, it's all inside the same function.
Also, apparently the behaviour is erratic: it works sometimes and sometimes
not.

Regards,

On Tue, Jun 14, 2011 at 10:26 AM, Adrien Vogt-Schilb
<vogt at centre-cired.fr>wrote:

>  Hi Edouardo
>
> This is not a bug, but a feature!
>
> you should read this :
> http://wiki.scilab.org/howto/global%20and%20local%20variables
>
> In the commented example, look for the following line :
>
>   x(3)=56          // note that at this point x=[0 0 56] because of 3. above, *not* [45 0 56]
>
> this is exactly want happened to you
>
>
>
> On 13/06/2011 22:11, Eduardo Torrecillas wrote:
>
> Hi all,
>
> I have some piece of code and noticed some strange behaviour. As the code
> is complex and involves external compilation, it is a little bit difficult
> to post it here, so I will try to explain it in a general way.
>
> At some point in the code, I have a structure called interpolados, having 3
> matrices. So, I have
>
> interpolados(1).matriz
> interpolados(2).matriz
> interpolados(3).matriz
>
> all of them are sized 497 x 95.
>
> Later on, when trying to substitute a column of one of this matrices, I get
> it destroyed.
> The syntax I'm using is:
>
> interpolados(1).matriz(:,3)=13
>
> I was expecting to maintain the matrix and change only its third column,
> all with 13. Actually what I'm getting is that interpolados(1).matriz is no
> longer 497 x 95, but 497 x 1 (only 1 column, all full with 13).
>
> System summary: Ubuntu 10.10 64bit + Scilab 5.3.2
>
> Has anybody experienced that? Any issues?
>
> Regards,
>
>
> --
> Eduardo Torrecillas
>
>
>
> --
>
> *Adrien Vogt-Schilb*
>
> Research Fellow
>
> vogt at centre-cired.fr
>
> Tel: (+33) 1 43 94 73 96
>
> Fax: (+33) 1 43 94 73 70
>
> [image: CIRED]
>
> 45 bis, Av de la Belle Gabrielle
>
> F-94736 Nogent-sur-Marne
>
> http://www.centre-cired.fr/
>
>
>



-- 
Eduardo Torrecillas
AER-09
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20110614/6d2b485d/attachment.htm>