Re: [OctDev] fminbnd incompatibility with octave-3.4.x

[fabien a. <fab...@fe...>]

On 21/08/2012 17:23, Martin Helm wrote:
> Am 21.08.2012 17:33, schrieb fabien amiot:
>    
>> On 21/08/2012 17:03, Martin Helm wrote:
>>      
>>> Am 21.08.2012 17:22, schrieb fabien amiot:
>>>
>>>        
>>>> signal | 1.1.3 |
>>>>
>>>>          
>>> There is no * which means signal is not loaded.
>>>
>>> Either do a
>>> pkg load signal
>>> or
>>> pkg rebuild -auto signal
>>>
>>>
>>>
>>>        
>> adding
>>
>> pkg load signal
>>
>> yields
>>
>> Package Name   | Version | Installation directory
>> ---------------+---------+-----------------------
>>         control *|  2.3.52 |
>> .../octave/3.6.2/share/octave/packages/control-2.3.52
>>         general  |   1.3.1 |
>> .../octave/3.6.2/share/octave/packages/general-1.3.1
>>         gnuplot  |   1.0.1 |
>> .../octave/3.6.2/share/octave/packages/gnuplot-1.0.1
>>             gsl *|   1.0.8 |
>> .../apps/octave/3.6.2/share/octave/packages/gsl-1.0.8
>>           image *|  1.0.15 |
>> .../octave/3.6.2/share/octave/packages/image-1.0.15
>> miscellaneous *|   1.1.0 |
>> .../3.6.2/share/octave/packages/miscellaneous-1.1.0
>>           optim *|   1.2.0 |
>> .../octave/3.6.2/share/octave/packages/optim-1.2.0
>>        parallel *|   2.0.5 |
>> .../octave/3.6.2/share/octave/packages/parallel-2.0.5
>>          signal *|   1.1.3 |
>> .../octave/3.6.2/share/octave/packages/signal-1.1.3
>>         specfun *|   1.1.0 |
>> .../octave/3.6.2/share/octave/packages/specfun-1.1.0
>>          struct *|  1.0.10 |
>> .../octave/3.6.2/share/octave/packages/struct-1.0.10
>>
>> but also the same error:
>> error: Invalid call to fminbnd.  Correct usage is:
>>
>> Function File: [X, FVAL, INFO, OUTPUT] = fminbnd (FUN, A, B,
>>             OPTIONS)
>>
>>
>> Fabien
>>
>>      
> I do not get that with 3.6.2 and signal 1.1.3
>
> check with
> which fminbnd
>
> if the fminbnd which is called is really from the signal package!
>
>
> octave:1>  pkg load signal
> octave:2>  pkg list
> Package Name   | Version | Installation directory
> ---------------+---------+-----------------------
>        control *|  2.3.52 | /usr/share/octave/packages/control-2.3.52
>          image *|  1.0.15 | /usr/share/octave/packages/image-1.0.15
>           java  |   1.2.9 | /usr/share/octave/packages/java-1.2.9
> miscellaneous *|   1.1.0 | /usr/share/octave/packages/miscellaneous-1.1.0
>         odepkg  |   0.8.2 | /usr/share/octave/packages/odepkg-0.8.2
>          optim *|   1.2.0 | /usr/share/octave/packages/optim-1.2.0
>         signal *|   1.1.3 | /usr/share/octave/packages/signal-1.1.3
>        specfun *|   1.1.0 | /usr/share/octave/packages/specfun-1.1.0
>         struct *|  1.0.10 | /usr/share/octave/packages/struct-1.0.10
> octave:3>  [b,a]=ellip(5,1,90,[.1,.2])
> b =
>
>   Columns 1 through 6:
>
>     0.0001321  -0.0006639   0.0014925  -0.0019624   0.0014425   0.0000000
>
>   Columns 7 through 11:
>
>    -0.0014425   0.0019624  -0.0014925   0.0006639  -0.0001321
>
> a =
>
>   Columns 1 through 6:
>
>       1.00000    -8.64826    34.60318   -84.21550   137.92762  -158.75980
>
>   Columns 7 through 11:
>
>     130.04250   -74.86358    29.00444    -6.83593     0.74556
>
> octave:4>
>
>
>    


That's the one from the optimization package, we upgraded together with 
signal package.
I do sincerely apologize: it appears that my first message was rather 
unclear : I'm not using ellip for testing purpose, but the functions I 
developed invoking fminbnd. These functions are running fine under 
3.2.4. So does the ellip test under 3.6.2 :

[b,a]=ellip(5,1,90,[.1,.2])


b =

  Columns 1 through 6:

    0.0001321  -0.0006639   0.0014925  -0.0019624   0.0014425   0.0000000

  Columns 7 through 11:

   -0.0014425   0.0019624  -0.0014925   0.0006639  -0.0001321

a =

  Columns 1 through 6:

      1.00000    -8.64826    34.60318   -84.21550   137.92762  -158.75980

  Columns 7 through 11:

    130.04250   -74.86358    29.00444    -6.83593     0.74556


I cannot find out where ellip calls fminbnd, but my codes running under 
3.2.4 call fminbnd like that :

fminbnd('function',lower_bound,upper_bound,[], 
parameters_to_be_passed_to_function)

and I suspect the  parameters_to_be_passed_to_function not to be 
accepted anymore...

Offering again my apologies,
Fabien

-- 
Fabien AMIOT
Chargé de recherche CNRS en colere / cross Research Associate
fab...@fe...
Tel : (+33) (0)3.81.66.60.14
Fax : (+33) (0)3.81.66.67.00
UMR6174 / FEMTO-ST, Dpt. Mécanique Appliquée
24, rue de l'Épitaphe
25000 Besancon
France