ribbon broken

[David Bateman]

John W. Eaton wrote:
> On  7-Dec-2007, kahacjde wrote:
>
> | 
> | 
> | I don't think ribbon is doing the right thing now.  With Octave built
> | from the current CVS sources, I see the attached for
> | 
> |   [x, y, z] = peaks ();
> |   ribbon (y, z);
> | 
> | when I think it should be like the normal peaks plot but with the
> | surface cut into strips (ribbons).
> | 
> | Any clues about what might have broken this?
> | 
> | jwe
> | 
> | 
> | 
> | I cannot check at the moment, but when looking at the code the surface call
> | could be the problem.
> | 
> |   tmp(c) = surface (xx, yy, zz, cc);
> |  
> | if xx contains a 1 or a valid axes value, then the first argument could be
> | intepreted as axes, and the surface call is still valid. 
> | 
> | If this is the problem, we have to fix the __plt_get_axis_arg__ function. I
> | think it would help if only scalar values are allowed for an axes argument
>
> I think that is already true.
>
> jwe
>
>   

I think its also due to the broken calculation of the axis limits in the
presence of multiple 3D objects.. See the message attached. With this
patch ribbon works correctly for me.

D.



-- 
David Bateman                                David.Bateman at motorola.com
Motorola Labs - Paris                        +33 1 69 35 48 04 (Ph) 
Parc Les Algorithmes, Commune de St Aubin    +33 6 72 01 06 33 (Mob) 
91193 Gif-Sur-Yvette FRANCE                  +33 1 69 35 77 01 (Fax) 

The information contained in this communication has been classified as: 

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

-------------- next part --------------
An embedded message was scrubbed...
From: David Bateman <David.Bateman at motorola.com>
Subject: Re: meshc, surfc, and contour are not working
Date: Fri, 07 Dec 2007 16:53:50 +0100
Size: 4019
Url: https://www.cae.wisc.edu/pipermail/bug-octave/attachments/20071207/a062aa9a/attachment-0001.mht