Re: mesh handles not working?

[Daniel J Sebald]

John W. Eaton wrote:
> On  6-Mar-2007, Daniel J Sebald wrote:
>
> | I think that mesh() in CVS isn't working correctly in terms of the
> | handles.  It appears the handle from __uiobject_get_handle__.m is
> | returning a handle value which is one too high so that every time
> | mesh() is called the old surface remains.
> | 
> | There may also be another problem inside __uiobject_draw_axes__.m
> | with surfaces but I can't get that far to isolate things.
> | 
> | Anyway, I think the problem is near lines 106 to 111, but these
> | globals with lock and unlock is too much for me to follow at the
> | moment.
>
> All the files that use the globals directly are generated
> automatically from the corresponding .in files.  But anyway, those are
> all going away in favor of managing the properties in C++.  Doing all
> of that bookkeeping in the scripting language was a noble experiment,
> but I think the "Matlab language" is just not up to the task.

If it weren't so slow.  Just curious why it is slow.  Does it have to do with 
cells?  Lock/unlock?


> I think the fix is just ot add a call to newplot at the beginning of
> mesh.m.

Yes, that fixes things.

BTW, there is a user who submitted to the graphics list about surface plots.  
Is there surface plots?  (I know it isn't implemented because I don't see the 
proper commands in __uiobject_draw_axes__.m.)  Or is mesh() all there is?

Dan