[Scilab-users] contour : z level use

[Samuel Gougeon]

Le 02/04/2017 à 14:49, Rafael Guerra a écrit :
> // NOTE: the previous contour code line should read:
> contour(X,Y,Z,[Zc Zc],flag=[0 0 4]);
Right.
When Zc is not a strictly positive integer, it should be taken as a 
single Z level.
But it is currently not the case. It is a bug : 
http://bugzilla.scilab.org/13467

Then, for plotting the only level z=0, specifying 1e-300 would be OK
And for plotting the only level z=2, 2*(1+%eps) would do it.

Doubling [Zc Zc] is another trick. contour() does not apply unique() on 
specified levels,
and so processes the same level twice. Another improvement to implement.