how I can copy the results

On Wed, Oct 31, 2012 at 04:01:16PM -0000, Ajeeb wrote:
> New question #212878 on FEniCS Project:
> https://answers.launchpad.net/fenics/+question/212878
>
> Hi, I'm new user of fenics, I write the problem in python and run in
> the fenics window, successfully i get the results, now how I can
> copy these results from fenics window . I'm using widow 7. Any help
> in this regard will be highly appreciated.

Press 'i' to save to .png. Or better, save the solution to .pvd and open
it in paraview for postprocessing.

--
Anders

Thanks Anders Logg, that solved my question.

Thanks Anders for quick replying, I want to ask one more question
I am trying to solve three variables, two (u,v) of them I have to defined on whole space and one (q) of them I have to define on boundary only, I mean q lives only on lower boundary where y=0 , I defined in this way but not gives correct results
mesh = UnitSquare (4,4)
X = FunctionSpace (mesh,'Lagrange',1)
Y = FunctionSpace (mesh,'Lagrange',1)
M = FunctionSpace (mesh,'Lagrange',1)
Z=MixedFunctionSpace([X,Y,M])
#Z = X*Y
(u,v,q) = TrialFunctions(Z)
(si,phi,xi) = TestFunctions(Z)

On Sun, Nov 04, 2012 at 02:11:14PM -0000, Ajeeb wrote:
> Question #212878 on FEniCS Project changed:
> https://answers.launchpad.net/fenics/+question/212878
>
> Status: Solved => Open
>
> Ajeeb is still having a problem:
> Thanks Anders for quick replying, I want to ask one more question
> I am trying to solve three variables, two (u,v) of them I have to defined on whole space and one (q) of them I have to define on boundary only, I mean q lives only on lower boundary where y=0 , I defined in this way but not gives correct results
> mesh = UnitSquare (4,4)
> X = FunctionSpace (mesh,'Lagrange',1)
> Y = FunctionSpace (mesh,'Lagrange',1)
> M = FunctionSpace (mesh,'Lagrange',1)
> Z=MixedFunctionSpace([X,Y,M])
> #Z = X*Y
> (u,v,q) = TrialFunctions(Z)
> (si,phi,xi) = TestFunctions(Z)

It's not possible to have variables living only on the boundary,
unless you hack it using ident_zeros (search the Q/A database) but it
will still be inefficient.

--
Anders