dolfin team mailing list archive

[Ken Kennedy <ken@xxxxxxxxxxxxxx>]

I seem to have fixed it. Replacing the subclass with the following
raises no problems and generates plots that are similar to the ones
given in the example. I have also attached the entire script. Sorry if
the formatting is not good, I am new to launchpad and I don't know if
there are special tags for code.

class BoundarySource(Expression):
    def __init__(self, mesh):
        self.mesh = mesh
    def eval_data(self, values, data):
        cell = Cell(self.mesh, data.cell().index())
        n = cell.normal(data.facet())
        x = data.x()
        g = sin(5*x[0])
        values[0] = g*n[0]
        values[1] = g*n[1]
    def value_shape(self):
        return (2,)
    def dim(self):
        return 2

If this can be updated on the website then that would be great!

** Attachment added: "demo4.py"
   https://bugs.launchpad.net/dolfin/+bug/769324/+attachment/2085560/+files/demo4.py

-- 
You received this bug notification because you are a member of DOLFIN
Team, which is subscribed to DOLFIN.
https://bugs.launchpad.net/bugs/769324

Title:
  Overloading eval_cell fails in demo4.py

Status in DOLFIN:
  New

Bug description:
  When running demo4.py (which demonstrates the mixed formulation) I get
  the following traceback:

  Traceback (most recent call last):
    File "demo4.py", line 50, in <module>
      class BoundarySource(Expression):
    File "/usr/lib/python2.6/dist-packages/dolfin/function/expression.py", line 221, in __new__
      raise TypeError, "expected an overload 'eval' or 'eval_data' method"
  TypeError: expected an overload 'eval' or 'eval_data' method

  This is installed on Ubuntu 10.10. If you need any more information,
  please let me know.

  I learned about FEniCS recently at a conference and I am interested in
  using it to solve a nonlinear problem which is best solved in the
  mixed formulation, so this example is of particular importance to me.