FFC breaks immutability of elements when using quadrature elements

[Martin Sandve Alnæs]

This has been there forever, just making an issue for the record.

In ffc/analysis.py:

    # FIXME: This modifies the elements depending on the form compiler parameters,
    #        this is a serious breach of the immutability of ufl objects, since the
    #        element quad scheme is part of the signature and hash of the element...
    for element in form_data.unique_sub_elements:
        if element.family() == "Quadrature":
            element._quad_scheme = scheme

If you're using quadrature elements in multiple forms with different schemes, this will cause trouble.

[Martin Sandve Alnæs]

I'm fixing this by requiring the degree and scheme to be specified by the user and consistent with the form it's used in. If someone has a problem with that they can put in the work to find a better solution.

[Martin Sandve Alnæs]

This should then be equivalent to todays

def QuadratureElement(scheme, cell, degree):
    return FiniteElement("Quadrature", cell, degree, quad_scheme=scheme)

[Jan Blechta]

Great!

[Jan Blechta]

@martinal, could you check also quadrature degree, not just scheme, so that this

mesh = UnitSquareMesh(1, 1)
e = FiniteElement("Quadrature", mesh.ufl_cell(), 2, quad_scheme='default')
f = Expression("x[0]", element=e)
V = FunctionSpace(mesh, "Lagrange", 1)
u = project(f, V) # complains about non-matching quadrature points
u = project(f, V, form_compiler_parameters={'quadrature_degree': 2}) # passes fine

complains earlier with message like non-matching quadrature degree of quadrature integrand ({}) and measure ({}) instead of letting this go so far and raising

('\npoints:\n', array([[ 0.65902762,  0.23193337],
       [ 0.65902762,  0.10903901],
       [ 0.23193337,  0.65902762],
       [ 0.23193337,  0.10903901],
       [ 0.10903901,  0.65902762],
       [ 0.10903901,  0.23193337]]))
('\nquad points:\n', array([[ 0.16666667,  0.16666667],
       [ 0.16666667,  0.66666667],
       [ 0.66666667,  0.16666667]]))
Points must be equal to coordinates of quadrature points
Traceback (most recent call last):
...
Exception: Points must be equal to coordinates of quadrature points

[Prof Garth Wells]

Removing milestone: 1.7 (automated comment)