from functools import partial
from operator import add, methodcaller
from itertools import chain
import numpy
import FIAT
import gem
from gem.utils import cached_property
from finat.finiteelementbase import FiniteElementBase
[docs]class EnrichedElement(FiniteElementBase):
"""A finite element whose basis functions are the union of the
basis functions of several other finite elements."""
def __new__(cls, elements):
elements = tuple(chain.from_iterable(e.elements if isinstance(e, EnrichedElement) else (e,) for e in elements))
if len(elements) == 1:
return elements[0]
else:
self = super().__new__(cls)
self.elements = elements
return self
[docs] @cached_property
def cell(self):
result, = set(elem.cell for elem in self.elements)
return result
[docs] @cached_property
def degree(self):
return tree_map(max, *[elem.degree for elem in self.elements])
[docs] def entity_dofs(self):
'''Return the map of topological entities to degrees of
freedom for the finite element.'''
return concatenate_entity_dofs(self.cell, self.elements,
methodcaller("entity_dofs"))
[docs] @cached_property
def entity_permutations(self):
'''Return the map of topological entities to the map of
orientations to permutation lists for the finite element'''
return concatenate_entity_permutations(self.elements)
@cached_property
def _entity_support_dofs(self):
return concatenate_entity_dofs(self.cell, self.elements,
methodcaller("entity_support_dofs"))
[docs] def space_dimension(self):
'''Return the dimension of the finite element space.'''
return sum(elem.space_dimension() for elem in self.elements)
[docs] @cached_property
def index_shape(self):
return (self.space_dimension(),)
[docs] @cached_property
def value_shape(self):
'''A tuple indicating the shape of the element.'''
shape, = set(elem.value_shape for elem in self.elements)
return shape
[docs] @cached_property
def fiat_equivalent(self):
if self.is_mixed:
# EnrichedElement is actually a MixedElement
return FIAT.MixedElement([e.element.fiat_equivalent
for e in self.elements], ref_el=self.cell)
else:
return FIAT.EnrichedElement(*(e.fiat_equivalent
for e in self.elements))
[docs] @cached_property
def is_mixed(self):
# Avoid circular import dependency
from finat.mixed import MixedSubElement
return all(isinstance(e, MixedSubElement) for e in self.elements)
def _compose_evaluations(self, results):
keys, = set(map(frozenset, results))
def merge(tables):
tables = tuple(tables)
zeta = self.get_value_indices()
tensors = []
for elem, table in zip(self.elements, tables):
beta_i = elem.get_indices()
tensors.append(gem.ComponentTensor(
gem.Indexed(table, beta_i + zeta),
beta_i
))
beta = self.get_indices()
return gem.ComponentTensor(
gem.Indexed(gem.Concatenate(*tensors), beta),
beta + zeta
)
return {key: merge(result[key] for result in results)
for key in keys}
[docs] def basis_evaluation(self, order, ps, entity=None, coordinate_mapping=None):
'''Return code for evaluating the element at known points on the
reference element.
:param order: return derivatives up to this order.
:param ps: the point set object.
:param entity: the cell entity on which to tabulate.
'''
results = [element.basis_evaluation(order, ps, entity, coordinate_mapping=coordinate_mapping)
for element in self.elements]
return self._compose_evaluations(results)
[docs] def point_evaluation(self, order, refcoords, entity=None):
'''Return code for evaluating the element at an arbitrary points on
the reference element.
:param order: return derivatives up to this order.
:param refcoords: GEM expression representing the coordinates
on the reference entity. Its shape must be
a vector with the correct dimension, its
free indices are arbitrary.
:param entity: the cell entity on which to tabulate.
'''
results = [element.point_evaluation(order, refcoords, entity)
for element in self.elements]
return self._compose_evaluations(results)
@property
def mapping(self):
mappings = set(elem.mapping for elem in self.elements)
if len(mappings) != 1:
return None
else:
result, = mappings
return result
[docs]def tree_map(f, *args):
"""Like the built-in :py:func:`map`, but applies to a tuple tree."""
nonleaf, = set(isinstance(arg, tuple) for arg in args)
if nonleaf:
ndim, = set(map(len, args)) # asserts equal arity of all args
return tuple(tree_map(f, *subargs) for subargs in zip(*args))
else:
return f(*args)
[docs]def concatenate_entity_dofs(ref_el, elements, method):
"""Combine the entity DoFs from a list of elements into a combined
dict containing the information for the concatenated DoFs of all
the elements.
:arg ref_el: the reference cell
:arg elements: subelement whose DoFs are concatenated
:arg method: method to obtain the entity DoFs dict
:returns: concatenated entity DoFs dict
"""
entity_dofs = {dim: {i: [] for i in entities}
for dim, entities in ref_el.get_topology().items()}
offsets = numpy.cumsum([0] + list(e.space_dimension()
for e in elements), dtype=int)
for i, d in enumerate(map(method, elements)):
for dim, dofs in d.items():
for ent, off in dofs.items():
entity_dofs[dim][ent] += list(map(partial(add, offsets[i]), off))
return entity_dofs
[docs]def concatenate_entity_permutations(elements):
"""For each dimension, for each entity, and for each possible
entity orientation, collect the DoF permutation lists from
entity_permutations dicts of elements and concatenate them.
:arg elements: subelements whose DoF permutation lists are concatenated
:returns: entity_permutation dict of the :class:`EnrichedElement` object
composed of elements.
"""
permutations = {}
for element in elements:
for dim, e_o_p_map in element.entity_permutations.items():
dim_permutations = permutations.setdefault(dim, {})
for e, o_p_map in e_o_p_map.items():
e_dim_permutations = dim_permutations.setdefault(e, {})
for o, p in o_p_map.items():
o_e_dim_permutations = e_dim_permutations.setdefault(o, [])
offset = len(o_e_dim_permutations)
o_e_dim_permutations += list(offset + q for q in p)
return permutations
