Work on Tensor Products

.github/workflows/test.yml

@@ -24,7 +24,7 @@ jobs:
         arch: [default]
     runs-on: [self-hosted, Linux]
     container:
-      image: ubuntu:latest
+      image: ubuntu:25.10
     env:
       OMPI_ALLOW_RUN_AS_ROOT: 1
       OMPI_ALLOW_RUN_AS_ROOT_CONFIRM: 1

Makefile

@@ -26,6 +26,12 @@ tests:
 	@echo "    Running all tests"
 	@FIREDRAKE_USE_FUSE=1 python3 -m coverage run -p -m pytest -rx test
 
+mini_tests:
+	@FIREDRAKE_USE_FUSE=1 python3 -m pytest test/test_2d_examples_docs.py
+	@FIREDRAKE_USE_FUSE=1 python3 -m pytest test/test_convert_to_fiat.py::test_1d
+	@FIREDRAKE_USE_FUSE=1 python3 -m pytest test/test_orientations.py::test_surface_vec_rt
+	@FIREDRAKE_USE_FUSE=1 python3 -m pytest test/test_convert_to_fiat.py::test_projection_convergence_3d\[construct_tet_ned-N1curl-1-0.8\]
+
 coverage:
 	@python3 -m coverage combine
 	@python3 -m coverage report -m 

fuse/__init__.py

@@ -1,9 +1,10 @@
-from fuse.cells import Point, Edge, polygon, make_tetrahedron, constructCellComplex
+
+from fuse.cells import Point, Edge, polygon, line, make_tetrahedron, constructCellComplex, TensorProductPoint
 from fuse.groups import S1, S2, S3, D4, Z3, Z4, C3, C4, S4, A4, diff_C3, tet_edges, tet_faces, sq_edges, GroupRepresentation, PermutationSetRepresentation, get_cyc_group, get_sym_group
 from fuse.dof import DeltaPairing, DOF, L2Pairing, FuseFunction, PointKernel, VectorKernel, BarycentricPolynomialKernel, PolynomialKernel, ComponentKernel
 from fuse.triples import ElementTriple, DOFGenerator, immerse
 from fuse.traces import TrH1, TrGrad, TrHess, TrHCurl, TrHDiv
-from fuse.tensor_products import tensor_product
+from fuse.tensor_products import tensor_product, symmetric_tensor_product
 
 from fuse.spaces.element_sobolev_spaces import CellH1, CellL2, CellHDiv, CellHCurl, CellH2
 from fuse.spaces.polynomial_spaces import P0, P1, P2, P3, Q2, PolynomialSpace

fuse/cells.py

@@ -151,6 +151,13 @@
         raise ValueError("Dimension {} not supported".format(d))
 
 
+def line():
+    """
+    Constructs the default 1D interval
+    """
+    return Point(1, [Point(0), Point(0)], vertex_num=2)
+
+
 def polygon(n):
     """
     Constructs the 2D default cell with n sides/vertices
@@ -373,6 +380,7 @@
         """
         verts = self.ordered_vertices()
         v_coords = [self.get_node(v, return_coords=True) for v in verts]
+
         n = len(verts)
         max_group = SymmetricGroup(n)
         edges = [edge.ordered_vertices() for edge in self.edges()]
@@ -502,6 +510,15 @@
         min_ids = [min(dimension) for dimension in structure]
         return min_ids
 
+    def local_id(self, node):
+        structure = [sorted(generation) for generation in nx.topological_generations(self.G)]
+        structure.reverse()
+        min_id = self.get_starter_ids()
+        for d in range(len(structure)):
+            if node.id in structure[d]:
+                return node.id - min_id[d]
+        raise ValueError("Node not found in cell")
+
     def graph_dim(self):
         if self.oriented:
             dim = self.dimension + 1
@@ -648,7 +665,6 @@
         self_levels = [generation for generation in nx.topological_generations(self.G)]
         vertices = entity.ordered_vertices()
         if self.dimension == 0:
-            # return [[]
             raise ValueError("Dimension 0 entities cannot have Basis Vectors")
         if self.oriented:
             # ordered_vertices() handles the orientation so we want to drop the orientation node
@@ -822,7 +838,7 @@
 
         return lambda *x: fold_reduce(attachments[0], *x)
 
-    def attachment_J(self, source, dst):
+    def attachment_J_det(self, source, dst):
         attachment = self.attachment(source, dst)
         symbol_names = ["x", "y", "z"]
         symbols = []
@@ -831,7 +847,7 @@
         for i in range(self.dim_of_node(dst)):
             symbols += [sp.Symbol(symbol_names[i])]
         J = sp.Matrix(attachment(*symbols)).jacobian(sp.Matrix(symbols))
-        return J
+        return np.sqrt(abs(float(sp.det(J.T * J))))
 
     def quadrature(self, degree):
         fiat_el = self.to_fiat()
@@ -903,6 +919,13 @@
     def _from_dict(o_dict):
         return Point(o_dict["dim"], o_dict["edges"], oriented=o_dict["oriented"], cell_id=o_dict["id"])
 
+    def equivalent(self, other):
+        if self.dimension != other.dimension:
+            return False
+        if set(self.ordered_vertex_coords()) != set(other.ordered_vertex_coords()):
+            return False
+        return self.get_topology() == other.get_topology()
+
 
 class Edge():
     """
@@ -926,7 +949,12 @@
             if hasattr(self.attachment, '__iter__'):
                 res = []
                 for attach_comp in self.attachment:
-                    res.append(sympy_to_numpy(attach_comp, syms, x))
+                    if len(attach_comp.atoms(sp.Symbol)) <= len(x):
+                        res.append(sympy_to_numpy(attach_comp, syms, x))
+                    else:
+                        res_val = attach_comp.subs({syms[i]: x[i] for i in range(len(x))})
+                        res.append(res_val)
+
                 return tuple(res)
             return sympy_to_numpy(self.attachment, syms, x)
         return x
@@ -957,50 +985,198 @@
 
 
 class TensorProductPoint():
+    id_iter = itertools.count()
 
-    def __init__(self, A, B, flat=False):
+    def __init__(self, A, B):
+        self.id = next(self.id_iter)
         self.A = A
         self.B = B
         self.dimension = self.A.dimension + self.B.dimension
-        self.flat = flat
+        self.flat = False
+        self.fiat_elem = None
+        self.group = self.compute_cell_group()
+        self.entities = {}
+        for d in [(a_d, b_d) for a_d in range(self.A.dimension + 1) for b_d in range(self.B.dimension + 1)]:
+            if d == (self.A.dimension, self.B.dimension):
+                self.entities[d] = [self]
+            else:
+                self.entities[d] = [TensorProductPoint(e_a, e_b) for e_a in self.A.d_entities(d[0], True) for e_b in self.B.d_entities(d[1], True)]
 
     def ordered_vertices(self):
         return self.A.ordered_vertices() + self.B.ordered_vertices()
 
+    def component_orientations(self):
+        from fuse.utils import orientation_value
+        self.component_os_to_os = {}
+        for dim in self.to_fiat().get_topology():
+            self.component_os_to_os[dim] = {}
+            a_ent = self.A.d_entities(dim[0])[0]
+            b_ent = self.B.d_entities(dim[1])[0]
+            verts = [(v_a, v_b) for v_a in a_ent.vertices() for v_b in b_ent.vertices()]
+            ident = [i for i in range(len(verts))]
+            group = [(g_a, g_b) for g_a in a_ent.group.members() for g_b in b_ent.group.members()]
+            for g_a, g_b in group:
+                new_verts = [(v_a, v_b) for v_a in g_a.permute(a_ent.vertices()) for v_b in g_b.permute(b_ent.vertices())]
+                perm = [verts.index(v) for v in new_verts]
+                o_val = orientation_value(ident, perm)
+                if sum(dim) == self.dimension and self.group.group_rep_numbering is not None:
+                    o_val = self.group.group_rep_numbering[o_val]
+                self.component_os_to_os[dim][(g_a.numeric_rep(), g_b.numeric_rep())] = o_val
+        return self.component_os_to_os
+
+    def compute_cell_group(self):
+        """
+        Systematically work out the symmetry group of the tensor product cell.
+        """
+        verts = self.vertices()
+        group = [(g_a, g_b) for g_a in self.A.group.members() for g_b in self.B.group.members()]
+        perms = []
+        for g_a, g_b in group:
+            new_verts = [(v_a, v_b) for v_a in g_a.permute(self.A.vertices()) for v_b in g_b.permute(self.B.vertices())]
+            perm = [verts.index(v) for v in new_verts]
+            perms += [fuse_groups.Permutation(perm)]
+
+        grp = fuse_groups.PermutationSetRepresentation(perms).add_cell(self)
+        return grp
+
+    def get_starter_ids(self):
+        # this doesn't actually make sense - remove when confirmed all changes to eliminate min ids from triple is done
+        a_starts = self.A.get_starter_ids()
+        b_starts = self.B.get_starter_ids()
+        ids = []
+        for a, b in zip(a_starts, b_starts):
+            ids += [max(a, b)]
+        return ids
+
     def get_spatial_dimension(self):
         return self.dimension
 
     def get_sub_entities(self):
-        self.A.get_sub_entities()
-        self.B.get_sub_entities()
+        return self.to_fiat().sub_entities
 
-    def dimension(self):
-        return tuple(self.A.dimension, self.B.dimension)
+    def dim(self):
+        return (self.A.dimension, self.B.dimension)
 
     def d_entities(self, d, get_class=True):
+        if isinstance(d, tuple):
+            if get_class:
+                return self.entities[d]
+            return [e.id for e in self.entities[d]]
+        raise NotImplementedError("not sure this is right")
         return self.A.d_entities(d, get_class) + self.B.d_entities(d, get_class)
 
     def vertices(self, get_class=True, return_coords=False):
         # TODO maybe refactor with get_node
-        verts = self.d_entities(0, get_class)
         if return_coords:
             a_verts = self.A.vertices(return_coords=return_coords)
             b_verts = self.B.vertices(return_coords=return_coords)
             return [a + b for a in a_verts for b in b_verts]
-        return verts
+        return [(a, b) for a in self.A.vertices() for b in self.B.vertices()]
 
     def to_ufl(self, name=None):
-        if self.flat:
-            return CellComplexToUFL(self, "quadrilateral")
         return TensorProductCell(self.A.to_ufl(), self.B.to_ufl())
 
     def to_fiat(self, name=None):
-        if self.flat:
-            return CellComplexToFiatHypercube(self, CellComplexToFiatTensorProduct(self, name))
-        return CellComplexToFiatTensorProduct(self, name)
+        if self.fiat_elem is None:
+            self.fiat_elem = CellComplexToFiatTensorProduct(self, name)
+        return self.fiat_elem
+
+    def flatten(self):
+        assert self.A.equivalent(self.B)
+        return FlattenedPoint(self.A, self.B)
+
+
+class FlattenedPoint(Point, TensorProductPoint):
+    d_entities_by_total_d = Point.d_entities
+
+    def __init__(self, A, B):
+        self.A = A
+        self.B = B
+        self.dimension = self.A.dimension + self.B.dimension
+        self.flat = True
+        fuse_edges = self.construct_fuse_rep()
+        super().__init__(self.dimension, fuse_edges)
+
+    def to_ufl(self, name=None):
+        return CellComplexToUFL(self, "quadrilateral")
+
+    def to_fiat(self, name=None):
+        # TODO this should check if it actually is a hypercube
+        fiat = CellComplexToFiatHypercube(self, CellComplexToFiatTensorProduct(self, name))
+        return fiat
+
+    def d_entities(self, d, get_class=True):
+        if isinstance(d, tuple):
+            if not get_class:
+                return [p.id for p in self.all_subpoints[d]]
+            return self.all_subpoints[d]
+        return self.d_entities_by_total_d(d, get_class)
+
+
+    def construct_fuse_rep(self):
+        sub_cells = [self.A, self.B]
+        dims = (self.A.dimension, self.B.dimension)
+        if sum(dims) > 2:
+            raise NotImplementedError("Flattening 3D tensor products not yet implemented")
+
+        self.all_subpoints = {(a_d, b_d): [] for a_d in range(self.A.dimension + 1) for b_d in range(self.B.dimension + 1)}
+        points = {cell: {i: [] for i in range(max(dims) + 1)} for cell in sub_cells}
+        attachments = {cell: {i: [] for i in range(max(dims) + 1)} for cell in sub_cells}
+
+        for d in range(max(dims) + 1):
+            for cell in sub_cells:
+                if d <= cell.dimension:
+                    sub_ent = cell.d_entities(d, get_class=True)
+                    points[cell][d].extend(sub_ent)
+                    for s in sub_ent:
+                        attachments[cell][d].extend(s.connections)
+
+        prod_points = list(itertools.product(*[points[cell][0] for cell in sub_cells]))
+        # temp = prod_points[1]
+        # prod_points[1] = prod_points[2]
+        # prod_points[2] = temp
+        point_cls = [Point(0) for i in range(len(prod_points))]
+        self.all_subpoints[(0,0)] = point_cls
+        edges = []
+
+        # generate edges of tensor product result
+        for a in prod_points:
+            for b in prod_points:
+                # of all combinations of point, take those where at least one changes and at least one is the same
+                if any(a[i] == b[i] for i in range(len(a))) and any(a[i] != b[i] for i in range(len(sub_cells))):
+                    # ensure if they change, that edge exists in the existing topology
+                    if all([a[i] == b[i] or (sub_cells[i].local_id(a[i]), sub_cells[i].local_id(b[i])) in list(sub_cells[i]._topology[1].values()) for i in range(len(sub_cells))]):
+                        edges.append((a, b))
+        # hasse level 1
+        edge_cls1 = {e: None for e in edges}
+        for i in range(len(sub_cells)):
+            for (a, b) in edges:
+                a_idx = prod_points.index(a)
+                b_idx = prod_points.index(b)
+                if a[i] != b[i]:
+                    a_edge = [att for att in attachments[sub_cells[i]][1] if att.point == a[i]][0]
+                    b_edge = [att for att in attachments[sub_cells[i]][1] if att.point == b[i]][0]
+                    edge_cls1[(a, b)] = Point(1, [Edge(point_cls[a_idx], a_edge.attachment, a_edge.o),
+                                                  Edge(point_cls[b_idx], b_edge.attachment, b_edge.o)])
+                    edge_tuple_dim = tuple(int((sub_cells[i].local_id(a[i]), sub_cells[i].local_id(b[i])) in list(sub_cells[i]._topology[1].values())) for i in range(len(sub_cells)))
+                    self.all_subpoints[edge_tuple_dim] += [edge_cls1[(a, b)]]
+        edge_cls2 = []
+        # hasse level 2
+        for i in range(len(sub_cells)):
+            for (a, b) in edges:
+                if a[i] == b[i]:
+                    x = sp.Symbol("x")
+                    a_edge = [att for att in attachments[sub_cells[i]][1] if att.point == a[i]][0]
+                    if i == 0:
+                        attach = (x,) + a_edge.attachment
+                    else:
+                        attach = a_edge.attachment + (x,)
+                    edge_cls2.append(Edge(edge_cls1[(a, b)], attach, a_edge.o))
+        self.all_subpoints[(1, 1)] = [self]
+        return edge_cls2
 
     def flatten(self):
-        return TensorProductPoint(self.A, self.B, True)
+        return self
 
 
 class CellComplexToFiatSimplex(Simplex):
@@ -1190,9 +1366,8 @@
         return polygon(3).to_ufl(name)
         # return ufc_triangle().to_ufl(name)
     elif name == "quadrilateral":
-        interval = Point(1, [Point(0), Point(0)], vertex_num=2)
-        return TensorProductPoint(interval, interval).flatten().to_ufl(name)
-        # return ufc_quad().to_ufl(name)
+        return TensorProductPoint(line(), line()).flatten().to_ufl(name)
+        # return firedrake_quad().to_ufl(name)
         # return polygon(4).to_ufl(name)
     elif name == "tetrahedron":
         # return ufc_tetrahedron().to_ufl(name)