Merge pull request #3202 from AayushSabharwal/as/tuple-observed

feat: support directly generating observed functions for tuples

Author: ChrisRackauckas

Project.toml

@@ -131,7 +131,7 @@ SimpleNonlinearSolve = "0.1.0, 1, 2"
 SparseArrays = "1"
 SpecialFunctions = "0.7, 0.8, 0.9, 0.10, 1.0, 2"
 StaticArrays = "0.10, 0.11, 0.12, 1.0"
-SymbolicIndexingInterface = "0.3.31"
+SymbolicIndexingInterface = "0.3.35"
 SymbolicUtils = "3.7"
 Symbolics = "6.15.4"
 URIs = "1"

src/systems/abstractsystem.jl

@@ -818,6 +818,8 @@ function SymbolicIndexingInterface.is_observed(sys::AbstractSystem, sym)
            !is_independent_variable(sys, sym) && symbolic_type(sym) != NotSymbolic()
 end
 
+SymbolicIndexingInterface.supports_tuple_observed(::AbstractSystem) = true
+
 function SymbolicIndexingInterface.observed(
         sys::AbstractSystem, sym; eval_expression = false, eval_module = @__MODULE__)
     if has_index_cache(sys) && (ic = get_index_cache(sys)) !== nothing
@@ -827,7 +829,8 @@ function SymbolicIndexingInterface.observed(
                 throw(ArgumentError("Symbol $sym does not exist in the system"))
             end
             sym = _sym
-        elseif sym isa AbstractArray && symbolic_type(sym) isa NotSymbolic &&
+        elseif (sym isa Tuple ||
+                (sym isa AbstractArray && symbolic_type(sym) isa NotSymbolic)) &&
                any(x -> x isa Symbol, sym)
             sym = map(sym) do s
                 if s isa Symbol

src/systems/diffeqs/odesystem.jl

@@ -427,6 +427,10 @@ function build_explicit_observed_function(sys, ts;
         param_only = false,
         op = Operator,
         throw = true)
+    is_tuple = ts isa Tuple
+    if is_tuple
+        ts = collect(ts)
+    end
     if (isscalar = symbolic_type(ts) !== NotSymbolic())
         ts = [ts]
     end
@@ -573,9 +577,16 @@ function build_explicit_observed_function(sys, ts;
 
     # Need to keep old method of building the function since it uses `output_type`,
     # which can't be provided to `build_function`
+    return_value = if isscalar
+        ts[1]
+    elseif is_tuple
+        MakeTuple(Tuple(ts))
+    else
+        MakeArray(ts, output_type)
+    end
     oop_fn = Func(args, [],
                  pre(Let(obsexprs,
-                     isscalar ? ts[1] : MakeArray(ts, output_type),
+                     return_value,
                      false))) |> array_wrapper[1] |> oop_mtkp_wrapper |> toexpr
     oop_fn = expression ? oop_fn : eval_or_rgf(oop_fn; eval_expression, eval_module)
 

test/symbolic_indexing_interface.jl

@@ -8,6 +8,7 @@ using SciMLStructures: Tunable
     eqs = [D(x) ~ a * y + t, D(y) ~ b * t]
     @named odesys = ODESystem(eqs, t, [x, y], [a, b]; observed = [xy ~ x + y])
     odesys = complete(odesys)
+    @test SymbolicIndexingInterface.supports_tuple_observed(odesys)
     @test all(is_variable.((odesys,), [x, y, 1, 2, :x, :y]))
     @test all(.!is_variable.((odesys,), [a, b, t, 3, 0, :a, :b]))
     @test variable_index.((odesys,), [x, y, a, b, t, 1, 2, :x, :y, :a, :b]) ==
@@ -33,6 +34,14 @@ using SciMLStructures: Tunable
     @test default_values(odesys)[y] == 2.0
     @test isequal(default_values(odesys)[xy], x + y)
 
+    prob = ODEProblem(odesys, [], (0.0, 1.0), [a => 1.0, b => 2.0])
+    getter = getu(odesys, (x + 1, x + 2))
+    @test getter(prob) isa Tuple
+    @test_nowarn @inferred getter(prob)
+    getter = getp(odesys, (a + 1, a + 2))
+    @test getter(prob) isa Tuple
+    @test_nowarn @inferred getter(prob)
+
     @named odesys = ODESystem(
         eqs, t, [x, y], [a, b]; defaults = [xy => 3.0], observed = [xy ~ x + y])
     odesys = complete(odesys)
@@ -99,6 +108,7 @@ end
         0 ~ x * y - β * z]
     @named ns = NonlinearSystem(eqs, [x, y, z], [σ, ρ, β])
     ns = complete(ns)
+    @test SymbolicIndexingInterface.supports_tuple_observed(ns)
     @test !is_time_dependent(ns)
     ps = ModelingToolkit.MTKParameters(ns, [σ => 1.0, ρ => 2.0, β => 3.0])
     pobs = parameter_observed(ns, σ + ρ)
@@ -107,6 +117,15 @@ end
     pobs = parameter_observed(ns, [σ + ρ, ρ + β])
     @test isempty(get_all_timeseries_indexes(ns, [σ + ρ, ρ + β]))
     @test pobs(ps) == [3.0, 5.0]
+
+    prob = NonlinearProblem(
+        ns, [x => 1.0, y => 2.0, z => 3.0], [σ => 1.0, ρ => 2.0, β => 3.0])
+    getter = getu(ns, (x + 1, x + 2))
+    @test getter(prob) isa Tuple
+    @test_nowarn @inferred getter(prob)
+    getter = getp(ns, (σ + 1, σ + 2))
+    @test getter(prob) isa Tuple
+    @test_nowarn @inferred getter(prob)
 end
 
 @testset "PDESystem" begin