Merge pull request #504 from MilesCranmer/format-v2

MilesCranmer · web-flow · commit 137db25c2af3 · 2025-09-07T18:21:54.000+01:00
style: update to formatter version 2
diff --git a/.JuliaFormatter.toml b/.JuliaFormatter.toml
@@ -1,3 +1,4 @@
 style = "blue"
 ignore = ["docs"]
+# Keep pipe operators as they are more readable
 pipe_to_function_call = false
diff --git a/.github/workflows/check-format.yml b/.github/workflows/check-format.yml
@@ -25,7 +25,7 @@ jobs:
         uses: julia-actions/cache@v2
       - name: Install JuliaFormatter and format
         run: |
-          julia --startup-file=no -e 'using Pkg; pkg"activate --temp"; pkg"add JuliaFormatter@1.0.61"; using JuliaFormatter; format("."; verbose=true)'
+          julia --startup-file=no -e 'using Pkg; pkg"activate --temp"; pkg"add JuliaFormatter@2"; using JuliaFormatter; format("."; verbose=true)'
       - name: "Format check"
         run: |
           julia -e '
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -11,7 +11,7 @@ repos:
       - id: check-yaml
       - id: check-added-large-files
   - repo: https://github.com/domluna/JuliaFormatter.jl
-    rev: v1.0.61
+    rev: v2.1.6
     hooks:
       - id: julia-formatter
   - repo: https://github.com/pre-commit/mirrors-prettier
diff --git a/benchmark/benchmarks.jl b/benchmark/benchmarks.jl
@@ -88,10 +88,10 @@ function create_utils_benchmark()
     suite["best_of_sample"] = @benchmarkable(
         best_of_sample(pop, rss, $options),
         setup = (
-            nfeatures = 1;
-            dataset = Dataset(randn(nfeatures, 32), randn(32));
-            pop = Population(dataset; npop=100, nlength=20, options=$options, nfeatures);
-            rss = RunningSearchStatistics(; options=$options)
+            nfeatures=1;
+            dataset=Dataset(randn(nfeatures, 32), randn(32));
+            pop=Population(dataset; npop=100, nlength=20, options=($options), nfeatures);
+            rss=RunningSearchStatistics(; options=($options))
         )
     )
 
@@ -110,9 +110,9 @@ function create_utils_benchmark()
             end
         end,
         setup = (
-            nfeatures = 1;
-            dataset = Dataset(randn(nfeatures, 32), randn(32));
-            mutation_weights = MutationWeights(;
+            nfeatures=1;
+            dataset=Dataset(randn(nfeatures, 32), randn(32));
+            mutation_weights=MutationWeights(;
                 mutate_constant=1.0,
                 mutate_operator=1.0,
                 swap_operands=1.0,
@@ -125,21 +125,23 @@ function create_utils_benchmark()
                 form_connection=0.0,
                 break_connection=0.0,
             );
-            options = Options(;
-                unary_operators=[sin, cos], binary_operators=[+, -, *, /], mutation_weights
+            options=Options(;
+                unary_operators=[sin, cos],
+                binary_operators=[+, -, *, /],
+                mutation_weights,
             );
-            recorder = RecordType();
-            temperature = 1.0;
-            curmaxsize = 20;
-            rss = RunningSearchStatistics(; options);
-            trees = [
+            recorder=RecordType();
+            temperature=1.0;
+            curmaxsize=20;
+            rss=RunningSearchStatistics(; options);
+            trees=[
                 gen_random_tree_fixed_size(15, options, nfeatures, Float64) for _ in 1:100
             ];
-            expressions = [
+            expressions=[
                 Expression(tree; operators=options.operators, variable_names=["x1"]) for
                 tree in trees
             ];
-            members = [
+            members=[
                 PopMember(dataset, expression, options; deterministic=false) for
                 expression in expressions
             ]
@@ -153,14 +155,14 @@ function create_utils_benchmark()
         end,
         seconds = 20,
         setup = (
-            nfeatures = 1;
-            T = Float64;
-            dataset = Dataset(randn(nfeatures, 512), randn(512));
-            ntrees = $ntrees;
-            trees = [
+            nfeatures=1;
+            T=Float64;
+            dataset=Dataset(randn(nfeatures, 512), randn(512));
+            ntrees=($ntrees);
+            trees=[
                 gen_random_tree_fixed_size(20, $options, nfeatures, T) for i in 1:ntrees
             ];
-            members = [
+            members=[
                 PopMember(dataset, tree, $options; deterministic=false) for tree in trees
             ]
         )
@@ -179,9 +181,9 @@ function create_utils_benchmark()
                     compute_complexity(tree, $options)
                 end,
                 setup = (
-                    T = Float64;
-                    nfeatures = 3;
-                    trees = [
+                    T=Float64;
+                    nfeatures=3;
+                    trees=[
                         gen_random_tree_fixed_size(20, $options, nfeatures, T) for
                         i in 1:($ntrees)
                     ]
@@ -197,9 +199,9 @@ function create_utils_benchmark()
                 SymbolicRegression.MutationFunctionsModule.randomly_rotate_tree!(tree)
             end,
             setup = (
-                T = Float64;
-                nfeatures = 3;
-                trees = [
+                T=Float64;
+                nfeatures=3;
+                trees=[
                     gen_random_tree_fixed_size(20, $options, nfeatures, T) for
                     i in 1:($ntrees)
                 ]
@@ -214,9 +216,9 @@ function create_utils_benchmark()
             )
         end,
         setup = (
-            T = Float64;
-            nfeatures = 3;
-            trees = [
+            T=Float64;
+            nfeatures=3;
+            trees=[
                 gen_random_tree_fixed_size(20, $options, nfeatures, T) for i in 1:($ntrees)
             ]
         )
@@ -240,9 +242,9 @@ function create_utils_benchmark()
             check_constraints(tree, $options, $options.maxsize)
         end,
         setup = (
-            T = Float64;
-            nfeatures = 3;
-            trees = [
+            T=Float64;
+            nfeatures=3;
+            trees=[
                 gen_random_tree_fixed_size(20, $options, nfeatures, T) for i in 1:($ntrees)
             ]
         )
diff --git a/src/CheckConstraints.jl b/src/CheckConstraints.jl
@@ -91,8 +91,8 @@ function check_constraints(
     return true
 end
 
-check_constraints(
-    ex::Union{AbstractExpression,AbstractExpressionNode}, options::AbstractOptions
-)::Bool = check_constraints(ex, options, options.maxsize)
+check_constraints(ex::Union{AbstractExpression,AbstractExpressionNode}, options::AbstractOptions)::Bool = check_constraints(
+    ex, options, options.maxsize
+)
 
 end
diff --git a/src/ComposableExpression.jl b/src/ComposableExpression.jl
@@ -271,8 +271,9 @@ function apply_operator(op::F, x::Vararg{Any,N}) where {F<:Function,N}
     if all(_is_valid, x)
         return _apply_operator(op, x...)
     else
-        example_vector =
-            something(map(xi -> xi isa ValidVector ? xi : nothing, x)...)::ValidVector
+        example_vector = something(
+            map(xi -> xi isa ValidVector ? xi : nothing, x)...
+        )::ValidVector
         expected_return_type = Base.promote_op(
             _apply_operator, typeof(op), map(typeof, x)...
         )
diff --git a/src/Configure.jl b/src/Configure.jl
@@ -243,12 +243,10 @@ function activate_env_on_workers(
 )
     verbosity > 0 && @info "Activating environment on workers."
     @everywhere procs begin
-        Base.MainInclude.eval(
-            quote
-                using Pkg
-                Pkg.activate($$project_path)
-            end,
-        )
+        Base.MainInclude.eval(quote
+            using Pkg
+            Pkg.activate($$project_path)
+        end)
     end
 end
 
@@ -291,12 +289,9 @@ function import_module_on_workers(
     all_extensions = vcat(relevant_extensions, @something(worker_imports, Symbol[]))
 
     for ext in all_extensions
-        push!(
-            expr.args,
-            quote
-                using $ext: $ext
-            end,
-        )
+        push!(expr.args, quote
+            using $ext: $ext
+        end)
     end
 
     verbosity > 0 && if isempty(relevant_extensions)
diff --git a/src/ExpressionBuilder.jl b/src/ExpressionBuilder.jl
@@ -17,7 +17,7 @@ import DynamicExpressions: get_operators
 import ..CoreModule: create_expression
 
 @unstable function create_expression(
-    t::T, options::AbstractOptions, dataset::Dataset{T,L}, ::Val{embed}=Val(false)
+    t::T, options::AbstractOptions, dataset::Dataset{T,L}, (::Val{embed})=Val(false)
 ) where {T,L,embed}
     return create_expression(
         t, options, dataset, options.node_type, options.expression_type, Val(embed)
@@ -27,7 +27,7 @@ end
     t::AbstractExpressionNode{T},
     options::AbstractOptions,
     dataset::Dataset{T,L},
-    ::Val{embed}=Val(false),
+    (::Val{embed})=Val(false),
 ) where {T,L,embed}
     return create_expression(
         t, options, dataset, options.node_type, options.expression_type, Val(embed)
@@ -37,7 +37,7 @@ function create_expression(
     ex::AbstractExpression{T},
     options::AbstractOptions,
     ::Dataset{T,L},
-    ::Val{embed}=Val(false),
+    (::Val{embed})=Val(false),
 ) where {T,L,embed}
     return ex::options.expression_type
 end
@@ -47,7 +47,7 @@ end
     dataset::Dataset{T,L},
     ::Type{N},
     ::Type{E},
-    ::Val{embed}=Val(false),
+    (::Val{embed})=Val(false),
 ) where {T,L,embed,N<:AbstractExpressionNode,E<:AbstractExpression}
     return create_expression(constructorof(N)(; val=t), options, dataset, N, E, Val(embed))
 end
@@ -57,7 +57,7 @@ end
     dataset::Dataset{T,L},
     ::Type{<:AbstractExpressionNode},
     ::Type{E},
-    ::Val{embed}=Val(false),
+    (::Val{embed})=Val(false),
 ) where {T,L,embed,E<:AbstractExpression}
     return constructorof(E)(t; init_params(options, dataset, nothing, Val(embed))...)
 end
diff --git a/src/MLJInterface.jl b/src/MLJInterface.jl
@@ -331,7 +331,7 @@ function _update(
         variable_names=variable_names,
         y_variable_names=y_variable_names,
         y_is_table=MMI.istable(y),
-        has_class=!isnothing(class),
+        has_class=(!isnothing(class)),
         X_units=X_units_clean,
         y_units=y_units_clean,
         types=SRFitResultTypes(;
@@ -595,9 +595,9 @@ end
 function get_equation_strings_for(
     ::AbstractSingletargetSRRegressor, trees, options, variable_names
 )
-    return (
-        t -> string_tree(t, options; variable_names=variable_names, pretty=false)
-    ).(trees)
+    return (t -> string_tree(t, options; variable_names=variable_names, pretty=false)).(
+        trees
+    )
 end
 function get_equation_strings_for(
     ::AbstractMultitargetSRRegressor, trees, options, variable_names
@@ -671,7 +671,7 @@ for model in [:SRRegressor, :SRTestRegressor]
             target_scitype=AbstractVector{<:MMI.Continuous},
             supports_weights=true,
             reports_feature_importances=false,
-            load_path=$("SymbolicRegression.MLJInterfaceModule." * string(model)),
+            load_path=($("SymbolicRegression.MLJInterfaceModule." * string(model))),
             human_name="Symbolic Regression via Evolutionary Search",
         )
     end
@@ -686,7 +686,7 @@ for model in [:MultitargetSRRegressor, :MultitargetSRTestRegressor]
             },
             supports_weights=true,
             reports_feature_importances=false,
-            load_path=$("SymbolicRegression.MLJInterfaceModule." * string(model)),
+            load_path=($("SymbolicRegression.MLJInterfaceModule." * string(model))),
             human_name="Multi-Target Symbolic Regression via Evolutionary Search",
         )
     end
diff --git a/src/Options.jl b/src/Options.jl
@@ -150,7 +150,8 @@ end
                         break
                     end
                 end
-                found_degree == 0 && error("Operator $(op) is not in the operator set.")
+                found_degree == 0 &&
+                    error("Operator $(op) is not in the operator set.")
                 (found_degree, found_idx)
             end,
             new_max_nesting_dict = [
@@ -166,7 +167,7 @@ end
                             end
                         end
                         found_degree == 0 &&
-                        error("Operator $(nested_op) is not in the operator set.")
+                            error("Operator $(nested_op) is not in the operator set.")
                         (found_degree, found_idx)
                     end
                     (nested_degree, nested_idx, max_nesting)
diff --git a/src/OptionsStruct.jl b/src/OptionsStruct.jl
@@ -113,9 +113,11 @@ Controls level of specialization we compile into `Options`.
 
 Overload if needed for custom expression types.
 """
-operator_specialization(
+function operator_specialization(
     ::Type{O}, ::Type{<:AbstractExpression}
-) where {O<:AbstractOperatorEnum} = O
+) where {O<:AbstractOperatorEnum}
+    return O
+end
 @unstable operator_specialization(::Type{<:OperatorEnum}, ::Type{<:AbstractExpression}) =
     OperatorEnum
 
diff --git a/src/SearchUtils.jl b/src/SearchUtils.jl
@@ -37,18 +37,16 @@ to avoid spam when worker processes exit normally.
 macro filtered_async(expr)
     return esc(
         quote
-            $(Base).errormonitor(
-                @async begin
-                    try
-                        $expr
-                    catch ex
-                        if !(ex isa $(Distributed).ProcessExitedException)
-                            rethrow(ex)
-                        end
+            $(Base).errormonitor(@async begin
+                try
+                    $expr
+                catch ex
+                    if !(ex isa $(Distributed).ProcessExitedException)
+                        rethrow(ex)
                     end
                 end
-            )
-        end,
+            end)
+        end
     )
 end
 
diff --git a/src/SymbolicRegression.jl b/src/SymbolicRegression.jl
@@ -904,9 +904,9 @@ function _main_search_loop!(
     resource_monitor = ResourceMonitor(;
         # Storing n times as many monitoring intervals as populations seems like it will
         # help get accurate resource estimates:
-        max_recordings=options.populations * 100 * nout,
-        start_reporting_at=options.populations * 3 * nout,
-        window_size=options.populations * 2 * nout,
+        max_recordings=(options.populations * 100 * nout),
+        start_reporting_at=(options.populations * 3 * nout),
+        window_size=(options.populations * 2 * nout),
     )
     while sum(state.cycles_remaining) > 0
         kappa += 1
diff --git a/src/TemplateExpression.jl b/src/TemplateExpression.jl
@@ -503,7 +503,7 @@ function EB.create_expression(
     dataset::Dataset{T,L},
     ::Type{<:AbstractExpressionNode},
     ::Type{E},
-    ::Val{embed}=Val(false),
+    (::Val{embed})=Val(false),
 ) where {T,L,embed,E<:TemplateExpression}
     function_keys = get_function_keys(options.expression_options.structure)
 
diff --git a/src/TemplateExpressionMacro.jl b/src/TemplateExpressionMacro.jl
diff --git a/src/Utils.jl b/src/Utils.jl
diff --git a/test/manual_distributed.jl b/test/manual_distributed.jl
diff --git a/test/test_mlj.jl b/test/test_mlj.jl
diff --git a/test/test_units.jl b/test/test_units.jl
diff --git a/test/test_utils.jl b/test/test_utils.jl
