style: apply runic formatting

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: SebastianM-C

lib/OptimizationBase/src/utils.jl

@@ -19,7 +19,7 @@ function apply_sense(f::OptimizationFunction{iip}, sense) where {iip}
     elseif iip
         function (G, u, p)
             f.grad(G, u, p)
-            G .*= -one(eltype(G))
+            return G .*= -one(eltype(G))
         end
     else
         (u, p) -> -f.grad(u, p)
@@ -45,7 +45,7 @@ function apply_sense(f::OptimizationFunction{iip}, sense) where {iip}
     elseif iip
         function (H, u, p)
             f.hess(H, u, p)
-            H .*= -one(eltype(H))
+            return H .*= -one(eltype(H))
         end
     else
         (u, p) -> -f.hess(u, p)
@@ -72,7 +72,7 @@ function apply_sense(f::OptimizationFunction{iip}, sense) where {iip}
     elseif iip
         function (Hv, u, v, p)
             f.hv(Hv, u, v, p)
-            Hv .*= -one(eltype(Hv))
+            return Hv .*= -one(eltype(Hv))
         end
     else
         (u, v, p) -> -f.hv(u, v, p)

lib/OptimizationMadNLP/src/OptimizationMadNLP.jl

@@ -206,7 +206,7 @@ function SciMLBase.__solve(cache::OptimizationCache{O}) where {O <: MadNLPOptimi
     results = MadNLP.solve!(solver)
 
     if cache.progress
-        Base.@logmsg Base.LogLevel(-1) progress="done" _id = :OptimizationMadNLP
+        Base.@logmsg Base.LogLevel(-1) progress = "done" _id = :OptimizationMadNLP
     end
 
     stats = OptimizationBase.OptimizationStats(;

lib/OptimizationMadNLP/test/runtests.jl

@@ -649,11 +649,11 @@ end
     cb(opt_state, loss) = (@info "obj=$loss"; opt_state.original.cnt.k >= 3)
     prob = OptimizationProblem(optfunc, x0, _p)
 
-    sol = @test_logs (:info, "obj=1.0") match_mode=:any solve(prob, MadNLPOptimizer(); callback = cb)
+    sol = @test_logs (:info, "obj=1.0") match_mode = :any solve(prob, MadNLPOptimizer(); callback = cb)
     @test sol.stats.iterations == 3
 
     logs, _ = Test.collect_test_logs(min_level = Base.LogLevel(-1)) do
-        solve(prob, MadNLPOptimizer(), progress = true, maxiters=10)
+        solve(prob, MadNLPOptimizer(), progress = true, maxiters = 10)
     end
     @test !isempty(logs)
     m = first(logs)

lib/OptimizationOptimJL/test/runtests.jl

@@ -39,15 +39,15 @@ end
 
         prob_min = OptimizationProblem(optf, x0, nothing; lb = lb, ub = ub)
         sol_min = solve(prob_min, LBFGS(); x_abstol = 0.1)
-        @test isapprox(sol_min.u, [0.0, 0.0]; atol = 1e-2)
-        @test isapprox(sol_min.objective, 0.0; atol = 1e-2)
+        @test isapprox(sol_min.u, [0.0, 0.0]; atol = 1.0e-2)
+        @test isapprox(sol_min.objective, 0.0; atol = 1.0e-2)
 
         prob_max = OptimizationProblem(
             optf, x0, nothing; lb = lb, ub = ub, sense = OptimizationBase.MaxSense
         )
         sol_max = solve(prob_max, LBFGS(); x_abstol = 0.1)
-        @test isapprox(sol_max.u, [10.0, 10.0]; atol = 1e-2)
-        @test isapprox(sol_max.objective, 20.0; atol = 1e-2)
+        @test isapprox(sol_max.u, [10.0, 10.0]; atol = 1.0e-2)
+        @test isapprox(sol_max.objective, 20.0; atol = 1.0e-2)
     end
 
     @testset "MaxSense with IPNewton" begin
@@ -61,8 +61,8 @@ end
             optf, x0, nothing; lb = lb, ub = ub, sense = OptimizationBase.MaxSense
         )
         sol_max = solve(prob_max, Optim.IPNewton(); x_abstol = 0.1)
-        @test isapprox(sol_max.u, [10.0, 10.0]; atol = 1e-2)
-        @test isapprox(sol_max.objective, 20.0; atol = 1e-2)
+        @test isapprox(sol_max.u, [10.0, 10.0]; atol = 1.0e-2)
+        @test isapprox(sol_max.objective, 20.0; atol = 1.0e-2)
     end
 
     rosenbrock(x, p) = (p[1] - x[1])^2 + p[2] * (x[2] - x[1]^2)^2

lib/OptimizationOptimisers/test/runtests.jl

@@ -18,10 +18,10 @@ using Lux, MLUtils, Random, ComponentArrays, Printf, MLDataDevices
         prob_max = OptimizationProblem(optf, x0, nothing; sense = OptimizationBase.MaxSense)
         cache = OptimizationBase.OptimizationCache(prob_max, Optimisers.Adam())
         G = zeros(2)
-        @test isapprox(cache.f.f(x0, nothing), -10.0; atol = 1e-12)
+        @test isapprox(cache.f.f(x0, nothing), -10.0; atol = 1.0e-12)
         cache.f.grad(G, x0, nothing)
-        @test isapprox(G[1], -1.0; atol = 1e-12)
-        @test isapprox(G[2], -1.0; atol = 1e-12)
+        @test isapprox(G[1], -1.0; atol = 1.0e-12)
+        @test isapprox(G[2], -1.0; atol = 1.0e-12)
     end
 
     rosenbrock(x, p) = (p[1] - x[1])^2 + p[2] * (x[2] - x[1]^2)^2

lib/SimpleOptimization/src/SimpleOptimization.jl

@@ -252,7 +252,7 @@ function SciMLBase.__solve(cache::OptimizationCache{O}) where {O <: SimpleGradie
 
     abstol = cache.solver_args.abstol
     if abstol === nothing
-        abstol = 1e-8
+        abstol = 1.0e-8
     end
 
     f = Base.Fix2(cache.f.f, cache.p)

test/sense_tests.jl

@@ -60,7 +60,8 @@ const P0 = nothing
         H[2, 2] = 0.0
     end
 
-    optf = OptimizationFunction(obj;
+    optf = OptimizationFunction(
+        obj;
         grad = grad!, fg = fg!, hess = hess!,
         fgh = fgh!, hv = hv!, lag_h = lag_h!
     )