Add and apply .JuliaFormatter.toml with SciML style

Author: jacobusmmsmit

.JuliaFormatter.toml

@@ -1 +1 @@
-always_for_in = true
+style = "sciml"

benchmark/agents.jl

@@ -54,33 +54,32 @@ end
 
 mutable struct ContinuousAgent <: AbstractAgent
     id::Int
-    pos::NTuple{3,Float64}
-    vel::NTuple{3,Float64}
+    pos::NTuple{3, Float64}
+    vel::NTuple{3, Float64}
     one::Float64
     two::Bool
 end
 
 mutable struct ContinuousAgentTwo <: AbstractAgent
     id::Int
-    pos::NTuple{3,Float64}
-    vel::NTuple{3,Float64}
+    pos::NTuple{3, Float64}
+    vel::NTuple{3, Float64}
 end
 
 mutable struct ContinuousAgentThree <: AbstractAgent
     id::Int
-    pos::NTuple{3,Float64}
-    vel::NTuple{3,Float64}
+    pos::NTuple{3, Float64}
+    vel::NTuple{3, Float64}
 end
 
 mutable struct ContinuousAgentFour <: AbstractAgent
     id::Int
-    pos::NTuple{3,Float64}
-    vel::NTuple{3,Float64}
+    pos::NTuple{3, Float64}
+    vel::NTuple{3, Float64}
 end
 
 mutable struct ContinuousAgentFive <: AbstractAgent
     id::Int
-    pos::NTuple{3,Float64}
-    vel::NTuple{3,Float64}
+    pos::NTuple{3, Float64}
+    vel::NTuple{3, Float64}
 end
-

benchmark/benchmarks.jl

@@ -26,15 +26,15 @@ function ensemble_benchmark(f, parallel, nreplicates)
         adf, mdf, _ = ensemblerun!(models, schelling_agent_step!, dummystep, nsteps;
                                    parallel, adata = [:pos, :mood, :group],
                                    showprogress = true,
-                                   when = (model, step) ->
-                                    ( (step) % whensteps == 0  ||  step == 0 ),
+                                   when = (model, step) -> ((step) % whensteps == 0 ||
+                                                            step == 0),
                                    mdata = [:min_to_be_happy])
     else
         # TODO: Why do we need `replicate_idx` here?
         # Can't we just use the `Models.schelling`?
         function initialize(;
-            replicate_idx = 1, numagents = 320, griddims = (20, 20), min_to_be_happy = 3
-        )
+                            replicate_idx = 1, numagents = 320, griddims = (20, 20),
+                            min_to_be_happy = 3)
             space = GridSpace(griddims, periodic = false)
             properties = Dict(:min_to_be_happy => min_to_be_happy)
 
@@ -49,17 +49,14 @@ function ensemble_benchmark(f, parallel, nreplicates)
             return model
         end
 
-        parameters = Dict(
-            :numagents => collect(numagents_low:numagents_high),
-            :replicate_idx => collect(1:nreplicates),
-            :griddims => (20, 20),
-        )
+        parameters = Dict(:numagents => collect(numagents_low:numagents_high),
+                          :replicate_idx => collect(1:nreplicates),
+                          :griddims => (20, 20))
         paramscan(parameters, initialize;
                   parallel, adata = [:pos, :mood, :group],
                   mdata = [:min_to_be_happy], showprogress = true,
                   agent_step! = schelling_agent_step!,
-                  when = (model, step) ->
-                   ( (step) % whensteps == 0  ||  step == 0 ),
+                  when = (model, step) -> ((step) % whensteps == 0 || step == 0),
                   n = nsteps)
     end
 end
@@ -72,7 +69,7 @@ for (f, parallel, nreplicates, name) in [
     (paramscan, false, 10, "paramscan serial 10 reps"),
     (paramscan, true, 10, "paramscan parallel 10 reps"),
     (paramscan, false, 20, "paramscan serial 20 reps"),
-    (paramscan, true, 20, "paramscan parallel 20 reps")
+    (paramscan, true, 20, "paramscan parallel 20 reps"),
 ]
-    ENSEMBLES_SUITE[name] = @benchmarkable ensemble_benchmark($f, $parallel, $nreplicates) samples = 1
+    ENSEMBLES_SUITE[name] = @benchmarkable ensemble_benchmark($f, $parallel, $nreplicates) samples=1
 end

benchmark/ensembles.jl

@@ -20,7 +20,7 @@ function fake_model(; nagents = 500, scheduler)
 end
 
 function fake_model_multi(; nagents = 500, scheduler)
-    model = ABM(Union{FakeAgent,OtherFakeAgent}; scheduler, warn = false)
+    model = ABM(Union{FakeAgent, OtherFakeAgent}; scheduler, warn = false)
     for i in 1:nagents
         if i % 2 == 0
             add_agent!(FakeAgent(i, 1), model)
@@ -42,7 +42,10 @@ for (model, name) in [
     (fake_model(; scheduler = Schedulers.partially(0.7)), "partially"),
     (fake_model(; scheduler = Schedulers.Partially(0.7)), "Partially"),
     (fake_model_multi(; scheduler = Schedulers.by_type(true, true)), "by_type"),
-    (fake_model_multi(; scheduler = Schedulers.ByType(true, true, Union{FakeAgent,OtherFakeAgent})), "ByType")
+    (fake_model_multi(;
+                      scheduler = Schedulers.ByType(true, true,
+                                                    Union{FakeAgent, OtherFakeAgent})),
+     "ByType"),
 ]
     SUITE[name] = @benchmarkable Agents.schedule($model)
 end
@@ -55,12 +58,18 @@ for (model, name) in [
     (fake_model(; nagents = 800000, scheduler = Schedulers.ByID()), "ByID"),
     (fake_model(; nagents = 800000, scheduler = Schedulers.randomly), "randomly"),
     (fake_model(; nagents = 800000, scheduler = Schedulers.Randomly()), "Randomly"),
-    (fake_model(; nagents = 800000, scheduler = Schedulers.by_property(:group)), "by_property"),
-    (fake_model(; nagents = 800000, scheduler = Schedulers.ByProperty(:group)), "ByProperty"),
+    (fake_model(; nagents = 800000, scheduler = Schedulers.by_property(:group)),
+     "by_property"),
+    (fake_model(; nagents = 800000, scheduler = Schedulers.ByProperty(:group)),
+     "ByProperty"),
     (fake_model(; nagents = 800000, scheduler = Schedulers.partially(0.7)), "partially"),
     (fake_model(; nagents = 800000, scheduler = Schedulers.Partially(0.7)), "Partially"),
-    (fake_model_multi(; nagents = 800000, scheduler = Schedulers.by_type(true, true)), "by_type"),
-    (fake_model_multi(; nagents = 800000, scheduler = Schedulers.ByType(true, true, Union{FakeAgent,OtherFakeAgent})), "ByType")
+    (fake_model_multi(; nagents = 800000, scheduler = Schedulers.by_type(true, true)),
+     "by_type"),
+    (fake_model_multi(; nagents = 800000,
+                      scheduler = Schedulers.ByType(true, true,
+                                                    Union{FakeAgent, OtherFakeAgent})),
+     "ByType"),
 ]
     SUITE["large_model"][name] = @benchmarkable Agents.schedule($model)
 end
@@ -77,8 +86,12 @@ for (model, name) in [
     (fake_model(; nagents = 50, scheduler = Schedulers.ByProperty(:group)), "ByProperty"),
     (fake_model(; nagents = 50, scheduler = Schedulers.partially(0.7)), "partially"),
     (fake_model(; nagents = 50, scheduler = Schedulers.Partially(0.7)), "Partially"),
-    (fake_model_multi(; nagents = 50, scheduler = Schedulers.by_type(true, true)), "by_type"),
-    (fake_model_multi(; nagents = 50, scheduler = Schedulers.ByType(true, true, Union{FakeAgent,OtherFakeAgent})), "ByType")
+    (fake_model_multi(; nagents = 50, scheduler = Schedulers.by_type(true, true)),
+     "by_type"),
+    (fake_model_multi(; nagents = 50,
+                      scheduler = Schedulers.ByType(true, true,
+                                                    Union{FakeAgent, OtherFakeAgent})),
+     "ByType"),
 ]
     SUITE["small_model"][name] = @benchmarkable Agents.schedule($model)
 end

benchmark/schedulers.jl

@@ -5,7 +5,7 @@
 # die in this simulation.
 # Based on https://github.com/JuliaDynamics/Agents.jl/issues/640#issuecomment-1166330815
 mutable struct AbusiveUnkillableAgent
-    pos::Tuple{Int,Int}
+    pos::Tuple{Int, Int}
     group::Int
     happy::Bool
 end
@@ -19,7 +19,7 @@ function initialize_abusiveunkillable()
     model = AbusiveUnkillableModel(AbusiveUnkillableAgent[], fill(0, grid_size...))
     N = floor(Int, grid_size[1] * grid_size[2] * grid_occupation * 0.5)
 
-    for i in 1:2N
+    for i in 1:(2N)
         grp = i <= N ? 1 : 2
         pos = (rand(1:grid_size[1]), rand(1:grid_size[2]))
         while model.space[pos...] > 0
@@ -67,7 +67,7 @@ end
 
 model_abusiveunkillable = initialize_abusiveunkillable()
 println("Benchmarking abusive unkillable version")
-@btime simulate_abusiveunkillable!($model_abusiveunkillable) setup = (model_abusiveunkillable = initialize_abusiveunkillable())
+@btime simulate_abusiveunkillable!($model_abusiveunkillable) setup=(model_abusiveunkillable = initialize_abusiveunkillable())
 println("Benchmarking abusive unkillable version: count nearby same")
 model_abusiveunkillable = initialize_abusiveunkillable()
-@btime count_nearby_same_abusiveunkillable(agent, model_abusiveunkillable) setup = (agent = rand(model_abusiveunkillable.agents))
+@btime count_nearby_same_abusiveunkillable(agent, model_abusiveunkillable) setup=(agent = rand(model_abusiveunkillable.agents))

benchmark/simple_grid/abusive_unkillable.jl

@@ -31,9 +31,9 @@ function count_nearby_same_exact(agent, model)
 end
 
 println("Continuous space count nearby ids, spacing=$spacing, r=$r, inexact")
-@btime count_nearby_same(agent, $model) setup = (agent = random_agent($model))
+@btime count_nearby_same(agent, $model) setup=(agent = random_agent($model))
 println("Continuous space count nearby ids, spacing=$spacing, r=$r, exact")
-@btime count_nearby_same_exact(agent, $model) setup = (agent = random_agent($model))
+@btime count_nearby_same_exact(agent, $model) setup=(agent = random_agent($model))
 
 #= Results
 # Current state
@@ -47,7 +47,6 @@ Continuous space count nearby ids, spacing=0.05, r=0.1, exact
 Continuous space count nearby ids, spacing=0.1, r=0.1, exact
   4.843 μs (58 allocations: 5.22 KiB)
 
-
 # master
 Continuous space count nearby ids, spacing=0.05, r=0.1, inexact
   3.612 μs (71 allocations: 4.91 KiB)
@@ -58,4 +57,4 @@ Continuous space count nearby ids, spacing=0.1, r=0.1, inexact
   5.900 μs (118 allocations: 8.05 KiB)
 Continuous space count nearby ids, spacing=0.1, r=0.1, exact
   10.300 μs (218 allocations: 32.88 KiB)
-  =#
+  =#

benchmark/simple_grid/continuous_space_impact.jl

@@ -5,7 +5,7 @@ end
 
 function initialize_dict()
     model = Dict{NTuple{2, Int}, DictAgent}()
-    N = grid_size[1]*grid_size[2]*grid_occupation
+    N = grid_size[1] * grid_size[2] * grid_occupation
     for n in 1:N
         group = n < N / 2 ? 1 : 2
         pos = (rand(1:grid_size[1]), rand(1:grid_size[2]))
@@ -49,4 +49,4 @@ end
 
 model_dict = initialize_dict()
 println("benchmarking dict-based step")
-@btime simulation_step!($model_dict) setup = (model_dict = initialize_dict())
+@btime simulation_step!($model_dict) setup=(model_dict = initialize_dict())

benchmark/simple_grid/dict_based.jl

@@ -10,7 +10,7 @@ function initialize_gridspace()
     properties = Dict(:min_to_be_happy => min_to_be_happy)
     rng = Random.Xoshiro(rand(UInt))
     model = ABM(GridSpaceAgent, space; properties, rng)
-    N = grid_size[1]*grid_size[2]*grid_occupation
+    N = grid_size[1] * grid_size[2] * grid_occupation
     for n in 1:N
         group = n < N / 2 ? 1 : 2
         agent = GridSpaceAgent(n, (1, 1), group, false)
@@ -41,18 +41,18 @@ end
 model = initialize_gridspace()
 agent = model[1]
 println("Benchmarking GridSpace version")
-@btime step!($model, agent_step_gridspace!) setup = (model = initialize_gridspace())
+@btime step!($model, agent_step_gridspace!) setup=(model = initialize_gridspace())
 
 println("Benchmarking GridSpace version: count nearby same")
 model = initialize_gridspace()
-@btime count_nearby_same(agent, model) setup = (agent = random_agent(model))
+@btime count_nearby_same(agent, model) setup=(agent = random_agent(model))
 
 function profile_nearby_same(model, agent = random_agent(model))
     x = 0
     N = 1000000
     for i in 1:N
         x += count_nearby_same(agent, model)
     end
-    return x/N
+    return x / N
 end
-# Then use @profview in VSCode
+# Then use @profview in VSCode

benchmark/simple_grid/gridspace_based.jl

@@ -5,12 +5,12 @@ const min_to_be_happy = 3 # how many nearby agents you need of same group
 const grid_occupation = 0.8 # percentage of space occupied by agents
 const grid_size = (30, 30)
 const moore = [
-    (1,0),  (1,1),  (1,-1),
-    (0,1),          (0,-1),
-    (-1,0), (-1,1), (-1,-1),
+    (1, 0), (1, 1), (1, -1),
+    (0, 1), (0, -1),
+    (-1, 0), (-1, 1), (-1, -1),
 ]
 
 include("dict_based.jl")
 include("gridspace_based.jl")
 include("simple_grid_space.jl")
-include("abusive_unkillable.jl")
+include("abusive_unkillable.jl")