Merge pull request #1246 from Mattriks/boxplot_dodge

Boxplot dodging

docs/src/gallery/geometries.md

@@ -63,9 +63,19 @@ plot(dataset("lattice", "singer"), x="VoicePart", y="Height", Geom.beeswarm)
 ## [`Geom.boxplot`](@ref)
 
 ```@example
-using Gadfly, RDatasets
-set_default_plot_size(14cm, 8cm)
-plot(dataset("lattice", "singer"), x="VoicePart", y="Height", Geom.boxplot)
+using Compose, Gadfly, RDatasets
+set_default_plot_size(21cm, 8cm)
+singers, salaries = dataset("lattice", "singer"), dataset("car","Salaries")
+salaries.Salary /= 1000.0
+salaries.Discipline = ["Discipline $(x)" for x in salaries.Discipline]
+p1 = plot(singers, x=:VoicePart, y=:Height, Geom.boxplot, 
+    Theme(default_color="MidnightBlue"))
+p2 = plot(salaries, x=:Discipline, y=:Salary, color=:Rank,
+    Scale.x_discrete(levels=["Discipline A", "Discipline B"]),
+    Geom.boxplot, Theme(boxplot_spacing=0.1cx),
+    Guide.colorkey(title="", pos=[0.78w,-0.4h])
+)
+hstack(p1, p2)
 ```
 
 

src/geom/boxplot.jl

@@ -17,6 +17,8 @@ outliers.
 
 Alternatively, if the `y` aesthetic is specified instead, the middle, hinges,
 fences, and outliers aesthetics will be computed using [`Stat.boxplot`](@ref).
+
+Boxplots will be automatically dodged by specifying a `color` aesthetic different to the `x` aesthetic.
 """
 const boxplot = BoxplotGeometry
 
@@ -125,11 +127,9 @@ function render(geom::BoxplotGeometry, theme::Gadfly.Theme, aes::Gadfly.Aestheti
         xys = collect(Iterators.flatten(zip(cycle([x]), ys, cycle([c]))
                              for (x, ys, c) in zip(xs, aes.outliers, cs)))
         compose!(ctx, (context(),
-            (context(), Shape.circle([x for (x, y, c) in xys],
-                    [y for (x, y, c) in xys],
-                    [theme.point_size], to), svgclass("marker")),
+            Shape.circle([x for (x, y, c) in xys], [y for (x, y, c) in xys], [theme.point_size], to), svgclass("marker"),
              stroke([theme.discrete_highlight_color(c) for (x, y, c) in xys]),
-             fill([c for (x, y, c) in xys])))
+             fill([c for (x, y, c) in xys]) ))
     end
 
     # Middle

src/statistics.jl

@@ -996,14 +996,17 @@ function apply_statistic(stat::BoxplotStatistic,
                          coord::Gadfly.CoordinateElement,
                          aes::Gadfly.Aesthetics)
 
-    if aes.x === nothing
-        aes_x = [1]
-        aes.x_label = x -> fill("", length(x))
-    else
+    xflag = aes.x != nothing
+    aes_x = (xflag ? eltype(aes.x) : Int)[]
+    if xflag
         aes_x = aes.x
+    else
+        aes_x = ones(Int, length(aes.y))
+        aes.x_label = x -> fill("", length(x))
     end
-    aes_color = aes.color === nothing ? [nothing] : aes.color
-
+    colorflag = aes.color != nothing
+    aes_color =  colorflag ? aes.color : fill(nothing, length(aes_x))
+
     if aes.y == nothing
         groups = Any[]
         for (x, c) in zip(aes.x, cycle(aes_color))
@@ -1024,30 +1027,30 @@ function apply_statistic(stat::BoxplotStatistic,
             aes.yviewmax = yviewmax
         end
     else
-        T = isempty(aes.y) ? eltype(aes.y) : typeof(aes.y[1] / 1)
-        groups = DefaultOrderedDict(() -> T[])
-
-        for (x, y, c) in zip(cycle(aes_x), aes.y, cycle(aes_color))
-            push!(groups[(x, c)], y)
-        end
-
-        m = length(groups)
-        aes.x = Array{eltype(aes.x)}(undef, m)
-        aes.middle = Array{T}(undef, m)
-        aes.lower_hinge = Array{T}(undef, m)
-        aes.upper_hinge = Array{T}(undef, m)
-        aes.lower_fence = Array{T}(undef, m)
-        aes.upper_fence = Array{T}(undef, m)
-        aes.outliers = Vector{T}[]
-
-        for (i, ((x, c), ys)) in enumerate(groups)
+        YT = isempty(aes.y) ? eltype(aes.y) : typeof(aes.y[1] / 1)        
+        CT, XT = eltype(aes_color), eltype(aes_x)    
+        groups = collect(Tuple{CT, XT}, zip(aes_color, aes_x))
+        ug = unique(groups)
+
+        K = Tuple{CT, XT}
+        grouped_y = Dict{K, Vector{YT}}(g=>aes.y[groups.==[g]] for g in ug)
+
+        m = length(ug)
+        aes.x = Vector{XT}(undef, m)
+        aes.middle = Vector{YT}(undef, m)
+        aes.lower_hinge = Vector{YT}(undef, m)
+        aes.upper_hinge = Vector{YT}(undef, m)
+        aes.lower_fence = Vector{YT}(undef, m)
+        aes.upper_fence = Vector{YT}(undef, m)
+        aes.outliers = Vector{YT}[]
+
+        for (i, ((c, x), ys)) in enumerate(grouped_y)
             sort!(ys)
 
             aes.x[i] = x
 
             if stat.method == :tukey
-                aes.lower_hinge[i], aes.middle[i], aes.upper_hinge[i] =
-                        quantile(ys, [0.25, 0.5, 0.75])
+                aes.lower_hinge[i], aes.middle[i], aes.upper_hinge[i] = quantile(ys, [0.25, 0.5, 0.75])
                 iqr = aes.upper_hinge[i] - aes.lower_hinge[i]
 
                 idx = searchsortedfirst(ys, aes.lower_hinge[i] - 1.5iqr)
@@ -1057,22 +1060,19 @@ function apply_statistic(stat::BoxplotStatistic,
                 aes.upper_fence[i] = ys[idx]
             elseif isa(stat.method, Vector)
                 qs = stat.method
-                if length(qs) != 5
-                    error("Stat.boxplot requires exactly five quantiles.")
-                end
+                (length(qs) != 5) && error("Stat.boxplot requires exactly five quantiles.")
 
                 aes.lower_fence[i], aes.lower_hinge[i], aes.middle[i],
                 aes.upper_hinge[i], aes.upper_fence[i] = quantile!(ys, qs)
             else
                 error("Invalid method specified for State.boxplot")
             end
 
-            push!(aes.outliers,
-                 filter(y -> y < aes.lower_fence[i] || y > aes.upper_fence[i], ys))
+            push!(aes.outliers, filter(y -> y < aes.lower_fence[i] || y > aes.upper_fence[i], ys))
         end
     end
 
-    if length(aes.x) > 1 && (haskey(scales, :x) && isa(scales[:x], Scale.ContinuousScale))
+    if length(aes.x) > 1 && haskey(scales, :x) && isa(scales[:x], Scale.ContinuousScale)
         xmin, xmax = minimum(aes.x), maximum(aes.x)
         minspan = minimum([xj - xi for (xi, xj) in zip(aes.x[1:end-1], aes.x[2:end])])
 
@@ -1088,14 +1088,17 @@ function apply_statistic(stat::BoxplotStatistic,
         end
     end
 
-    if isa(aes_x, IndirectArray)
-        aes.x = discretize_make_ia(aes.x, aes_x.values)
+    if haskey(scales, :x) && isa(scales[:x], Scale.DiscreteScale)
+        aes.xviewmin = minimum(aes.x)-0.5
+        aes.xviewmax = maximum(aes.x)+0.5
+        aes.pad_categorical_x = false
     end
 
-    if aes.color !== nothing
-        aes.color = discretize_make_ia(RGB{Float32}[c for (x, c) in keys(groups)],
-            aes.color.values)
-    end
+    isa(aes_x, IndirectArray) && (aes.x = discretize_make_ia(aes.x, aes_x.values))
+
+    colorflag && (aes.color = discretize_make_ia(first.(keys(grouped_y)), aes.color.values))
+
+    Stat.apply_statistic(Stat.dodge(), scales, coord, aes)
 
     nothing
 end

test/testscripts/colored_boxplot.jl

@@ -1,6 +1,12 @@
-using Gadfly, RDatasets
+using Compose, Gadfly, RDatasets
 
-set_default_plot_size(6inch, 3inch)
+set_default_plot_size(21cm, 16cm)
 
 mpg = dataset("ggplot2","mpg")
-plot(mpg, x=:Class, y=:Hwy, Geom.boxplot, color=:Class)
+p1 = plot(mpg, x=:Class, y=:Hwy, color=:Class, Geom.boxplot, Theme(boxplot_spacing=0.3cx) )
+
+p2 = plot(mpg, x=:Class, y=:Hwy, color=:Drv, Geom.boxplot,
+    Guide.colorkey(labels=["front", "4-wheel\t\t","rear"]),
+    Theme(boxplot_spacing=0.1cx) )
+
+vstack(p1, p2)