Enable construction and queries for empty trees (#36)

* Allow 0 point tree

* Clean up check_k function

* Add tests for empty trees

src/ball_tree.jl

@@ -48,7 +48,7 @@ function BallTree{V <: AbstractArray, M <: Metric}(data::Vector{V},
     # Bottom up creation of hyper spheres so need spheres even for leafs)
     hyper_spheres = Array(HyperSphere{length(V), eltype(V)}, tree_data.n_internal_nodes + tree_data.n_leafs)
 
-        if reorder
+    if reorder
         indices_reordered = Vector{Int}(n_p)
         if isempty(reorderbuffer)
             data_reordered = Vector{V}(n_p)
@@ -61,9 +61,11 @@ function BallTree{V <: AbstractArray, M <: Metric}(data::Vector{V},
         data_reordered = Vector{V}()
     end
 
-    # Call the recursive BallTree builder
-    build_BallTree(1, data, data_reordered, hyper_spheres, metric, indices, indices_reordered,
-                   1,  length(data), tree_data, array_buffs, reorder)
+    if n_p > 0
+        # Call the recursive BallTree builder
+        build_BallTree(1, data, data_reordered, hyper_spheres, metric, indices, indices_reordered,
+                       1,  length(data), tree_data, array_buffs, reorder)
+    end
 
     if reorder
        data = data_reordered
@@ -201,6 +203,11 @@ function inrange_kernel!(tree::BallTree,
                          query_ball::HyperSphere,
                          idx_in_ball::Vector{Int})
     @NODE 1
+
+    if index > length(tree.hyper_spheres)
+        return
+    end
+
     sphere = tree.hyper_spheres[index]
 
     # If the query ball in the bounding sphere for the current sub tree

src/hyperrectangles.jl

@@ -7,7 +7,6 @@ end
 
 # Computes a bounding box around a point cloud
 function compute_bbox{V <: AbstractVector}(data::Vector{V})
-    @assert length(data) != 0
     T = eltype(V)
     n_dim = length(V)
     maxes = Vector{T}(n_dim)

src/knn.jl

@@ -1,4 +1,8 @@
-check_k(tree, k) = (k > length(tree.data)|| k <= 0) && throw(ArgumentError("k > number of points in tree or ≦ 0"))
+function check_k(tree, k)
+    if k > length(tree.data) || k < 0
+        throw(ArgumentError("k > number of points in tree or < 0"))
+    end
+end
 
 """
     knn(tree::NNTree, points, k [, sortres=false]) -> indices, distances
@@ -11,7 +15,6 @@ to determine if a point that would be returned should be skipped.
 function knn{V, T <: AbstractVector}(tree::NNTree{V}, points::Vector{T}, k::Int, sortres=false, skip::Function=always_false)
     check_input(tree, points)
     check_k(tree, k)
-
     n_points = length(points)
     dists = [Vector{get_T(eltype(V))}(k) for _ in 1:n_points]
     idxs = [Vector{Int}(k) for _ in 1:n_points]
@@ -34,10 +37,7 @@ function knn_point!{V, T <: Number}(tree::NNTree{V}, point::AbstractVector{T}, s
 end
 
 function knn{V, T <: Number}(tree::NNTree{V}, point::AbstractVector{T}, k::Int, sortres=false, skip::Function=always_false)
-    if k > length(tree.data)|| k <= 0
-        throw(ArgumentError("k > number of points in tree or ≦ 0"))
-    end
-
+    check_k(tree, k)
     idx = Vector{Int}(k)
     dist = Vector{get_T(eltype(V))}(k)
     knn_point!(tree, point, sortres, dist, idx, skip)

test/test_inrange.jl

@@ -37,6 +37,10 @@
 
             @test_throws ArgumentError inrange(tree, rand(3), -0.1)
             @test_throws ArgumentError inrange(tree, rand(5), 1.0)
+
+            empty_tree = TreeType(rand(3,0), metric)
+            idxs = inrange(empty_tree, [0.5, 0.5, 0.5], 1.0)
+            @test idxs == []
         end
     end
 end

test/test_knn.jl

@@ -29,9 +29,15 @@ import Distances.evaluate
             idxs, dists = knn(tree, [1//10, 8//10], 3, true)
             @test idxs == [3, 2, 5]
 
+            @test_throws ArgumentError knn(tree, [0.1, 0.8], -1) # k < 0
             @test_throws ArgumentError knn(tree, [0.1, 0.8], 10) # k > n_points
             @test_throws ArgumentError knn(tree, [0.1], 10) # n_dim != trees dim
 
+            empty_tree = TreeType(rand(2,0), metric; leafsize=2)
+            idxs, dists = knn(empty_tree, [0.5, 0.5], 0, true)
+            @test idxs == Int[]
+            @test_throws ArgumentError knn(empty_tree, [0.1, 0.8], -1) # k < 0
+            @test_throws ArgumentError knn(empty_tree, [0.1, 0.8], 1)  # k > n_points
         end
     end
 end