Initialize indices with 1, allows for Infs and NaNs in the data

src/knn.jl

@@ -11,7 +11,7 @@ end
 Performs a lookup of the `k` nearest neigbours to the `points` from the data
 in the `tree`. If `sortres = true` the result is sorted such that the results are
 in the order of increasing distance to the point. `skip` is an optional predicate
-to determine if a point that would be returned should be skipped based on its 
+to determine if a point that would be returned should be skipped based on its
 index.
 """
 function knn(tree::NNTree{V}, points::Vector{T}, k::Int, sortres=false, skip::F=always_false) where {V, T <: AbstractVector, F<:Function}
@@ -27,7 +27,7 @@ function knn(tree::NNTree{V}, points::Vector{T}, k::Int, sortres=false, skip::F=
 end
 
 function knn_point!(tree::NNTree{V}, point::AbstractVector{T}, sortres, dist, idx, skip::F) where {V, T <: Number, F}
-    fill!(idx, -1)
+    fill!(idx, 1)
     fill!(dist, typemax(get_T(eltype(V))))
     _knn(tree, point, idx, dist, skip)
     sortres && heap_sort_inplace!(dist, idx)

test/runtests.jl

@@ -31,6 +31,7 @@ include("test_knn.jl")
 include("test_inrange.jl")
 include("test_monkey.jl")
 include("test_datafreetree.jl")
+include("test_specialfloats.jl")
 
 @testset "periodic euclidean" begin
     pred = PeriodicEuclidean([Inf, 2.5])

test/test_specialfloats.jl

@@ -0,0 +1,53 @@
+using NearestNeighbors
+using Test
+
+# Test for issue #125
+@testset "nan on query" begin
+    for _ in 1:111
+        Ndim = 35
+        Npt = 408
+
+        data = randn(Ndim, Npt)
+        tree = KDTree(data)
+
+        pointnan = repeat([NaN], Ndim)
+        indnan,distnan = nn(tree, pointnan)
+        @test 1 <= indnan <= Npt
+    end
+end
+
+# # Test for issue #78
+# @testset "infs on data" begin
+#     for _ in 1:11
+#         coords = [
+#             29882.5 25974.3 Inf  Inf 17821.8 Inf Inf Inf Inf Inf 16322.0;
+#             9279.86 9286.35 Inf Inf 10320.4 Inf Inf Inf Inf Inf 11459.0;
+#             0.0 0.0 Inf Inf 0.0 Inf Inf Inf Inf Inf 0.0]
+#         point =  [17889.55, 2094.45, 0.0]
+
+#         tree = BallTree(coords)
+#         idx, _ = knn(tree, point, 1)
+#         @test idx[1] == 5
+#     end
+# end
+
+# @testset "nan on data" begin
+#     for _ in 1:11
+#         # Ndim = 35
+#         # Npt = 408
+
+#         # data = randn(Ndim, Npt)
+#         # tree = KDTree(data)
+
+#         # datanan = copy(data)
+#         # datanan[rand(1:Ndim),rand(1:Npt)] = NaN
+#         # treenan = KDTree(datanan)
+
+#         # pointrand = randn(Ndim)
+
+#         # @show indnan2, distnan2 = nn(tree, pointrand)
+#         # @show indnan2, distnan2 = nn(treenan, pointrand)
+#         # @test 1 <= indnan2 <= Npt
+
+#     end
+# end