Merge pull request #109 from alan-turing-institute/dev

For a 0.8.1 release

Author: ablaom

.travis.yml

@@ -4,10 +4,7 @@ os:
   - linux
 julia:
   - 1.0
-  - 1.1
-  - 1.2
-  - 1.3
-  - 1.4
+  - 1.5
   - nightly
 matrix:
   allow_failures:

Project.toml

@@ -1,7 +1,7 @@
 name = "ScientificTypes"
 uuid = "321657f4-b219-11e9-178b-2701a2544e81"
 authors = ["Anthony D. Blaom <[email protected]>"]
-version = "0.8.0"
+version = "0.8.1"
 
 [compat]
 julia = "1"

src/scitype.jl

@@ -4,7 +4,35 @@
 """
     scitype(X)
 
-The scientific type that `X` may represent.
+The scientific type (interpretation) of `X`, as distinct from its
+machine type, as specified by the active convention.
+
+### Examples from the MLJ convention
+
+```
+julia> using MLJScientificTypes # or `using MLJ`
+julia> scitype(3.14)
+Continuous
+
+julia> scitype([1, 2, 3, missing])
+AbstractArray{Union{Missing, Count},1}
+
+julia> scitype((5, "beige"))
+Tuple{Count, Textual}
+
+julia> using CategoricalArrays
+julia> X = (gender = categorical([:M, :M, :F, :M, :F]),
+            ndevices = [1, 3, 2, 3, 2])
+julia> scitype(X)
+Table{Union{AbstractArray{Count,1}, AbstractArray{Multiclass{2},1}}}
+```
+
+The specific behavior of `scitype` is governed by the active
+convention, as returned by `ScientificTypes.convention()`. The
+[MLJScientificTypes.jl
+documentation](https://alan-turing-institute.github.io/MLJScientificTypes.jl/dev/)
+details the convention demonstrated above.
+
 """
 scitype(X;    kw...) = scitype(X, convention();     kw...)
 scitype(X, C; kw...) = scitype(X, C, Val(trait(X)); kw...)
@@ -23,7 +51,10 @@ scitype(t::Tuple, ::Convention; kw...) = Tuple{scitype.(t; kw...)...}
 Return the type union, over all elements `x` generated by the iterable `A`,
 of `scitype(x)`. See also [`scitype`](@ref).
 """
-scitype_union(A) = reduce((a,b)->Union{a,b}, (scitype(el) for el in A))
+function scitype_union(A)
+    isempty(A) && return scitype(eltype(A))
+    reduce((a,b)->Union{a,b}, (scitype(el) for el in A))
+end
 
 
 # -----------------------------------------------------------------

test/scitype.jl

@@ -16,7 +16,7 @@
 
     Xm = Any[missing, missing]
     @test scitype(Xm) == AbstractVector{Missing}
-    
+
     @test scitype([missing, missing]) == AbstractVector{Missing}
 end
 
@@ -41,3 +41,13 @@ end
     @test ScientificDateTime <: ScientificTimeType
     @test ScientificTime <: ScientificTimeType
 end
+
+@testset "Empty array" begin
+    set_convention(MockMLJ())
+    ScientificTypes.Scitype(::Type{<:Integer}, ::MockMLJ) = Count
+    ScientificTypes.Scitype(::Type{Missing}, ::MockMLJ) = Missing
+    @test scitype(Int[]) == AbstractVector{Count}
+    @test scitype(Any[]) == AbstractVector{Unknown}
+    @test scitype(Missing[]) == AbstractVector{Missing}
+    @test scitype(Vector{Union{Int,Missing}}()) == AbstractVector{Union{Missing,Count}}
+end