Merge pull request #10 from N5N3/N5N3-patch-1

replace all AbstractArray with AbstractVecOrMat

Author: N5N3

stdlib/LinearAlgebra/src/diagonal.jl

@@ -316,10 +316,11 @@ function mul!(C::AbstractMatrix, Da::Diagonal, Db::Diagonal, alpha::Number, beta
     return C
 end
 
+#TODO: many of /, \ related function has no size check and singular check
+(/)(A::AbstractVecOrMat, D::Diagonal) =
+    rdiv!((typeof(oneunit(eltype(D))/oneunit(eltype(A)))).(A), D)
 (/)(Da::Diagonal, Db::Diagonal) = Diagonal(Da.diag ./ Db.diag)
 
-ldiv!(x::AbstractArray, A::Diagonal, b::AbstractArray) = (x .= A.diag .\ b)
-
 function rdiv!(A::AbstractMatrix, D::Diagonal)
     require_one_based_indexing(A)
     dd = D.diag
@@ -339,8 +340,30 @@ function rdiv!(A::AbstractMatrix, D::Diagonal)
     A
 end
 
-(/)(A::AbstractArray, D::Diagonal) =
-    rdiv!((typeof(oneunit(eltype(D))/oneunit(eltype(A)))).(A), D)
+(\)(D::Diagonal, A::AbstractMatrix) =
+    ldiv!(D, (typeof(oneunit(eltype(D))/oneunit(eltype(A)))).(A))
+(\)(D::Diagonal, b::AbstractVector) = D.diag .\ b
+(\)(Da::Diagonal, Db::Diagonal) = Diagonal(Da.diag .\ Db.diag)
+
+ldiv!(x::AbstractVecOrMat, A::Diagonal, b::AbstractVecOrMat) = (x .= A.diag .\ b)
+
+function ldiv!(D::Diagonal, B::AbstractVecOrMat)
+    m, n = size(B, 1), size(B, 2)
+    if m != length(D.diag)
+        throw(DimensionMismatch("diagonal matrix is $(length(D.diag)) by $(length(D.diag)) but right hand side has $m rows"))
+    end
+    (m == 0 || n == 0) && return B
+    for j = 1:n
+        for i = 1:m
+            di = D.diag[i]
+            if di == 0
+                throw(SingularException(i))
+            end
+            B[i,j] = di \ B[i,j]
+        end
+    end
+    return B
+end
 
 # (l/r)mul!, l/rdiv!, *, / and \ Optimization for AbstractTriangular.
 # These functions are generally more efficient if we calculate the whole data field.
@@ -469,30 +492,6 @@ for f in (:exp, :cis, :log, :sqrt,
     @eval $f(D::Diagonal) = Diagonal($f.(D.diag))
 end
 
-(\)(D::Diagonal, A::AbstractMatrix) =
-    ldiv!(D, (typeof(oneunit(eltype(D))/oneunit(eltype(A)))).(A))
-
-(\)(D::Diagonal, b::AbstractVector) = D.diag .\ b
-(\)(Da::Diagonal, Db::Diagonal) = Diagonal(Da.diag .\ Db.diag)
-
-function ldiv!(D::Diagonal, B::AbstractVecOrMat)
-    m, n = size(B, 1), size(B, 2)
-    if m != length(D.diag)
-        throw(DimensionMismatch("diagonal matrix is $(length(D.diag)) by $(length(D.diag)) but right hand side has $m rows"))
-    end
-    (m == 0 || n == 0) && return B
-    for j = 1:n
-        for i = 1:m
-            di = D.diag[i]
-            if di == 0
-                throw(SingularException(i))
-            end
-            B[i,j] = di \ B[i,j]
-        end
-    end
-    return B
-end
-
 function inv(D::Diagonal{T}) where T
     Di = similar(D.diag, typeof(inv(zero(T))))
     for i = 1:length(D.diag)
@@ -572,7 +571,6 @@ function _mapreduce_prod(f, x, D::Diagonal, y)
     end
 end
 
-
 function cholesky!(A::Diagonal, ::Val{false} = Val(false); check::Bool = true)
     info = 0
     for (i, di) in enumerate(A.diag)