more verbose type names

NEWS.md

@@ -103,9 +103,10 @@ Standard library changes
 * The shape of an `UpperHessenberg` matrix is preserved under certain arithmetic operations, e.g. when multiplying or dividing by an `UpperTriangular` matrix. ([#40039])
 * `cis(A)` now supports matrix arguments ([#40194]).
 * `dot` now supports `UniformScaling` with `AbstractMatrix` ([#40250]).
-* `qr[!]` and `lu[!]` now support `PivotingStrategy` values as their optional `pivot` argument:
-  defaults are `qr(A, NoPivot())` (vs. `qr(A, ColNorm())` for pivoting) and `lu(A, RowMax())`
-  (vs. `lu(A, NoPivot())` without pivoting); the former `Val{true/false}`-based calls are deprecated. ([#40623])
+* `qr[!]` and `lu[!]` now support `LinearAlgebra.PivotingStrategy` (singleton type) values
+  as their optional `pivot` argument: defaults are `qr(A, NoPivot())` (vs.
+  `qr(A, ColumnNorm())` for pivoting) and `lu(A, RowMaximum())` (vs. `lu(A, NoPivot())`
+  without pivoting); the former `Val{true/false}`-based calls are deprecated. ([#40623])
 
 #### Markdown
 

stdlib/LinearAlgebra/src/LinearAlgebra.jl

@@ -35,7 +35,7 @@ export
     BunchKaufman,
     Cholesky,
     CholeskyPivoted,
-    ColNorm,
+    ColumnNorm,
     Eigen,
     GeneralizedEigen,
     GeneralizedSVD,
@@ -50,7 +50,7 @@ export
     Schur,
     SVD,
     Hermitian,
-    RowMax,
+    RowMaximum,
     Symmetric,
     LowerTriangular,
     UpperTriangular,
@@ -169,8 +169,8 @@ struct QRIteration <: Algorithm end
 
 abstract type PivotingStrategy end
 struct NoPivot <: PivotingStrategy end
-struct RowMax <: PivotingStrategy end
-struct ColNorm <: PivotingStrategy end
+struct RowMaximum <: PivotingStrategy end
+struct ColumnNorm <: PivotingStrategy end
 
 # Check that stride of matrix/vector is 1
 # Writing like this to avoid splatting penalty when called with multiple arguments,

stdlib/LinearAlgebra/src/dense.jl

@@ -1371,7 +1371,7 @@ function factorize(A::StridedMatrix{T}) where T
         end
         return lu(A)
     end
-    qr(A, ColNorm())
+    qr(A, ColumnNorm())
 end
 factorize(A::Adjoint)   =   adjoint(factorize(parent(A)))
 factorize(A::Transpose) = transpose(factorize(parent(A)))

stdlib/LinearAlgebra/src/factorization.jl

@@ -16,9 +16,9 @@ size(F::Adjoint{<:Any,<:Factorization}) = reverse(size(parent(F)))
 size(F::Transpose{<:Any,<:Factorization}) = reverse(size(parent(F)))
 
 checkpositivedefinite(info) = info == 0 || throw(PosDefException(info))
-checknonsingular(info, ::RowMax) = info == 0 || throw(SingularException(info))
+checknonsingular(info, ::RowMaximum) = info == 0 || throw(SingularException(info))
 checknonsingular(info, ::NoPivot) = info == 0 || throw(ZeroPivotException(info))
-checknonsingular(info) = checknonsingular(info, RowMax())
+checknonsingular(info) = checknonsingular(info, RowMaximum())
 
 """
     issuccess(F::Factorization)

stdlib/LinearAlgebra/src/generic.jl

@@ -1141,7 +1141,7 @@ function (\)(A::AbstractMatrix, B::AbstractVecOrMat)
         end
         return lu(A) \ B
     end
-    return qr(A, ColNorm()) \ B
+    return qr(A, ColumnNorm()) \ B
 end
 
 (\)(a::AbstractVector, b::AbstractArray) = pinv(a) * b

stdlib/LinearAlgebra/src/lu.jl

@@ -76,26 +76,26 @@ adjoint(F::LU) = Adjoint(F)
 transpose(F::LU) = Transpose(F)
 
 # StridedMatrix
-lu!(A::StridedMatrix{<:BlasFloat}; check::Bool = true) = lu!(A, RowMax(); check=check)
-function lu!(A::StridedMatrix{T}, ::RowMax; check::Bool = true) where {T<:BlasFloat}
+lu!(A::StridedMatrix{<:BlasFloat}; check::Bool = true) = lu!(A, RowMaximum(); check=check)
+function lu!(A::StridedMatrix{T}, ::RowMaximum; check::Bool = true) where {T<:BlasFloat}
     lpt = LAPACK.getrf!(A)
     check && checknonsingular(lpt[3])
     return LU{T,typeof(A)}(lpt[1], lpt[2], lpt[3])
 end
 function lu!(A::StridedMatrix{<:BlasFloat}, pivot::NoPivot; check::Bool = true)
     return generic_lufact!(A, pivot; check = check)
 end
-function lu!(A::HermOrSym, pivot::PivotingStrategy = RowMax(); check::Bool = true)
+function lu!(A::HermOrSym, pivot::Union{RowMaximum,NoPivot} = RowMaximum(); check::Bool = true)
     copytri!(A.data, A.uplo, isa(A, Hermitian))
     lu!(A.data, pivot; check = check)
 end
 # for backward compatibility
 # TODO: remove towards Julia v2
-@deprecate lu!(A::Union{StridedMatrix,HermOrSym,Tridiagonal}, ::Val{true}; check::Bool = true) lu!(A, RowMax(); check=check)
+@deprecate lu!(A::Union{StridedMatrix,HermOrSym,Tridiagonal}, ::Val{true}; check::Bool = true) lu!(A, RowMaximum(); check=check)
 @deprecate lu!(A::Union{StridedMatrix,HermOrSym,Tridiagonal}, ::Val{false}; check::Bool = true) lu!(A, NoPivot(); check=check)
 
 """
-    lu!(A, pivot = RowMax(); check = true) -> LU
+    lu!(A, pivot = RowMaximum(); check = true) -> LU
 
 `lu!` is the same as [`lu`](@ref), but saves space by overwriting the
 input `A`, instead of creating a copy. An [`InexactError`](@ref)
@@ -131,26 +131,22 @@ Stacktrace:
 [...]
 ```
 """
-lu!(A::StridedMatrix, pivot::PivotingStrategy = RowMax(); check::Bool = true) =
+lu!(A::StridedMatrix, pivot::Union{RowMaximum,NoPivot} = RowMaximum(); check::Bool = true) =
     generic_lufact!(A, pivot; check = check)
-function generic_lufact!(A::StridedMatrix{T}, pivot::PivotingStrategy = RowMax();
+function generic_lufact!(A::StridedMatrix{T}, pivot::Union{RowMaximum,NoPivot} = RowMaximum();
                          check::Bool = true) where {T}
     # Extract values
     m, n = size(A)
     minmn = min(m,n)
 
-    if pivot !== RowMax() && pivot !== NoPivot()
-        throw(ArgumentError("only `RowMax()` and `NoPivot()` are supported as `pivot` argument but you supplied `$pivot`"))
-    end
-
     # Initialize variables
     info = 0
     ipiv = Vector{BlasInt}(undef, minmn)
     @inbounds begin
         for k = 1:minmn
             # find index max
             kp = k
-            if pivot === RowMax() && k < m
+            if pivot === RowMaximum() && k < m
                 amax = abs(A[k, k])
                 for i = k+1:m
                     absi = abs(A[i,k])
@@ -211,7 +207,7 @@ end
 
 # for all other types we must promote to a type which is stable under division
 """
-    lu(A, pivot = RowMax(); check = true) -> F::LU
+    lu(A, pivot = RowMaximum(); check = true) -> F::LU
 
 Compute the LU factorization of `A`.
 
@@ -278,12 +274,12 @@ julia> l == F.L && u == F.U && p == F.p
 true
 ```
 """
-function lu(A::AbstractMatrix{T}, pivot::PivotingStrategy = RowMax(); check::Bool = true) where {T}
+function lu(A::AbstractMatrix{T}, pivot::Union{RowMaximum,NoPivot} = RowMaximum(); check::Bool = true) where {T}
     S = lutype(T)
     lu!(copy_oftype(A, S), pivot; check = check)
 end
 # TODO: remove for Julia v2.0
-@deprecate lu(A::AbstractMatrix, ::Val{true}; check::Bool = true) lu(A, RowMax(); check=check)
+@deprecate lu(A::AbstractMatrix, ::Val{true}; check::Bool = true) lu(A, RowMaximum(); check=check)
 @deprecate lu(A::AbstractMatrix, ::Val{false}; check::Bool = true) lu(A, NoPivot(); check=check)
 
 
@@ -495,14 +491,10 @@ inv(A::LU{<:BlasFloat,<:StridedMatrix}) = inv!(copy(A))
 # Tridiagonal
 
 # See dgttrf.f
-function lu!(A::Tridiagonal{T,V}, pivot::PivotingStrategy = RowMax(); check::Bool = true) where {T,V}
+function lu!(A::Tridiagonal{T,V}, pivot::Union{RowMaximum,NoPivot} = RowMaximum(); check::Bool = true) where {T,V}
     # Extract values
     n = size(A, 1)
 
-    if pivot !== RowMax() && pivot !== NoPivot()
-        throw(ArgumentError("only `RowMax()` and `NoPivot()` are supported as `pivot` argument but you supplied `$pivot`"))
-    end
-
     # Initialize variables
     info = 0
     ipiv = Vector{BlasInt}(undef, n)

stdlib/LinearAlgebra/src/qr.jl

@@ -248,7 +248,7 @@ end
 # LAPACK version
 qr!(A::StridedMatrix{<:BlasFloat}, ::NoPivot; blocksize=36) =
     QRCompactWY(LAPACK.geqrt!(A, min(min(size(A)...), blocksize))...)
-qr!(A::StridedMatrix{<:BlasFloat}, ::ColNorm) = QRPivoted(LAPACK.geqp3!(A)...)
+qr!(A::StridedMatrix{<:BlasFloat}, ::ColumnNorm) = QRPivoted(LAPACK.geqp3!(A)...)
 
 # Generic fallbacks
 
@@ -293,10 +293,10 @@ Stacktrace:
 ```
 """
 qr!(A::AbstractMatrix, ::NoPivot) = qrfactUnblocked!(A)
-qr!(A::AbstractMatrix, ::ColNorm) = qrfactPivotedUnblocked!(A)
+qr!(A::AbstractMatrix, ::ColumnNorm) = qrfactPivotedUnblocked!(A)
 qr!(A::AbstractMatrix) = qr!(A, NoPivot())
 # TODO: Remove in Julia v2.0
-@deprecate qr!(A::AbstractMatrix, ::Val{true})  qr!(A, ColNorm())
+@deprecate qr!(A::AbstractMatrix, ::Val{true})  qr!(A, ColumnNorm())
 @deprecate qr!(A::AbstractMatrix, ::Val{false}) qr!(A, NoPivot())
 
 _qreltype(::Type{T}) where T = typeof(zero(T)/sqrt(abs2(one(T))))
@@ -313,7 +313,7 @@ A = Q R
 
 The returned object `F` stores the factorization in a packed format:
 
- - if `pivot == ColNorm()` then `F` is a [`QRPivoted`](@ref) object,
+ - if `pivot == ColumnNorm()` then `F` is a [`QRPivoted`](@ref) object,
 
  - otherwise if the element type of `A` is a BLAS type ([`Float32`](@ref), [`Float64`](@ref),
    `ComplexF32` or `ComplexF64`), then `F` is a [`QRCompactWY`](@ref) object,
@@ -387,7 +387,7 @@ function qr(A::AbstractMatrix{T}, arg...; kwargs...) where T
 end
 # TODO: remove in Julia v2.0
 @deprecate qr(A::AbstractMatrix, ::Val{false}; kwargs...) qr(A, NoPivot(); kwargs...)
-@deprecate qr(A::AbstractMatrix, ::Val{true}; kwargs...)  qr(A, ColNorm(); kwargs...)
+@deprecate qr(A::AbstractMatrix, ::Val{true}; kwargs...)  qr(A, ColumnNorm(); kwargs...)
 
 qr(x::Number) = qr(fill(x,1,1))
 function qr(v::AbstractVector)

stdlib/LinearAlgebra/test/diagonal.jl

@@ -565,7 +565,7 @@ end
     D = Diagonal(randn(5))
     Q = qr(randn(5, 5)).Q
     @test D * Q' == Array(D) * Q'
-    Q = qr(randn(5, 5), ColNorm()).Q
+    Q = qr(randn(5, 5), ColumnNorm()).Q
     @test_throws ArgumentError lmul!(Q, D)
 end
 

stdlib/LinearAlgebra/test/generic.jl

@@ -388,7 +388,7 @@ LinearAlgebra.Transpose(a::ModInt{n}) where {n} = transpose(a)
     Base.abs(a::ModInt{n}) where {n} = a
     Base.:<(a::ModInt{n}, b::ModInt{n}) where {n} = a.k < b.k
 
-    @test A*(lu(A, RowMax())\b) == b
+    @test A*(lu(A, RowMaximum())\b) == b
 end
 
 @testset "Issue 18742" begin

stdlib/LinearAlgebra/test/lq.jl

@@ -40,7 +40,7 @@ rectangularQ(Q::LinearAlgebra.LQPackedQ) = convert(Array, Q)
                 lqa   = lq(a)
                 x = lqa\b
                 l,q   = lqa.L, lqa.Q
-                qra   = qr(a, ColNorm())
+                qra   = qr(a, ColumnNorm())
                 @testset "Basic ops" begin
                     @test size(lqa,1) == size(a,1)
                     @test size(lqa,3) == 1

stdlib/LinearAlgebra/test/qr.jl

@@ -102,7 +102,7 @@ rectangularQ(Q::LinearAlgebra.AbstractQ) = convert(Array, Q)
                 @test Base.propertynames(qra)       == (:R, :Q)
             end
             @testset "(Automatic) Fat (pivoted) QR decomposition" begin
-                @inferred qr(a, ColNorm())
+                @inferred qr(a, ColumnNorm())
 
                 qrpa  = factorize(a[1:n1,:])
                 q,r = qrpa.Q, qrpa.R
@@ -254,7 +254,7 @@ end
     A = zeros(1, 2)
     B = zeros(1, 1)
     @test A \ B == zeros(2, 1)
-    @test qr(A, ColNorm()) \ B == zeros(2, 1)
+    @test qr(A, ColumnNorm()) \ B == zeros(2, 1)
 end
 
 @testset "Issue 24107" begin
@@ -276,7 +276,7 @@ end
     @test A \b ≈ ldiv!(c, qr(A ), b)
     @test b == b0
     c0 = copy(c)
-    @test Ac\c ≈ ldiv!(b, qr(Ac, ColNorm()), c)
+    @test Ac\c ≈ ldiv!(b, qr(Ac, ColumnNorm()), c)
     @test c0 == c
 end
 
@@ -293,7 +293,7 @@ end
 
 @testset "det(Q::Union{QRCompactWYQ, QRPackedQ})" begin
     # 40 is the number larger than the default block size 36 of QRCompactWY
-    @testset for n in [1:3; 40], m in [1:3; 40], pivot in (NoPivot(), ColNorm())
+    @testset for n in [1:3; 40], m in [1:3; 40], pivot in (NoPivot(), ColumnNorm())
         @testset "real" begin
             @testset for k in 0:min(n, m, 5)
                 A = cat(Array(I(k)), randn(n - k, m - k); dims=(1, 2))

stdlib/LinearAlgebra/test/special.jl

@@ -192,7 +192,7 @@ end
         a = rand(n,n)
         atri = typ(a)
         b = rand(n,n)
-        qrb = qr(b, ColNorm())
+        qrb = qr(b, ColumnNorm())
         @test *(atri, adjoint(qrb.Q)) ≈ Matrix(atri) * qrb.Q'
         @test rmul!(copy(atri), adjoint(qrb.Q)) ≈ Matrix(atri) * qrb.Q'
         qrb = qr(b, NoPivot())

stdlib/LinearAlgebra/test/uniformscaling.jl

@@ -500,7 +500,7 @@ end
 
 @testset "Factorization solutions" begin
     J = complex(randn(),randn()) * I
-    qrp = A -> qr(A, ColNorm())
+    qrp = A -> qr(A, ColumnNorm())
 
     # thin matrices
     X = randn(3,2)