Add sparse!(I, J, V) and spzeros!(I, J) (#397)

This patch adds convenience wrappers around the "full" calls to
`sparse!` and `spzeros` that allocates the intermediate arrays of
appropriate size, but reuses the input vectors for the final matrix
storage.

These methods are useful in the quite common case where the COO
representation is constructed with the sole purpose of constructing the
sparse matrix and then thrown away.

docs/src/index.md

@@ -211,12 +211,14 @@ SparseArrays.AbstractSparseMatrix
 SparseArrays.SparseVector
 SparseArrays.SparseMatrixCSC
 SparseArrays.sparse
+SparseArrays.sparse!
 SparseArrays.sparsevec
 Base.similar(::SparseArrays.AbstractSparseMatrixCSC, ::Type)
 SparseArrays.issparse
 SparseArrays.nnz
 SparseArrays.findnz
 SparseArrays.spzeros
+SparseArrays.spzeros!
 SparseArrays.spdiagm
 SparseArrays.sparse_hcat
 SparseArrays.sparse_vcat

src/sparsematrix.jl

@@ -1270,6 +1270,34 @@ function sparse!(I::AbstractVector{Ti}, J::AbstractVector{Ti},
  Vector{Ti}(undef, n+1), Vector{Ti}(), Vector{Tv}())
 end
 
+"""
+ SparseArrays.sparse!(I, J, V, [m, n, combine]) -> SparseMatrixCSC
+
+Variant of `sparse!` that re-uses the input vectors (`I`, `J`, `V`) for the final matrix
+storage. After construction the input vectors will alias the matrix buffers; `S.colptr ===
+I`, `S.rowval === J`, and `S.nzval === V` holds, and they will be `resize!`d as necessary.
+
+Note that some work buffers will still be allocated. Specifically, this method is a
+convenience wrapper around `sparse!(I, J, V, m, n, combine, klasttouch, csrrowptr,
+csrcolval, csrnzval, csccolptr, cscrowval, cscnzval)` where this method allocates
+`klasttouch`, `csrrowptr`, `csrcolval`, and `csrnzval` of appropriate size, but reuses `I`,
+`J`, and `V` for `csccolptr`, `cscrowval`, and `cscnzval`.
+
+Arguments `m`, `n`, and `combine` defaults to `maximum(I)`, `maximum(J)`, and `+`,
+respectively.
+
+!!! compat "Julia 1.10"
+ This method requires Julia version 1.10 or later.
+"""
+function sparse!(I::AbstractVector{Ti}, J::AbstractVector{Ti}, V::AbstractVector{Tv},
+ m::Integer=dimlub(I), n::Integer=dimlub(J), combine::Function=+) where {Tv, Ti<:Integer}
+ klasttouch = Vector{Ti}(undef, n)
+ csrrowptr = Vector{Ti}(undef, m + 1)
+ csrcolval = Vector{Ti}(undef, length(I))
+ csrnzval = Vector{Tv}(undef, length(I))
+ sparse!(I, J, V, Int(m), Int(n), combine, klasttouch, csrrowptr, csrcolval, csrnzval, I, J, V)
+end
+
 dimlub(I) = isempty(I) ? 0 : Int(maximum(I)) #least upper bound on required sparse matrix dimension
 
 sparse(I,J,v::Number) = sparse(I, J, fill(v,length(I)))
@@ -2116,6 +2144,31 @@ function spzeros!(::Type{Tv}, I::AbstractVector{Ti}, J::AbstractVector{Ti}, m::I
  csrrowptr, csrcolval, cscnzval, csccolptr, cscrowval, cscnzval)
 end
 
+"""
+ SparseArrays.spzeros!(::Type{Tv}, I, J, [m, n]) -> SparseMatrixCSC{Tv}
+
+Variant of `spzeros!` that re-uses the input vectors `I` and `J` for the final matrix
+storage. After construction the input vectors will alias the matrix buffers; `S.colptr ===
+I` and `S.rowval === J` holds, and they will be `resize!`d as necessary.
+
+Note that some work buffers will still be allocated. Specifically, this method is a
+convenience wrapper around `spzeros!(Tv, I, J, m, n, klasttouch, csrrowptr, csrcolval,
+csccolptr, cscrowval)` where this method allocates `klasttouch`, `csrrowptr`, and
+`csrcolval` of appropriate size, but reuses `I` and `J` for `csccolptr` and `cscrowval`.
+
+Arguments `m` and `n` defaults to `maximum(I)` and `maximum(J)`.
+
+!!! compat "Julia 1.10"
+ This method requires Julia version 1.10 or later.
+"""
+function spzeros!(::Type{Tv}, I::AbstractVector{Ti}, J::AbstractVector{Ti},
+ m::Integer=dimlub(I), n::Integer=dimlub(J)) where {Tv, Ti <: Integer}
+ klasttouch = Vector{Ti}(undef, n)
+ csrrowptr = Vector{Ti}(undef, m + 1)
+ csrcolval = Vector{Ti}(undef, length(I))
+ return spzeros!(Tv, I, J, Int(m), Int(n), klasttouch, csrrowptr, csrcolval, I, J)
+end
+
 import Base._one
 function Base._one(unit::T, S::AbstractSparseMatrixCSC) where T
  size(S, 1) == size(S, 2) || throw(DimensionMismatch("multiplicative identity only defined for square matrices"))

test/sparsematrix_constructors_indexing.jl

@@ -1751,6 +1751,52 @@ end
  @test getcolptr(S!) === I
  @test getrowval(S!) === J
  @test nonzeros(S!) === V
+
+ # Test reuse of I, J, V for the matrix buffers in
+ # sparse!(I, J, V), sparse!(I, J, V, m, n), sparse!(I, J, V, m, n, combine),
+ # spzeros!(T, I, J), and spzeros!(T, I, J, m, n).
+ I, J, V = allocate_arrays(m, n)
+ S = sparse(I, J, V)
+ S! = sparse!(I, J, V)
+ @test S == S!
+ @test same_structure(S, S!)
+ @test getcolptr(S!) === I
+ @test getrowval(S!) === J
+ @test nonzeros(S!) === V
+ I, J, V = allocate_arrays(m, n)
+ S = sparse(I, J, V, 2m, 2n)
+ S! = sparse!(I, J, V, 2m, 2n)
+ @test S == S!
+ @test same_structure(S, S!)
+ @test getcolptr(S!) === I
+ @test getrowval(S!) === J
+ @test nonzeros(S!) === V
+ I, J, V = allocate_arrays(m, n)
+ S = sparse(I, J, V, 2m, 2n, *)
+ S! = sparse!(I, J, V, 2m, 2n, *)
+ @test S == S!
+ @test same_structure(S, S!)
+ @test getcolptr(S!) === I
+ @test getrowval(S!) === J
+ @test nonzeros(S!) === V
+ for T in (Float32, Float64)
+ I, J, = allocate_arrays(m, n)
+ S = spzeros(T, I, J)
+ S! = spzeros!(T, I, J)
+ @test S == S!
+ @test same_structure(S, S!)
+ @test eltype(S) == eltype(S!) == T
+ @test getcolptr(S!) === I
+ @test getrowval(S!) === J
+ I, J, = allocate_arrays(m, n)
+ S = spzeros(T, I, J, 2m, 2n)
+ S! = spzeros!(T, I, J, 2m, 2n)
+ @test S == S!
+ @test same_structure(S, S!)
+ @test eltype(S) == eltype(S!) == T
+ @test getcolptr(S!) === I
+ @test getrowval(S!) === J
+ end
  end
 end