Change to ArrayInterfaceCore

Project.toml

@@ -4,7 +4,7 @@ authors = ["Kanav Gupta <[email protected]>"]
 version = "0.3.16"
 
 [deps]
-ArrayInterface = "4fba245c-0d91-5ea0-9b3e-6abc04ee57a9"
+ArrayInterfaceCore = "30b0a656-2188-435a-8636-2ec0e6a096e2"
 FiniteDiff = "6a86dc24-6348-571c-b903-95158fe2bd41"
 ForwardDiff = "f6369f11-7733-5829-9624-2563aa707210"
 IterativeSolvers = "42fd0dbc-a981-5370-80f2-aaf504508153"
@@ -18,7 +18,7 @@ StaticArrays = "90137ffa-7385-5640-81b9-e52037218182"
 UnPack = "3a884ed6-31ef-47d7-9d2a-63182c4928ed"
 
 [compat]
-ArrayInterface = "3, 4, 5"
+ArrayInterfaceCore = "0.1.1"
 FiniteDiff = "2"
 ForwardDiff = "0.10.3"
 IterativeSolvers = "0.9"

src/NonlinearSolve.jl

@@ -8,7 +8,7 @@ using Setfield
 using StaticArrays
 using RecursiveArrayTools
 using LinearAlgebra
-import ArrayInterface
+import ArrayInterfaceCore
 import IterativeSolvers
 import RecursiveFactorization
 

src/utils.jl

@@ -128,7 +128,7 @@ function (p::DefaultLinSolve)(x,A,b,update_matrix=false;tol=nothing, kwargs...)
       #
       # RecursiveFactorization seems to be consistantly winning below 100
       # https://discourse.julialang.org/t/ann-recursivefactorization-jl/39213
-      if ArrayInterface.can_setindex(x) && (size(A,1) <= 100 || ((blasvendor === :openblas || blasvendor === :openblas64) && size(A,1) <= 500))
+      if ArrayInterfaceCore.can_setindex(x) && (size(A,1) <= 100 || ((blasvendor === :openblas || blasvendor === :openblas64) && size(A,1) <= 500))
         p.A = RecursiveFactorization.lu!(A)
       else
         p.A = lu!(A)
@@ -141,7 +141,7 @@ function (p::DefaultLinSolve)(x,A,b,update_matrix=false;tol=nothing, kwargs...)
       p.A = bunchkaufman!(A)
     elseif typeof(A) <: SparseMatrixCSC
       p.A = lu(A)
-    elseif ArrayInterface.isstructured(A)
+    elseif ArrayInterfaceCore.isstructured(A)
       p.A = factorize(A)
     elseif !(typeof(A) <: AbstractDiffEqOperator)
       # Most likely QR is the one that is overloaded
@@ -156,7 +156,7 @@ function (p::DefaultLinSolve)(x,A,b,update_matrix=false;tol=nothing, kwargs...)
   # Missing a little bit of efficiency in a rare case
   #elseif typeof(A) <: DiffEqArrayOperator
   #  ldiv!(x,p.A,b)
-  elseif ArrayInterface.isstructured(A) || A isa SparseMatrixCSC
+  elseif ArrayInterfaceCore.isstructured(A) || A isa SparseMatrixCSC
     ldiv!(x,p.A,b)
   elseif typeof(A) <: AbstractDiffEqOperator
     # No good starting guess, so guess zero