Merge pull request #216 from mohamed82008/master

Final 0.7 deprecations

Author: haampie

benchmark/benchmark-linear-systems.jl

@@ -45,7 +45,7 @@ end
 
 function indefinite(n)
     # Generate an indefinite "hard" matrix
-    srand(1)
+    Random.seed!(1)
     A = speye(n) + sprand(n, n, 5.0 / n)
     A = (A + A') / 2
     x = ones(n)
@@ -68,10 +68,10 @@ end
 function bicgstabl()
     A, b = advection_dominated()
 
-    b1 = @benchmark IterativeSolvers.bicgstabl($A, $b, 2, max_mv_products = 1000, convex_combination = false) setup = (srand(1))
-    b2 = @benchmark IterativeSolvers.bicgstabl($A, $b, 2, max_mv_products = 1000, convex_combination = true) setup = (srand(1))
-    b3 = @benchmark IterativeSolvers.bicgstabl($A, $b, 4, max_mv_products = 1000, convex_combination = false) setup = (srand(1))
-    b4 = @benchmark IterativeSolvers.bicgstabl($A, $b, 4, max_mv_products = 1000, convex_combination = true) setup = (srand(1))
+    b1 = @benchmark IterativeSolvers.bicgstabl($A, $b, 2, max_mv_products = 1000, convex_combination = false) setup = (Random.seed!(1))
+    b2 = @benchmark IterativeSolvers.bicgstabl($A, $b, 2, max_mv_products = 1000, convex_combination = true) setup = (Random.seed!(1))
+    b3 = @benchmark IterativeSolvers.bicgstabl($A, $b, 4, max_mv_products = 1000, convex_combination = false) setup = (Random.seed!(1))
+    b4 = @benchmark IterativeSolvers.bicgstabl($A, $b, 4, max_mv_products = 1000, convex_combination = true) setup = (Random.seed!(1))
 
     b1, b2, b3, b4
 end

src/simple.jl

@@ -52,7 +52,7 @@ end
 
 function powm_iterable!(A, x; tol = eps(real(eltype(A))) * size(A, 2) ^ 3, maxiter = size(A, 1))
     T = eltype(x)
-    PowerMethodIterable(A, x, tol, maxiter, zero(T), similar(x), similar(x), realmax(real(T)))
+    PowerMethodIterable(A, x, tol, maxiter, zero(T), similar(x), similar(x), floatmax(real(T)))
 end
 
 """

test/bicgstabl.jl

@@ -5,7 +5,7 @@ using LinearAlgebra
 
 @testset ("BiCGStab(l)") begin
 
-srand(1234321)
+Random.seed!(1234321)
 n = 20
 
 @testset "Matrix{$T}" for T in (Float32, Float64, ComplexF32, ComplexF64)

test/cg.jl

@@ -17,7 +17,7 @@ ldiv!(y, P::JacobiPrec, x) = y .= x ./ P.diagonal
 
 @testset "Conjugate Gradients" begin
 
-srand(1234321)
+Random.seed!(1234321)
 
 @testset "Small full system" begin
     n = 10

test/chebyshev.jl

@@ -19,7 +19,7 @@ end
 @testset "Chebyshev" begin
 
 n = 10
-srand(1234321)
+Random.seed!(1234321)
 
 @testset "Matrix{$T}" for T in (Float32, Float64, ComplexF32, ComplexF64)
     A = randSPD(T, n)

test/common.jl

@@ -3,7 +3,7 @@ using IterativeSolvers
 using Test
 using Random
 
-srand(1234321)
+Random.seed!(1234321)
 
 @testset "Basic operations" begin
 

test/gmres.jl

@@ -8,7 +8,7 @@ using SparseArrays
 #GMRES
 @testset "GMRES" begin
 
-srand(1234321)
+Random.seed!(1234321)
 n = 10
 
 @testset "Matrix{$T}" for T in (Float32, Float64, ComplexF32, ComplexF64)

test/idrs.jl

@@ -7,7 +7,7 @@ using LinearAlgebra
 
 n = 10
 m = 6
-srand(1234567)
+Random.seed!(1234567)
 
 @testset "Matrix{$T}" for T in (Float32, Float64, ComplexF32, ComplexF64)
     A = rand(T, n, n) + n * I

test/lobpcg.jl

@@ -27,7 +27,7 @@ function max_err(R)
 end
 
 @testset "Locally Optimal Block Preconditioned Conjugate Gradient" begin
-    srand(1234323)
+    Random.seed!(1234323)
     @testset "Single eigenvalue" begin
         n = 10
         @testset "Small full system" begin

test/lsmr.jl

@@ -60,7 +60,7 @@ function sol_matrix(m, n)
 end
 
 @testset "LSMR" begin
-    srand(1234321)
+    Random.seed!(1234321)
 
     @testset "Small dense matrix" for T = (Float32, Float64)
         A = rand(T, 10, 5)