Fix depwarns on 0.6 (#26)

appveyor.yml

@@ -1,7 +1,9 @@
 environment:
-  matrix: 
+  matrix:
   - JULIAVERSION: "julialang/bin/winnt/x86/0.4/julia-0.4-latest-win32.exe"
   - JULIAVERSION: "julialang/bin/winnt/x64/0.4/julia-0.4-latest-win64.exe"
+  - JULIAVERSION: "julialang/bin/winnt/x86/0.5/julia-0.5-latest-win32.exe"
+  - JULIAVERSION: "julialang/bin/winnt/x64/0.5/julia-0.5-latest-win64.exe"
   - JULIAVERSION: "julianightlies/bin/winnt/x86/julia-latest-win32.exe"
   - JULIAVERSION: "julianightlies/bin/winnt/x64/julia-latest-win64.exe"
 
@@ -32,7 +34,7 @@ install:
 build_script:
 # Need to convert from shallow to complete for Pkg.clone to work
   - IF EXIST .git\shallow (git fetch --unshallow)
-  - C:\projects\julia\bin\julia -F -e "versioninfo();
+  - C:\projects\julia\bin\julia -e "versioninfo();
       Pkg.clone(pwd(), \"Cubature\"); Pkg.build(\"Cubature\")"
 
 test_script:

src/Cubature.jl

@@ -32,10 +32,10 @@ const FAILURE = convert(Int32, 1)
 # type to distinguish cubature error codes from thrown exceptions
 type NoError <: Exception end # used for integrand_error when nothing thrown
 
-type IntegrandData{F}
+@compat type IntegrandData{F}
     integrand_func::F
     integrand_error::Any
-    IntegrandData(f) = new(f, NoError())
+    (::Type{IntegrandData{F}}){F}(f) = new{F}(f, NoError())
 end
 IntegrandData{F}(f::F) = IntegrandData{F}(f)
 
@@ -161,8 +161,8 @@ function cubature{F}(xscalar::Bool, fscalar::Bool,
     end
     xmin = Float64[xmin_...]
     xmax = Float64[xmax_...]
-    val = Array(Float64, fdim)
-    err = Array(Float64, fdim)
+    val = Vector{Float64}(fdim)
+    err = Vector{Float64}(fdim)
     d = IntegrandData(f)
     fwrap = integrands(d, xscalar, fscalar, vectorized)
     # ccall's first arg needs to be a constant expression, so

test/runtests.jl

@@ -5,16 +5,16 @@ using Compat
 # The downside of having multiple interfaces to simplify 1d, scalar
 # integrands, etcetera, is that we have lot of interfaces to test:
 
-@test_approx_eq_eps hquadrature(cos, 0,1, abstol=1e-8)[1] sin(1) 1e-8
-@test_approx_eq_eps pquadrature(cos, 0,1, abstol=1e-8)[1] sin(1) 1e-8
+@test isapprox(hquadrature(cos, 0,1, abstol=1e-8)[1], sin(1), atol=1e-8)
+@test isapprox(pquadrature(cos, 0,1, abstol=1e-8)[1], sin(1), atol=1e-8)
 
 quad_tst1(x) = @compat prod(cos.(x))
 for dim = 0:3
     xmin = zeros(dim)
     xmax = 1:dim
     ans = @compat prod(sin.(xmax))
-    @test_approx_eq_eps hcubature(quad_tst1, xmin,xmax, abstol=1e-8)[1] ans 1e-8
-    @test_approx_eq_eps pcubature(quad_tst1, xmin,xmax, abstol=1e-8)[1] ans 1e-8
+    @test isapprox(hcubature(quad_tst1, xmin,xmax, abstol=1e-8)[1], ans, atol=1e-8)
+    @test isapprox(pcubature(quad_tst1, xmin,xmax, abstol=1e-8)[1], ans, atol=1e-8)
 end
 
 quad_tst2(x, v) = for j = 1:length(v)
@@ -26,19 +26,19 @@ for dim = 0:3
     xmax = 1:dim
     ans = @compat prod(sin.(xmax)) * (1:fdim)
     if dim == 1
-        @test_approx_eq_eps hquadrature(fdim, quad_tst2, xmin,xmax, abstol=1e-8)[1] ans 1e-8
-        @test_approx_eq_eps pquadrature(fdim, quad_tst2, xmin,xmax, abstol=1e-8)[1] ans 1e-8
+        @test isapprox(hquadrature(fdim, quad_tst2, xmin,xmax, abstol=1e-8)[1], ans, atol=1e-8)
+        @test isapprox(pquadrature(fdim, quad_tst2, xmin,xmax, abstol=1e-8)[1], ans, atol=1e-8)
     end
-    @test_approx_eq_eps hcubature(fdim, quad_tst2, xmin,xmax, abstol=1e-8)[1] ans 1e-8
-    @test_approx_eq_eps pcubature(fdim, quad_tst2, xmin,xmax, abstol=1e-8)[1] ans 1e-8
+    @test isapprox(hcubature(fdim, quad_tst2, xmin,xmax, abstol=1e-8)[1], ans, atol=1e-8)
+    @test isapprox(pcubature(fdim, quad_tst2, xmin,xmax, abstol=1e-8)[1], ans, atol=1e-8)
 end
 end
 
 quad_tst0v(x,v) = for i = 1:length(x)
     v[i] = cos(x[i])
 end
-@test_approx_eq_eps hquadrature_v(quad_tst0v, 0,1, abstol=1e-8)[1] sin(1) 1e-8
-@test_approx_eq_eps pquadrature_v(quad_tst0v, 0,1, abstol=1e-8)[1] sin(1) 1e-8
+@test isapprox(hquadrature_v(quad_tst0v, 0,1, abstol=1e-8)[1], sin(1), atol=1e-8)
+@test isapprox(pquadrature_v(quad_tst0v, 0,1, abstol=1e-8)[1], sin(1), atol=1e-8)
 
 quad_tst1v(x,v) = for i = 1:length(v)
     v[i] = @compat prod(cos.(x[:,i]))
@@ -47,8 +47,8 @@ for dim = 0:3
     xmin = zeros(dim)
     xmax = 1:dim
     ans = @compat prod(sin.(xmax))
-    @test_approx_eq_eps hcubature_v(quad_tst1v, xmin,xmax, abstol=1e-8)[1] ans 1e-8
-    @test_approx_eq_eps pcubature_v(quad_tst1v, xmin,xmax, abstol=1e-8)[1] ans 1e-8
+    @test isapprox(hcubature_v(quad_tst1v, xmin,xmax, abstol=1e-8)[1], ans, atol=1e-8)
+    @test isapprox(pcubature_v(quad_tst1v, xmin,xmax, abstol=1e-8)[1], ans, atol=1e-8)
 end
 
 quad_tst2v(x::Array{Float64,2}, v) = for i = 1:size(v,2)
@@ -67,16 +67,16 @@ for dim = 0:3
     xmax = 1:dim
     ans = @compat prod(sin.(xmax)) * (1:fdim)
     if dim == 1
-        @test_approx_eq_eps hquadrature_v(fdim, quad_tst2v, xmin,xmax, abstol=1e-8)[1] ans 1e-8
-        @test_approx_eq_eps pquadrature_v(fdim, quad_tst2v, xmin,xmax, abstol=1e-8)[1] ans 1e-8
+        @test isapprox(hquadrature_v(fdim, quad_tst2v, xmin,xmax, abstol=1e-8)[1], ans, atol=1e-8)
+        @test isapprox(pquadrature_v(fdim, quad_tst2v, xmin,xmax, abstol=1e-8)[1], ans, atol=1e-8)
     end
-    @test_approx_eq_eps hcubature_v(fdim, quad_tst2v, xmin,xmax, abstol=1e-8)[1] ans 1e-8
-    @test_approx_eq_eps pcubature_v(fdim, quad_tst2v, xmin,xmax, abstol=1e-8)[1] ans 1e-8
+    @test isapprox(hcubature_v(fdim, quad_tst2v, xmin,xmax, abstol=1e-8)[1], ans, atol=1e-8)
+    @test isapprox(pcubature_v(fdim, quad_tst2v, xmin,xmax, abstol=1e-8)[1], ans, atol=1e-8)
 end
 end
 
-# Check that nested calls work (although this is a suboptimal way to 
+# Check that nested calls work (although this is a suboptimal way to
 # compute multidimensional integrals):
 
-@test_approx_eq_eps hquadrature(x -> hquadrature(cos, 0,2, abstol=1e-8)[1]*cos(x),
-                                0,1, abstol=1e-8)[1] sin(1)*sin(2) 1e-8
+@test isapprox(hquadrature(x -> hquadrature(cos, 0,2, abstol=1e-8)[1]*cos(x),
+                           0,1, abstol=1e-8)[1], sin(1)*sin(2), atol=1e-8)