Upgrade to Julia 1.3+, use JLL package to provide FFTW (#122)

This is the first package where I've tried to mix JLL packages with more
custom installation procedures, but it seems to more-or-less work.

.cirrus.yml

@@ -4,8 +4,7 @@ task:
   name: FreeBSD
   env:
     matrix:
-      - JULIA_VERSION: 1.0
-      - JULIA_VERSION: 1.1
+      - JULIA_VERSION: 1.3
       - JULIA_VERSION: nightly
   install_script:
     - sh -c "$(fetch https://raw.githubusercontent.com/ararslan/CirrusCI.jl/master/bin/install.sh -o -)"

.drone.yml

@@ -12,34 +12,6 @@ steps:
   commands:
   - "julia --project=. --check-bounds=yes --color=yes -e 'using InteractiveUtils; versioninfo(verbose=true); using Pkg; Pkg.build(); Pkg.test(coverage=true)'"
 
----
-kind: pipeline
-name: linux - arm64 - Julia 1.0
-
-platform:
-  os: linux
-  arch: arm64
-
-steps:
-- name: build
-  image: julia:1.0
-  commands:
-  - "julia --project=. --check-bounds=yes --color=yes -e 'using InteractiveUtils; versioninfo(verbose=true); using Pkg; Pkg.build(); Pkg.test(coverage=true)'"
-
----
-kind: pipeline
-name: linux - arm64 - Julia 1.2
-
-platform:
-  os: linux
-  arch: arm64
-
-steps:
-- name: build
-  image: julia:1.2
-  commands:
-  - "julia --project=. --check-bounds=yes --color=yes -e 'using InteractiveUtils; versioninfo(verbose=true); using Pkg; Pkg.build(); Pkg.test(coverage=true)'"
-
 ---
 kind: pipeline
 name: linux - arm64 - Julia 1.3

.travis.yml

@@ -4,8 +4,7 @@ os:
   - linux
   - osx
 julia:
-  - 1.0
-  - 1.1
+  - 1.3
   - nightly
 env:
   matrix:
@@ -24,7 +23,7 @@ after_success:
 jobs:
   include:
     - stage: "Documentation"
-      julia: 1.0
+      julia: 1.3
       os: linux
       script:
         - julia --project=docs/ -e 'using Pkg; Pkg.develop(PackageSpec(path=pwd()));

Project.toml

@@ -4,18 +4,17 @@ version = "1.1.0"
 
 [deps]
 AbstractFFTs = "621f4979-c628-5d54-868e-fcf4e3e8185c"
-BinaryProvider = "b99e7846-7c00-51b0-8f62-c81ae34c0232"
 Conda = "8f4d0f93-b110-5947-807f-2305c1781a2d"
+FFTW_jll = "f5851436-0d7a-5f13-b9de-f02708fd171a"
 Libdl = "8f399da3-3557-5675-b5ff-fb832c97cbdb"
 LinearAlgebra = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
 Reexport = "189a3867-3050-52da-a836-e630ba90ab69"
 
 [compat]
 AbstractFFTs = "0.5"
-BinaryProvider = "0.5"
 Conda = "1"
 Reexport = "0.2"
-julia = "1"
+julia = "1.3"
 
 [extras]
 Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"

appveyor.yml

@@ -1,7 +1,6 @@
 environment:
   matrix:
-  - julia_version: 1
-  - julia_version: 1.1
+  - julia_version: 1.3
   - julia_version: nightly
 
 platform:

deps/build.jl

@@ -3,9 +3,6 @@ using LinearAlgebra
 import Libdl
 const depsfile = joinpath(@__DIR__, "deps.jl")
 
-# If BLAS was compiled with MKL and the user wants MKL-based FFTs, we'll oblige.
-# In that case, we have to do this little dance to get around having to use BinDeps
-# for a library that's already linked to Julia.
 settings = joinpath(first(DEPOT_PATH), "prefs", "FFTW")
 mkpath(dirname(settings))
 if haskey(ENV, "JULIA_FFTW_PROVIDER")
@@ -17,7 +14,9 @@ else
     provider = "FFTW"
     open(f -> println(f, provider), settings, "w")
 end
+
 if provider == "MKL"
+    # If BLAS was compiled with MKL and the user wants MKL-based FFTs, we'll oblige.
     if BLAS.vendor() === :mkl
         mklpath = Libdl.dlpath("libmkl_rt")
     else
@@ -41,10 +40,15 @@ if provider == "MKL"
             end
         """)
     end
-elseif provider != "FFTW"
+elseif provider == "FFTW"
+    open(depsfile, "w") do io
+        println(io, """
+            using FFTW_jll
+            check_deps() = nothing
+        """)
+    end
+else
     error("Unrecognized JULIA_FFTW_PROVIDER \"$provider\".\n",
           "To fix this, set ENV[\"JULIA_FFTW_PROVIDER\"] to \"FFTW\" or \"MKL\"\n",
           "and rerun Pkg.build(\"FFTW\").")
-else
-    include("build_fftw.jl")
 end