gnu: Add apache-arrow-for-ceph.

* gnu/packages/databases.scm (apache-arrow-for-ceph): New variable.

1 files changed, 132 insertions, 0 deletions

diff --git a/gnu/packages/databases.scm b/gnu/packages/databases.scm
index 83515ad17e..2a24ba580d 100644
--- a/gnu/packages/databases.scm
+++ b/gnu/packages/databases.scm
@@ -4310,6 +4310,138 @@ language-bindings for structure manipulation.  It also provides IPC and common
 algorithm implementations.")
     (license license:asl2.0)))
 
+(define-public apache-arrow-for-ceph
+  (package
+    (name "apache-arrow")
+    (version "6.0.1")
+    (source
+     (origin
+       (method git-fetch)
+       (uri (git-reference
+             (url "https://github.com/apache/arrow")
+             (commit (string-append "apache-arrow-" version))))
+       (file-name (git-file-name name version))
+       (sha256
+        (base32
+         "0mcw361akqw4sxnnpnr9c9v1zk4hphk6gcq763pcb19yzljh88ig"))))
+    (build-system cmake-build-system)
+    (arguments
+     `(#:tests? #f
+       #:phases
+       (modify-phases %standard-phases
+         (add-before 'configure 'enter-source-directory
+           (lambda _ (chdir "cpp")))
+         (add-after 'unpack 'set-env
+           (lambda* (#:key inputs #:allow-other-keys)
+             (substitute* "cpp/src/parquet/parquet.pc.in"
+               (("includedir=\\$\\{prefix\\}/")
+                "includedir="))
+             (substitute* "cpp/cmake_modules/ThirdpartyToolchain.cmake"
+               (("set\\(xsimd_SOURCE.*") ""))
+             (setenv "BOOST_ROOT" (assoc-ref inputs "boost"))
+             (setenv "BROTLI_HOME" (assoc-ref inputs "brotli"))
+             (setenv "FLATBUFFERS_HOME" (assoc-ref inputs "flatbuffers"))
+             (setenv "RAPIDJSON_HOME" (assoc-ref inputs "rapidjson")))))
+       #:build-type "Release"
+       #:configure-flags
+       (list "-DARROW_PYTHON=ON"
+             "-DARROW_GLOG=ON"
+             ;; Parquet options
+             "-DARROW_PARQUET=ON"
+             "-DPARQUET_BUILD_EXECUTABLES=ON"
+             ;; The maintainers disallow using system versions of
+             ;; jemalloc:
+             ;; https://issues.apache.org/jira/browse/ARROW-3507. This
+             ;; is unfortunate because jemalloc increases performance:
+             ;; https://arrow.apache.org/blog/2018/07/20/jemalloc/.
+             "-DARROW_JEMALLOC=OFF"
+
+             ;; The CMake option ARROW_DEPENDENCY_SOURCE is a global
+             ;; option that instructs the build system how to resolve
+             ;; each dependency. SYSTEM = Finding the dependency in
+             ;; system paths using CMake's built-in find_package
+             ;; function, or using pkg-config for packages that do not
+             ;; have this feature
+             "-DARROW_DEPENDENCY_SOURCE=SYSTEM"
+             "-Dxsimd_SOURCE=SYSTEM"
+
+             "-DARROW_RUNTIME_SIMD_LEVEL=NONE"
+             "-DARROW_SIMD_LEVEL=NONE"
+             "-DARROW_PACKAGE_KIND=Guix"
+
+             ;; Split output into its component packages.
+             (string-append "-DCMAKE_INSTALL_PREFIX="
+                            (assoc-ref %outputs "lib"))
+             (string-append "-DCMAKE_INSTALL_RPATH="
+                            (assoc-ref %outputs "lib")
+                            "/lib")
+             (string-append "-DCMAKE_INSTALL_BINDIR="
+                            (assoc-ref %outputs "out")
+                            "/bin")
+             (string-append "-DCMAKE_INSTALL_INCLUDEDIR="
+                            (assoc-ref %outputs "include")
+                            "/share/include")
+
+             "-DARROW_WITH_SNAPPY=ON"
+             "-DARROW_WITH_ZLIB=ON"
+             "-DARROW_WITH_ZSTD=ON"
+             "-DARROW_WITH_LZ4=ON"
+             "-DARROW_COMPUTE=ON"
+             "-DARROW_CSV=ON"
+             "-DARROW_DATASET=ON"
+             "-DARROW_FILESYSTEM=ON"
+             "-DARROW_HDFS=ON"
+             "-DARROW_JSON=ON"
+             ;; Arrow Python C++ integration library (required for
+             ;; building pyarrow). This library must be built against
+             ;; the same Python version for which you are building
+             ;; pyarrow. NumPy must also be installed. Enabling this
+             ;; option also enables ARROW_COMPUTE, ARROW_CSV,
+             ;; ARROW_DATASET, ARROW_FILESYSTEM, ARROW_HDFS, and
+             ;; ARROW_JSON.
+             "-DARROW_PYTHON=ON"
+
+             ;; Building the tests forces on all the
+             ;; optional features and the use of static
+             ;; libraries.
+             "-DARROW_BUILD_TESTS=OFF"
+             "-DBENCHMARK_ENABLE_GTEST_TESTS=OFF"
+             ;;"-DBENCHMARK_ENABLE_TESTING=OFF"
+             "-DARROW_BUILD_STATIC=OFF")))
+    (inputs
+     (list boost
+           brotli
+           bzip2
+           double-conversion
+           gflags
+           glog
+           grpc
+           protobuf
+           python
+           python-numpy
+           rapidjson
+           re2
+           snappy
+           xsimd))
+    ;; These are all listed under Requires.private in arrow.pc
+    (propagated-inputs
+     (list (list apache-thrift "lib")
+           lz4
+           utf8proc
+           zlib
+           (list zstd "lib")))
+    (native-inputs
+     (list pkg-config))
+    (outputs '("out" "lib" "include"))
+    (home-page "https://arrow.apache.org/")
+    (synopsis "Columnar in-memory analytics")
+    (description "Apache Arrow is a columnar in-memory analytics layer
+designed to accelerate big data.  It houses a set of canonical in-memory
+representations of flat and hierarchical data along with multiple
+language-bindings for structure manipulation.  It also provides IPC and common
+algorithm implementations.")
+    (license license:asl2.0)))
+
 (define-public apache-arrow-0.16
   (package
     (name "apache-arrow")