gnu: apache-arrow: Build with ORC.

* gnu/packages/databases.scm (apache-arrow)[arguments]: Set options needed to
build with ORC support; remove unused options
-DBENCHMARK_ENABLE_GTEST_TESTS=OFF and -DARROW_GLOG=ON.
[inputs]: Add apache-orc.

Change-Id: Icabe6e3a73478c8464ea1e8efc103899da8e5a86
This commit is contained in:
Ricardo Wurmus 2024-04-30 18:36:35 +02:00
parent 442eba98be
commit 8647aaaa14
No known key found for this signature in database
GPG key ID: 197A5888235FACAC

View file

@ -4626,68 +4626,69 @@ (define-public apache-arrow
(setenv "BROTLI_HOME" #$(this-package-input "brotli"))
(setenv "FLATBUFFERS_HOME" #$(this-package-input "flatbuffers"))
(setenv "RAPIDJSON_HOME" #$(this-package-input "rapidjson")))))
#:build-type "Release"
#:configure-flags
#~(list "-DARROW_PYTHON=ON"
"-DARROW_GLOG=ON"
;; Parquet options
"-DARROW_PARQUET=ON"
"-DPARQUET_BUILD_EXECUTABLES=ON"
;; The maintainers disallow using system versions of
;; jemalloc:
;; https://issues.apache.org/jira/browse/ARROW-3507. This
;; is unfortunate because jemalloc increases performance:
;; https://arrow.apache.org/blog/2018/07/20/jemalloc/.
"-DARROW_JEMALLOC=OFF"
#:build-type "Release"
#:configure-flags
#~(list "-DARROW_PYTHON=ON"
;; Parquet options
"-DARROW_PARQUET=ON"
"-DPARQUET_BUILD_EXECUTABLES=ON"
;; The maintainers disallow using system versions of
;; jemalloc:
;; https://issues.apache.org/jira/browse/ARROW-3507. This
;; is unfortunate because jemalloc increases performance:
;; https://arrow.apache.org/blog/2018/07/20/jemalloc/.
"-DARROW_JEMALLOC=OFF"
;; The CMake option ARROW_DEPENDENCY_SOURCE is a global
;; option that instructs the build system how to resolve
;; each dependency. SYSTEM = Finding the dependency in
;; system paths using CMake's built-in find_package
;; function, or using pkg-config for packages that do not
;; have this feature
"-DARROW_DEPENDENCY_SOURCE=SYSTEM"
"-Dxsimd_SOURCE=SYSTEM"
;; The CMake option ARROW_DEPENDENCY_SOURCE is a global
;; option that instructs the build system how to resolve
;; each dependency. SYSTEM = Finding the dependency in
;; system paths using CMake's built-in find_package
;; function, or using pkg-config for packages that do not
;; have this feature
"-DARROW_DEPENDENCY_SOURCE=SYSTEM"
"-Dxsimd_SOURCE=SYSTEM"
"-DARROW_RUNTIME_SIMD_LEVEL=NONE"
"-DARROW_SIMD_LEVEL=NONE"
"-DARROW_PACKAGE_KIND=Guix"
"-DARROW_RUNTIME_SIMD_LEVEL=NONE"
"-DARROW_SIMD_LEVEL=NONE"
"-DARROW_PACKAGE_KIND=Guix"
;; Split output into its component packages.
(string-append "-DCMAKE_INSTALL_PREFIX=" #$output:lib)
(string-append "-DCMAKE_INSTALL_RPATH=" #$output:lib "/lib")
(string-append "-DCMAKE_INSTALL_BINDIR=" #$output "/bin")
(string-append "-DCMAKE_INSTALL_INCLUDEDIR=" #$output:include
"/share/include")
;; Split output into its component packages.
(string-append "-DCMAKE_INSTALL_PREFIX=" #$output:lib)
(string-append "-DCMAKE_INSTALL_RPATH=" #$output:lib "/lib")
(string-append "-DCMAKE_INSTALL_BINDIR=" #$output "/bin")
(string-append "-DCMAKE_INSTALL_INCLUDEDIR=" #$output: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"
"-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"
"-DARROW_ORC=ON"
"-DORC_SOURCE=SYSTEM"
;; 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")))
;; Building the tests forces on all the
;; optional features and the use of static
;; libraries.
"-DARROW_BUILD_TESTS=OFF"
;;"-DBENCHMARK_ENABLE_TESTING=OFF"
"-DARROW_BUILD_STATIC=OFF")))
(inputs
(list boost
(list apache-orc
boost
brotli
bzip2
double-conversion