gnu: Add python-pyarrow-0.16.

* gnu/packages/databases.scm (python-pyarrow-0.16): New variable.

1 files changed, 58 insertions, 0 deletions

diff --git a/gnu/packages/databases.scm b/gnu/packages/databases.scm
index 03ea55c6e1..92ab94f453 100644
--- a/gnu/packages/databases.scm
+++ b/gnu/packages/databases.scm
@@ -4442,6 +4442,64 @@ implementation, along with tools for interoperability with pandas, NumPy, and
 other traditional Python scientific computing packages.")
     (license license:asl2.0)))
 
+(define-public python-pyarrow-0.16
+  (package
+    (inherit apache-arrow-0.16)
+    (name "python-pyarrow")
+    (build-system python-build-system)
+    (arguments
+     '(#:tests? #f          ; XXX There are no tests in the "python" directory
+       #:phases
+       (modify-phases %standard-phases
+         (delete 'build) ; XXX the build is performed again during the install phase
+         (add-after 'unpack 'enter-source-directory
+           (lambda _ (chdir "python")))
+         (add-after 'unpack 'make-git-checkout-writable
+           (lambda _
+             (for-each make-file-writable (find-files "."))))
+         (add-before 'install 'patch-cmake-variables
+           (lambda* (#:key inputs #:allow-other-keys)
+             ;; Replace cmake locations with hardcoded guix links for the
+             ;; underlying C++ library and headers.  This is a pretty awful
+             ;; hack.
+             (substitute* "cmake_modules/FindParquet.cmake"
+               (("# Licensed to the Apache Software Foundation" m)
+                (string-append "set(PARQUET_INCLUDE_DIR \""
+                               (assoc-ref inputs "apache-arrow:include")
+                               "/share/include\")\n" m))
+               (("find_package_handle_standard_args" m)
+                (string-append "set(PARQUET_LIB_DIR \""
+                               (assoc-ref inputs "apache-arrow:lib")
+                               "/lib\")\n" m)))))
+         (add-before 'install 'patch-parquet-library
+           (lambda _
+             (substitute* "CMakeLists.txt"
+               (("parquet_shared") "parquet"))))
+         (add-before 'install 'set-PYARROW_WITH_PARQUET
+           (lambda _
+             (setenv "PYARROW_WITH_PARQUET" "1"))))))
+    (propagated-inputs
+     `(("apache-arrow:lib" ,apache-arrow-0.16)
+       ("apache-arrow:include" ,apache-arrow-0.16 "include")
+       ("python-numpy" ,python-numpy)
+       ("python-pandas" ,python-pandas)
+       ("python-six" ,python-six)))
+    (native-inputs
+     (list cmake-minimal
+           pkg-config
+           python-cython
+           python-pytest
+           python-pytest-runner
+           python-setuptools-scm))
+    (outputs '("out"))
+    (home-page "https://arrow.apache.org/docs/python/")
+    (synopsis "Python bindings for Apache Arrow")
+    (description
+     "This library provides a Pythonic API wrapper for the reference Arrow C++
+implementation, along with tools for interoperability with pandas, NumPy, and
+other traditional Python scientific computing packages.")
+    (license license:asl2.0)))
+
 (define-public python-crate
   (package
     (name "python-crate")