Move pyarrow detection to get_arrow.cmake so that it also affects C++…

… builds

cpp/CMakeLists.txt

@@ -90,9 +90,6 @@ option(
 )
 mark_as_advanced(CUDF_BUILD_STREAMS_TEST_UTIL)
 
-option(USE_LIBARROW_FROM_PYARROW "Use the libarrow contained within pyarrow." OFF)
-mark_as_advanced(USE_LIBARROW_FROM_PYARROW)
-
 message(VERBOSE "CUDF: Build with NVTX support: ${USE_NVTX}")
 message(VERBOSE "CUDF: Configure CMake to build tests: ${BUILD_TESTS}")
 message(VERBOSE "CUDF: Configure CMake to build (google & nvbench) benchmarks: ${BUILD_BENCHMARKS}")

cpp/cmake/thirdparty/get_arrow.cmake

@@ -32,7 +32,7 @@ function(find_libarrow_in_python_wheel PYARROW_VERSION)
   # `${MINOR_VERSION}${PATCH_VERSION}` is almost always equivalent to "00"),
   # the soname is not generated by concatenating the major, minor, and patch versions into a single
   # version number soname, just `${MAJOR_VERSION}00`
-  set(PYARROW_LIB "libarrow.so.${PYARROW_SO_VER}00")
+  set(PYARROW_LIB "libarrow.${CMAKE_SHARED_LIBRARY_SUFFIX}.${PYARROW_SO_VER}00")
 
   execute_process(
     COMMAND "${Python_EXECUTABLE}" -c "import pyarrow; print(pyarrow.get_library_dirs()[0])"
@@ -90,10 +90,10 @@ endfunction()
 
 # This function finds arrow and sets any additional necessary environment variables.
 function(find_and_configure_arrow VERSION BUILD_STATIC ENABLE_S3 ENABLE_ORC ENABLE_PYTHON
-         ENABLE_PARQUET
+         ENABLE_PARQUET PYARROW_LIBARROW
 )
 
-  if(USE_LIBARROW_FROM_PYARROW)
+  if(PYARROW_LIBARROW)
     # Generate a FindArrow.cmake to find pyarrow's libarrow.so
     find_libarrow_in_python_wheel(${VERSION})
     set(ARROW_FOUND
@@ -431,7 +431,35 @@ if(NOT DEFINED CUDF_VERSION_Arrow)
   )
 endif()
 
+# If the pyarrow package contains libarrow we want to link against it directly instead of searching
+# for a separate libarrow. Even if libcudf was built against a different libarrow, that library
+# would have to be ABI-compatible with the one in pyarrow for the packages to work together, and for
+# wheels we must use the library in the pyarrow wheel, so it's best to simply be consistent.
+find_package(Python 3.9 COMPONENTS Interpreter)
+
+if (${Python_FOUND})
+  execute_process(
+    COMMAND "${Python_EXECUTABLE}" -c "import importlib; print(importlib.util.find_spec('pyarrow') is not None)"
+    OUTPUT_VARIABLE _pyarrow_installed
+    OUTPUT_STRIP_TRAILING_WHITESPACE
+  )
+  if (${_pyarrow_installed} STREQUAL "True")
+    execute_process(
+      COMMAND "${Python_EXECUTABLE}" -c "import pyarrow; print(pyarrow.get_library_dirs()[0])"
+      OUTPUT_VARIABLE _pyarrow_lib_dir
+      OUTPUT_STRIP_TRAILING_WHITESPACE
+    )
+    file(GLOB _pyarrow_libs "${_pyarrow_lib_dir}/libarrow.${CMAKE_SHARED_LIBRARY_SUFFIX}*")
+    list(LENGTH _pyarrow_libs _pyarrow_libs_len)
+    if(_pyarrow_libs_len GREATER 0)
+      set(USE_LIBARROW_FROM_PYARROW ON)
+    else()
+      set(USE_LIBARROW_FROM_PYARROW OFF)
+    endif()
+  endif()
+endif()
+
 find_and_configure_arrow(
   ${CUDF_VERSION_Arrow} ${CUDF_USE_ARROW_STATIC} ${CUDF_ENABLE_ARROW_S3} ${CUDF_ENABLE_ARROW_ORC}
-  ${CUDF_ENABLE_ARROW_PYTHON} ${CUDF_ENABLE_ARROW_PARQUET}
+  ${CUDF_ENABLE_ARROW_PYTHON} ${CUDF_ENABLE_ARROW_PARQUET} ${USE_LIBARROW_FROM_PYARROW}
 )

python/cudf/CMakeLists.txt

@@ -34,25 +34,9 @@ option(FIND_CUDF_CPP "Search for existing CUDF C++ installations before defaulti
        OFF
 )
 
-# If the pyarrow package contains libarrow we want to link against it directly instead of searching
-# for a separate libarrow. Even if libcudf was built against a different libarrow, that library
-# would have to be ABI-compatible with the one in pyarrow for the packages to work together, and for
-# wheels we must use the library in the pyarrow wheel.
+# Find Python early so that later commands can use it
 find_package(Python 3.9 REQUIRED COMPONENTS Interpreter)
 
-execute_process(
-  COMMAND "${Python_EXECUTABLE}" -c "import pyarrow; print(pyarrow.get_library_dirs()[0])"
-  OUTPUT_VARIABLE _pyarrow_lib_dir
-  OUTPUT_STRIP_TRAILING_WHITESPACE
-)
-file(GLOB _pyarrow_libs "${_pyarrow_lib_dir}/libarrow.${CMAKE_SHARED_LIBRARY_SUFFIX}*")
-list(LENGTH _pyarrow_libs _pyarrow_libs_len)
-if(_pyarrow_libs_len GREATER 0)
-  set(USE_LIBARROW_FROM_PYARROW ON)
-else()
-  set(USE_LIBARROW_FROM_PYARROW OFF)
-endif()
-
 # If the user requested it we attempt to find CUDF.
 if(FIND_CUDF_CPP)
   # We need to find arrow before libcudf since libcudf requires it but doesn't bundle arrow