Installation

There are two ways of installing pyPDAF.

Conda

The easiest approach is using conda. Currently, pyPDAF is available from conda for Windows, Linux and MacOS. The installation can be obtained via:

conda create -n pypdaf -c conda-forge yumengch::pypdaf

After installation, pyPDAF can be used by activating the conda environment conda activate pypdaf.

Source code

In some cases, it is desirable to compile pyPDAF so that one can use their favourite compiler as well as MPI and BLAS implementation.

In this case, pyPDAF source code can be obtained from source

git clone --recurse-submodules https://github.com/yumengch/pyPDAF.git
cd pyPDAF

The package can be installed with:

python -m pip install . -v \
    --config-settings=setup-args="-Dblas_lib=[LIBS]" \
    --config-settings=setup-args="-Dincdirs=[INCDIRS]" \
    --config-settings=setup-args="-Dlibdirs=[LIBDIRS]" \
    --config-settings=setup-args="-Dmpi_mod=MPIF90"^
    --config-settings=setup-args="-Dbuildtype=release"

Here, LIBS, INCDIRS, and LIBDIRS are elements of a list, separated by , similar to a Python list.

• LIBS are all the required library names for BLAS libraries.

• LIBDIRS are directories of these libraries

• INCDIRS are include directories of libraries

• MPIF90 is the path to mpi.f90. This is useful for the case where mpi.mod is not directly provided but requires compiling by the user. This is optional.

One can adjust the compiler, compiler and linker flags by changing environment variables such as CC, FC, CFLAGS and FCFLAGS. See flags and compiler table for references. One could also directly modify meson.build which might require more knowledge of meson.

An example of installing pyPDAF in Linux or Mac:

CC=mpicc FC=mpifort python -m pip install . -v -Cbuild-dir=build \
    --config-settings=setup-args="-Dblas_lib=['openblas']" \
    --config-settings=setup-args="-Dincdirs=['/usr/lib/']" \
    --config-settings=setup-args="-Dlibdirs=['/usr/include']" \
    --config-settings=setup-args="-Dbuildtype=release"

In Windows, one can use

set CXX=clang-cl
set CC=clang-cl
set FC=flang-new
set MSMPI_INC=C:\Program Files (x86)\Microsoft SDKs\MPI\Include
set MSMPI_LIB64=C:\Program Files (x86)\Microsoft SDKs\MPI\Lib\x64

python -m pip install . -v ^
    -Cbuild-dir=build --config-settings=setup-args="-Dblas_lib=openblas"^
    --config-settings=setup-args="-Dincdirs="C:\Program Files (x86)\blas\include"^
    --config-settings=setup-args="-Dlibdirs="C:\Program Files (x86)\blas\lib"^
    --config-settings=setup-args="-Dmpi_mod=C:\Program Files (x86)\Microsoft SDKs\MPI\Include\mpi.f90"^
    --config-settings=setup-args="-Dbuildtype=release"

where the variable MSMPI_INC and MSMPI_LIB64 are required environment variable for using MSMPI.

Please raise an issue if you have any questions or problems with this.