In 4D-STAR, we strive to make our software widely accessible. To achieve this goal, simplifying the installation process is key. Inspired by the MESA installation approach, we aim to streamline the setup of Ester by creating a Software Development Kit (SDK) that packages all the necessary libraries for compilation into a single bundle. Given the interest in integrating MESA functionality within Ester, it makes sense to compile both with the same SDK, hence the idea of extending the already existing MESA SDK.
Currently, the 4D-STAR SDK extends the MESA SDK by including cmake, CBLAS, SWIG, and the C++ API of HDF-5. This repo was made to test the 4DSTAR SDK.
The following assummes that git, curl, make and tar exist in your system. For example in Ubuntu you can install them with:
apt install git curl make tar
You will also need Python3 with the NumPy and Matplotlib libraries, as these are not provided by the SDK. You can install Python either using your package manager or conda/mamba. For example in Ubuntu
apt install python3 python3-dev python3-numpy python3-matplotlib
or with conda
conda create -n ester_env python=3.12 numpy matplotlib -c conda-forge
conda activate ester_env
Clone or download this repository, open a bash shell and then run the script
./installEsterWithSDK.sh
which will
- Download and initialize the SDK
- Clone Ester, check the
evolutionbranch and patch it to make it fully compatible with the SDK - Compile and install Ester
- Execute
star1dandstar_evol
If no errors are encountered during this process you should see a install/bin folder with the star1d, star2d and star_evol binaries ready to be used.
- This setup is currently supported on Linux and it has been tested on Ubuntu, Arch Linux, Rocky Linux, and also in Windows through the WSL
- If you find any errors please share them with us, along with the details of your system