Installing cling (from source, on macOS) and connect it to Jupyter, without conda or xeus_cling
cd to local dev directory (could be anything)
cd Documents/dev
clone the repo
git clone https://github.com/root-project/cling.git
follow the instructions provided by the README
cd tools/packaging
start the building process, this will be quite long and space demanding (~23 GB)
./cpt.py --check-requirements && ./cpt.py --create-dev-env Debug --with-workdir=build
install it (i.e. put it on your path, or just symlink it)
ln -s ~/Documents/dev/cling/tools/packaging/build/builddir/bin/cling /usr/local/bin
go back to the jupyter/kernel folder and finalize the process.
cd ../Jupyter/kernel
and finally
pip install .
jupyter-kernelspec install [--user] cling-cpp17
jupyter-kernelspec install [--user] cling-cpp1z
jupyter-kernelspec install [--user] cling-cpp14
jupyter-kernelspec install [--user] cling-cpp11
now if your launch jupyter
jupyter-notebook
you should have the option to start a c++ kernel.
you're welcome.
Wed Nov 3 22:08:14 CET 2021