Installation

If the automatic (pip based) install doesn’t work below, you may need to follow some of the manual steps.

Python Environment

If you don’t yet have a working python installation/environment, then e.g. on the MPCDF machines (vera, freya, raven, virgo, and so on) you can:

  1. Set up a clean anaconda environment:

module load anaconda/3/2023.03
mkdir -p ~/.local/envs
conda create --prefix=~/.local/envs/myenv python=3.12
source activate ~/.local/envs/myenv

  1. Add the following lines to your ~/.bashrc file for permanence

module load gcc/12
module load openmpi/4.1
module load fftw-serial/3.3.10
module load hdf5-serial/1.12.2
module load gsl/2.4

module load anaconda/3/2023.03
source activate ~/.local/envs/myenv

Automatic Installation

  1. If you do not plan on editing this package and hope to simply use it as is (not a common use case), you can simply install it

pip install git+ssh://git@github.com/dnelson86/temet.git

  1. Or, if you plan on editing, making changes, and adding functionality to this package (this is expected to be the most common choice), then first clone the repository into your home directory, here into a temet directory, then install the package in ‘editable’ mode. This means that the files in this directory are not copied anywhere, but are used as is. Any changes you make are reflected (i.e. usable) in your python environment.

cd ~
git clone git@github.com:dnelson86/temet.git
pip install -e temet

Manual Installation

If the above didn’t work or gave errors, you may need to go through a manual installation:

  1. Clone the repository into your home directory, here into a temet directory

cd ~
git clone git@github.com:dnelson86/temet.git

  1. Clone the public Illustris data analysis python scripts

mkdir ~/illustris_release
cd ~/illustris_release
git clone git@github.com:illustristng/illustris_python.git

3. Make sure both are set in the $PYTHONPATH environment variable. For example, add the following lines to your ~/.bashrc file

export PYTHONPATH+=:$HOME/temet/:$HOME/illustris_release/

  1. Install all python dependencies as required

pip install --user -r ~/temet/requirements.txt

  1. Point matplotlib to the default settings file

mkdir -p ~/.config/matplotlib
ln -s ~/temet/temet/matplotlibrc ~/.config/matplotlib/

and install the Roboto font used by default

mkdir -p ~/.fonts/Roboto
cd ~/.fonts/Roboto/
wget https://github.com/google/fonts/raw/main/apache/roboto/static/Roboto-Light.ttf
wget https://github.com/google/fonts/raw/main/apache/roboto/static/Roboto-LightItalic.ttf

  1. Several large tabulated data files are used to compute e.g. stellar luminosities (from FSPS), ion abundances and emissivities (from CLOUDY), and X-ray emission (from APEC/XPSEC). For convenience these can be downloaded as

cd ~/temet/temet/tables/
wget -r -nH --cut-dirs=1 --no-parent --reject="index.html*" -e robots=off temet.tng-project.org/tables/

  1. Organize simulation directories as follows

mkdir ~/sims.TNG
mkdir ~/sims.TNG/L75n1820TNG
mkdir ~/sims.TNG/L75n1820TNG/data.files
cd ~/sims.TNG/L75n1820TNG/
ln -s /virgotng/universe/IllustrisTNG/L75n1820TNG/output .
ln -s /virgotng/universe/IllustrisTNG/L75n1820TNG/postprocessing .

note that the last two lines create symlinks to the actual output directory where the simulation data files (groupcat_* and snapdir_*) reside, as well as to the postprocessing directory (containing trees, etc). Replace as needed with the actual path on your machine.

Manual Installation (continued)

Several external tools and post-processing codes are used, for specific analysis routines. The following additional installation steps are therefore optional, depending on application.

  1. Although most stellar light/magnitude tables of relevance are pre-computed and have been downloaded in the previous steps, the FSPS stellar population synthesis package is required to generate new SPS tables. To install:

mkdir ~/code
cd ~/code/
git clone https://github.com/cconroy20/fsps

edit the src/sps_vars.f90 file and switch the defaults spectral and isochrone libraries to

MILES 1
PADOVA 1 (and so MIST 0)

edit src/Makefile and make sure the F90FLAGS line contains -fPIC, then compile FSPS

make

add the following line to your ~/.bashrc file

export SPS_HOME=$HOME/code/fsps/

  1. Although the X-ray emission tables have been pre-computed, the creation of new tables requires the AtomDB APEC files.

mkdir ~/code/atomdb/
cd ~/code/atomdb/
wget --content-disposition http://www.atomdb.org/download_process.php?fname=apec_v3_0_9
wget --content-disposition http://www.atomdb.org/download_process.php?fname=apec_v3_0_9_nei
tar -xvf *.bz2 --strip-components 1
rm *.bz2

  1. The SKIRT dust radiative transfer code can be used to compute dust-attenuated stellar light images and spectra, dust infrared emission, and many further sophisticated observables.

mkdir ~/code/SKIRT9/
cd ~/code/SKIRT9/
git clone https://github.com/SKIRT/SKIRT9.git git
cd git
chmod +rx *.sh
./makeSKIRT.sh
./downloadResources.sh

link the executable into your local bin directory

mkdir ~/.local/bin
cd ~/.local/bin
ln -s ~/code/SKIRT9/release/SKIRT/main/skirt .

add the following lines to your ~/.bashrc file for permanence

export PATH=$HOME/.local/bin:$PATH

Now you’re all set!

Updating

The instructions above install a local copy into your home directory, which you are free to edit and modify as required. At any time you can then update your copy to the newest version, pulling in any changes, bugfixes, and so on, with:

git pull

However if you have made changes in the meantime, you may see a message similar to “error: Your local changes to the following files would be overwritten by merge. Please commit your changes or stash them before you merge.” In this case, you want to keep your local work, but also make the update. Please read this quick git tutorial on the topic.