PyHyp 2016

Introduction

pyHyp uses hyperbolic volume mesh marching schemes to extrude structured surface meshes into volume meshes. pyHyp is used as a preprocessing step in the geometry and mesh-creation process prior to an optimization. ¹

Installed version(s)

The following versions are installed and currently available on environment hpc-env/8.3:

• pyHyp/2.6.0-Singularity

Loading pyHyp

To load the desired version of the module, use the module load command, e.g.

module load hpc-env/8.3
module load pyHyp 

Always remember: these commands are case-sensitive!

Using pyHyp

pyHyp is built as a singularity image. This means that the softwares executables are accessible only by calling the image with singaularity. The process is similar to Docker containers, but the command options differ.
To make using pyHyp easier, we created a binary folder with "pseudo binaries" which are in fact just small wrapper scripts to call the singularity image in specific ways. This is why loading the mopdule automatically prints out the following quick-start tutorial:

$ ml pyHyp

This pyHyp package (version 2.6.0) works from inside a container file, called by Singularity (similar to Docker).
Please make sure that you don't have any Python module, or (Ana-)Conda environment loaded. The packages will only work inside of the containarized Python version!

Use this container by calling one of the following commands: 
 
    python3.8 <arguments>  
        => Calls python from inside the container like this: "singularity exec <container_image> python3 <your_arguments>"
        => Without any arguments, an interactive python shell session will start from inside the container.
 
    pyhyp-exec <executable> <arguments> <input_file> 
        => Executes any command from inside the container image. 
        => e.g.: "pyhyp-exec cgns_utils --help" ## print help output of cgns_utils and exit container afterwards.
 
    pyhyp-shell 
        => Opens a bash shell within the container. Here, you can browse through the container directories and start any of the installed packages.
        => The complete mdolab bundle is installed at /home/mdolabuser. 

Documentation

The full documentation can be found here. More information can be found at the projects GitHub page.