[Tutorial/Workflow/Doc] Recommended Software Installations

[TinasheMTapera]

🧾 Summary

We want to have one page that's dedicated to listing recommended software
installations for our lab members. This will be a living document that we can
update as needed. The goal is to provide a clear and concise list of tools that are
commonly used in our lab, along with instructions for installation and any relevant
notes.

🎯 Learning Goals / Outcomes

What should a reader understand or be able to do after following this tutorial/workflow?

• Why it is important to have a standardized, minimal set of tools for our lab
• Why HPC clusters limit software installations and how we work around that
• Why we don't just use conda environments for everything
• Why we don't wait on IT to install software for us
• Why we use a local ~/software directory for our installations
• How to install software in a local directory
• How to manage software installations and updates over time
• When to use FASRC's modules system vs. when to install software locally

📦 Required Tools, Packages, or Data

List any software packages, datasets, or computing platforms used (e.g., R, Python, FASRC, GeoPandas, etc.).

• Tools:
  • github CLI
  • quarto
  • arf
  • uv
  • rv
  • conda
• Libraries:
  • geospatial software (e.g., QGIS, ArcGIS, GDAL)
• Data source(s):

📄 Outline / Structure

Provide a brief outline of the sections or steps in the document:

1. Introduction
2. Binaries > Modules > Environments > Containers
3. Each section, i.e. user-specific global software -> project-specific
   environments -> Python, R
4. Conclusion / Next Steps

🔗 Related Materials

Link to related GitHub repos, issues, lab protocols, or previous examples:

• example-link-1
• example-link-2

🚦Status & Next Steps

What is the current status?

• Proposal
• In Progress
• Ready for Review
• Completed

Any notes on who will write or review the content?

[TinasheMTapera]

For gh cli, use webi:

(base) [ttapera@holylogin07 software]$ curl -sS https://webi.sh/gh | sh   


>>> Welcome to Webi! - modern tools, instant installs.  <<<
    We expect your experience to be absolutely perfect!

    Success? Star it!   https://github.com/webinstall/webi-installers
    Problem? Report it: https://github.com/webinstall/webi-installers/issues
                        (your system is GNU/Linux/x86_64 with libc & curl+wget)

Bootstrapping Webi
    Downloading https://webi.sh/packages/webi/webi.sh
        to ~/.local/bin/webi
    Running ~/.local/bin/webi gh@stable

Installing gh ...
    Found  ~/.local/bin
    Initializing ~/.config/envman/
    Edit ~/.profile to source ~/.config/envman/load.sh
    Edit ~/.bashrc to source ~/.config/envman/load.sh
    Downloading gh from
      https://github.com/cli/cli/releases/download/v2.90.0/gh_2.90.0_linux_amd64.tar.gz
    Saved as ~/Downloads/webi/gh/2.90.0/gh_2.90.0_linux_amd64.tar.gz
    Extracting ~/Downloads/webi/gh/2.90.0/gh_2.90.0_linux_amd64.tar.gz
    Installing to ~/.local/opt/gh-v2.90.0/bin/gh

    Installed 'gh v2.90.0' as ~/.local/bin/gh

    Edit ~/.config/envman/PATH.env to add:
        ~/.local/bin

>>> ACTION REQUIRED <<<
        Copy, paste & run the following command:
        source ~/.config/envman/PATH.env
        (newly opened terminal windows will update automatically)

(base) [ttapera@holylogin07 software]$ source ~/.config/envman/PATH.env
(base) [ttapera@holylogin07 software]$ which gh
~/.local/bin/gh

[TinasheMTapera]

HPC modules:
only when required by the platform
- maybe gcc
- maybe cuda
- maybe mpi

mamba:
runtime layer
- python
- r-base
- gdal
- geos
- proj
- netcdf
- other native libs

uv:
python packages

rv:
r packages

[TinasheMTapera]

[ HPC / MODULE LAYER ]

• scheduler (slurm)
• mpi / cuda (if needed)
• maybe compiler

[ USER SYSTEM LAYER ]

• mamba ← primary
• brew ← optional / edge

[ LANGUAGE / PROJECT LAYER ]

• rv (R)
• uv (Python)
• cargo (Rust)