Skip to content

pip_install.md

Latest commit

 

History

History
58 lines (38 loc) · 1.62 KB

pip_install.md

File metadata and controls

58 lines (38 loc) · 1.62 KB

Pip Install Usage

daylily-ephemeral-cluster can be installed with pip and used from any working directory. This is mainly useful for downstream repos, automation environments, or operator machines where you do not want to keep a full checkout around.

Install

From a clean virtual environment:

python -m venv .venv
source .venv/bin/activate

# From a local checkout
pip install /path/to/daylily-ephemeral-cluster

# Or from git
pip install "git+https://github.com/Daylily-Informatics/daylily-ephemeral-cluster.git@<ref>"

Verify The Install

daylily-ec --help
daylily-ec version
daylily-ec info
daylily-ec resources-dir

Packaged Resources

The wheel includes packaged repo assets such as config/, etc/, and selected bin/ helpers. They are extracted at runtime under:

~/.config/daylily/resources/<package-version>/

Use this command to resolve the active resource directory:

daylily-ec resources-dir

Override the resource root when needed:

export DAYLILY_EC_RESOURCES_DIR=/path/to/override-root

The override directory must contain the Daylily config/, etc/, and bin/ trees expected by the helper scripts.

Host Requirements

pip installs the Python dependencies, but some workflows still expect host tools or external configuration:

  • AWS CLI v2 for commands that shell out to aws
  • pcluster CLI for cluster management commands (e.g. daylily-ec cluster-info)
  • ssh for connecting to head nodes
  • a configured AWS profile when operating on real infrastructure

If you want the managed conda workflow instead, use DAY_EC_ENVIRONMENT.md.