ROOT CERN Conda Environment creation script

README.md

Setup ROOT Environment Script

This script automates the creation of a Conda environment for working with ROOT and Python, including optional configuration for additional packages, Conda channels, and Python versions. It also sets up environment variables for seamless integration with tools like VSCode and Jupyter Notebook.

Features

• Automates Environment Creation: Creates a Conda environment with ROOT and Python.
• Customizable Options: Allows specifying additional packages, Conda channels, and Python versions.
• Jupyter Kernel Integration: Automatically registers the environment as a Jupyter kernel.
• Environment Variable Management: Dynamically sets up and cleans up ROOT-specific environment variables.

---

Usage

1. Basic Usage

Run the script to create a default ROOT environment (root-env) with Python 3.9:

./setup_root_env.sh

2. Customize the Environment

You can specify optional arguments to customize the environment:

• -e or --env: Specify a custom environment name (default: root-env).
• -p or --packages: Add additional packages to install (comma-separated).
• -c or --channels: Specify Conda channels to use (comma-separated, default: conda-forge).
• -v or --python-version: Specify the Python version (default: 3.9).

Example:

./setup_root_env.sh --env my-env --python-version 3.8 --packages numpy,matplotlib --channels defaults,conda-forge

---

Options Summary

Option	Description	Default
-h, --help	Display help message and exit	-
-e, --env	Conda environment name	root-env
-p, --packages	Additional packages to install (comma-separated)	-
-c, --channels	Conda channels to use (comma-separated)	conda-forge
-v, --python-version	Python version	3.9

---

After Running

Once the script completes:

1. Activate the environment:

conda activate root-env

(Replace root-env with your custom environment name if specified.)

2. Use Jupyter Notebook or VSCode:

• The environment is registered as a Jupyter kernel named Python (root-env).

3. Test ROOT:

python -c "import ROOT; print(ROOT.__version__)"

setup_root_env.sh

#!/bin/bash

# Define color codes
GREEN='\033[1;32m'   # Info
YELLOW='\033[1;33m'  # Warning
RED='\033[1;31m'     # Error
NC='\033[0m'         # No Color

# Helper functions to print messages
info() { echo -e "${GREEN}[INFO]${NC} $1"; }
warn() { echo -e "${YELLOW}[WARNING]${NC} $1"; }
error() { echo -e "${RED}[ERROR]${NC} $1"; exit 1; }

# Function to display help
print_help() {
    echo -e "Usage: ./setup_root_env.sh [OPTIONS]"
    echo -e ""
    echo -e "Options:"
    echo -e "  -h, --help                   Show this help message and exit"
    echo -e "  -e, --env ENV_NAME           Specify the name of the Conda environment to create (default: root-env)"
    echo -e "  -p, --packages PACKAGES      Additional packages to install (comma-separated)"
    echo -e "  -c, --channels CHANNELS      Conda channels to use (comma-separated, default: conda-forge)"
    echo -e "  -v, --python-version VERSION Specify the Python version to use (default: 3.9)"
}

# Default values
ENV_NAME="root-env"
PACKAGES=""
CHANNELS="conda-forge"
PYTHON_VERSION="3.9"

# Parse arguments
while [[ $# -gt 0 ]]; do
    case $1 in
        -h|--help)
            print_help
            exit 0
            ;;
        -e|--env)
            ENV_NAME="$2"
            shift 2
            ;;
        -p|--packages)
            PACKAGES="$2"
            shift 2
            ;;
        -c|--channels)
            CHANNELS="$2"
            shift 2
            ;;
        -v|--python-version)
            PYTHON_VERSION="$2"
            shift 2
            ;;
        *)
            error "Unknown option: $1. Use -h or --help for usage."
            ;;
    esac
done

# Check if Conda is installed
if ! command -v conda &> /dev/null; then
    error "Conda is not installed. Please install Conda or Miniconda and try again."
fi

# Convert comma-separated lists to space-separated lists
IFS=',' read -r -a CHANNEL_LIST <<< "$CHANNELS"
CHANNEL_ARGS=""
for CHANNEL in "${CHANNEL_LIST[@]}"; do
    CHANNEL_ARGS+="-c $CHANNEL "
done

IFS=',' read -r -a PACKAGE_LIST <<< "$PACKAGES"

# Step 1: Remove existing environment if requested
if conda info --envs | grep -qE "^\s*$ENV_NAME\s"; then
    warn "Environment '$ENV_NAME' already exists."
    read -p "Do you want to remove it and create a fresh environment? (y/N): " choice
    if [[ "$choice" =~ ^[Yy]$ ]]; then
        info "Removing existing environment '$ENV_NAME'..."
        conda remove -n "$ENV_NAME" --all -y || error "Failed to remove environment '$ENV_NAME'."
        info "Environment '$ENV_NAME' removed successfully."
    else
        info "Exiting without making changes."
        exit 0
    fi
fi

# Step 2: Create a new environment
info "Creating new Conda environment '$ENV_NAME' with ROOT and Python $PYTHON_VERSION..."
CREATE_CMD="conda create -n \"$ENV_NAME\" python=\"$PYTHON_VERSION\" root ${PACKAGE_LIST[@]} ${CHANNEL_ARGS} -y"
eval $CREATE_CMD || error "Failed to create environment '$ENV_NAME'."

# Step 3: Activate the environment
info "Activating the environment..."

# Ensure Conda is initialized
CONDA_BASE=$(conda info --base)  # Get the base directory of Conda
source "$CONDA_BASE/etc/profile.d/conda.sh" || error "Failed to initialize Conda."

# Activate the environment
conda activate "$ENV_NAME" || error "Failed to activate environment '$ENV_NAME'."


# Step 4: Install Jupyter kernel
info "Installing Jupyter kernel..."
conda install ipykernel -y || error "Failed to install Jupyter kernel."
python -m ipykernel install --user --name="$ENV_NAME" --display-name "Python ($ENV_NAME)" || error "Failed to register Jupyter kernel."

# Step 5: Create activation and deactivation scripts
info "Setting up environment variables for ROOT..."

# Create activation script
ACTIVATE_SCRIPT="$CONDA_PREFIX/etc/conda/activate.d/env_vars.sh"
mkdir -p "$(dirname "$ACTIVATE_SCRIPT")"
cat > "$ACTIVATE_SCRIPT" <<EOF
export ROOTSYS=\$CONDA_PREFIX
export LD_LIBRARY_PATH=\$CONDA_PREFIX/lib:\$LD_LIBRARY_PATH
export PYTHONPATH=\$CONDA_PREFIX/lib:\$PYTHONPATH
EOF

# Create deactivation script
DEACTIVATE_SCRIPT="$CONDA_PREFIX/etc/conda/deactivate.d/env_vars.sh"
mkdir -p "$(dirname "$DEACTIVATE_SCRIPT")"
cat > "$DEACTIVATE_SCRIPT" <<EOF
unset ROOTSYS
unset LD_LIBRARY_PATH
unset PYTHONPATH
EOF

info "Environment variable scripts created successfully."

# Step 6: Test PyROOT
info "Testing PyROOT setup..."
python -c "import ROOT; print(ROOT.__version__)" &> /dev/null
if [[ $? -eq 0 ]]; then
    info "PyROOT is working correctly."
else
    error "PyROOT is not working. Please check the installation or environment setup."
fi

info "Environment '$ENV_NAME' has been successfully set up!"
info "You can now use it with VSCode or Jupyter Notebook. Select the 'Python ($ENV_NAME)' kernel."