FlowvVIPER

Workflow examples for radio astronomy data processing with astroviper.

Requires Python >= 3.11, < 3.14.

Installation

Base install

pip install flowviper

This is sufficient for the bundled Prefect demo (downloads a small real MSv4 processing set via toolviper) and workflows that start from MSv4 or zarr data already converted. It does not include reading legacy CASA MeasurementSets (.ms files).

macOS — MSv2 support

On macOS, python-casacore is not available via pip. Install a CASA table backend before using workflows that read or convert MSv2 data.

Path A (pip-only, recommended)

pip install --extra-index-url https://casa-pip.nrao.edu/repository/pypi-group/simple \
  casaconfig casatools
pip install flowviper

Works with a normal Python virtual environment. Do not install python-casacore alongside casatools (namespace conflict).

Path B (conda)

For users already using Miniforge, mamba, or conda:

conda create -n flowviper python=3.12 pip
conda activate flowviper
conda install -c conda-forge python-casacore
pip install flowviper

casatools is published on the NRAO pip index, not on PyPI, so it must be installed manually as shown above.

Prefect workflows

Example workflows live in prefect_workflow/. See prefect_workflow/README.md for usage notes.

Quick start:

pip install flowviper
python prefect_workflow/cube_imaging_example.py