Installation
Installing DP3
The batch pre-processing pipeline needs to call DP3 (LOFAR’s own preprocessing pipeline) as a subprocess. All that matters is that the DP3 executable is available in the PATH.
Detailed DP3 installation instructions will be provided in the future. At the moment, we recommend building it with Spack, using the DP3 build recipe provided as part of the ska-sdp-spack repository.
Installing poetry
poetry
is the official Python package manager for SKAO repositories. The recommended method
for installing poetry
is via pipx
, because pipx
allows running poetry in its own separate
virtual environment while making it available system-wide.
Instructions for installing pipx can be found here. Once that is done, just run:
pipx install poetry
Installing the batch pre-processing pipeline
Clone the repository
Navigate to the parent directory of your choice, clone preferably with SSH if you plan on contributing:
git clone git@gitlab.com:ska-telescope/sdp/science-pipeline-workflows/ska-sdp-batch-preprocess.git
Create a Python environment
It is highly recommended to install the pipeline inside a dedicated environment;
here we use venv
, which part of the standard library – again, alternatives are available.
We will store the pipeline’s virtual environment in a .venv/
sub-directory of the repository,
which is what poetry
expects by default.
cd ska-sdp-batch-preprocess/
python -m venv .venv/
We can now install the pipeline and its dependencies; they will all be placed inside the .venv/
directory.
poetry install
Activate the environment and verify the installation
The environment we just created needs to be activated before we can use the pipeline:
source .venv/bin/activate
It should now be possible to run:
ska-sdp-batch-preprocess --version