PyVO is a package providing access to remote data and services of the Virtual observatory (VO) using Python.
Its development was launched by the NSF/NASA-funded Virtual Astronomical Observatory (VAO, www.usvao.org) project (formerly under the name VAOpy) as part of its initiative to bring VO capabilities to desktop. Its goal is to allow astronomers and tool developers to access data and services from remote archives and other web resources. It takes advantage of VO standards to give access to thousands of catalogs, data archives, information services, and analysis tools. It also takes advantage of the general capabilities of Astopy (and numpy), and so a secondary goal is to provide a development platform for migrating more VO capabilities into Astropy.
Source code can be found on GitHub
Releases of PyVO are available from PyPI
thus, it and its prerequisites can be most easily installed using pip
:
pip install pyvo
Releases are also conda packaged and available on the conda-forge
channel.
PyVO requires Python 3.8 or later.
The following packages are required for PyVO:
The following packages are optional dependencies and are required for the full functionality:
- pillow
- defusedxml
For running the tests, and building the documentation, the following infrastructure packages are required:
- pytest-astropy
- requests-mock
- sphinx-astropy
To install from source use pip
:
pip install .[all]
We recommend and encourage testing against the development version of PyVO in CI, both for dependent libraries and notebook providers. As PyVO is a pure Python library, this can be done as easily as pip installing the developer version from GitHub:
pip install git+https://github.com/astropy/pyvo.git#egg=pyvo
An example for setting up development version testing for a library as a GitHub Actions Workflow can be found in astroquery.
Many instructive examples can be found in the PyVO Documentation. Additional examples can be found in the examples directory.
PyVO uses the Astropy framework for unit tests which is built into the setup script. To run the tests, type:
pip install .[test] pytest
This will run all unit tests that do not require a network connection. To run all tests, including those that access the network, add the --remote-data option:
pytest --remote-data