We discuss the development of SunPy and the ecosystem of related Python packages that provide specialized functionality to the heliophysics community. SunPy is a community-developed, free and open-source software package for solar physics. It aims to provide a comprehensive data-analysis environment that allows researchers within the field of solar physics to perform common tasks simply. SunPy has released 9 beta versions (0.x) and work is now ongoing on SunPy 1.0 whose purpose will be to provide a stable API. SunPy 1.0 will provide the ability to search and download data from multiple locations including the Virtual Solar Observatory (VSO), Joint Science Operations Center (JSOC) and the Heliophysics Event Knowledgebase (HEK); offers TimeSeries and Map objects that enable the manipulation of time series and two-dimensional image data respectively; and enables transformations between solar and astronomical coordinate systems. SunPy 1.0 will support Python 3.6+ only, and the SunPy 0.9.x series will provide bugfix support to existing functionality for Python 2.7.x until the scheduled end of Python 2 support at the end of 2019. SunPy is built using functionality from Astropy, pandas, matplotlib and other scientific packages commonly used in astronomy. SunPy is supported by the development of affiliated packages that provide additional functionality built on top of the core library. For example, IRISPy is an SunPy-affiliated package designed for reading, manipulating and visualizing data taken with the Interface Region Imaging Spectrometer (IRIS). The package ndcube is a SunPy-affiliated package designed for handling n-dimensional datacubes described by a World Coordinate System translation, permitting powerful manipulation of multi-dimensional data. The package drms provides access to Michelson Doppler Imager (MDI), and Solar Dynamics Observatory (SDO) data. Other packages in the SunPy ecosystem are fiasco, a Python interface to the Chianti atomic database and solarbextrapolation, a library for extrapolating 3D magnetic fields from line-of-sight magnetograms. We discuss how SunPy can support an expanding heliophysical data-analysis environment.
Link to PDF (may not be available yet): P7-3.pdf