[AstroPy] Astropy Proposal for Enhancement: A roadmap for Python 3-only support

Thomas Robitaille thomas.robitaille at gmail.com
Tue Aug 9 12:23:51 EDT 2016


Hi everyone,

[note that this was previously sent to astropy-dev only, but we are
now sending it out more widely to make sure users have a chance to
comment]

Python 2.7 will cease to be supported on January 1st 2020. As a
result, a number of scientific Python projects, including IPython,
Matplotlib, Sympy, and others have now signed the Python 3 statement,
which states that scientific Python packages will drop support for
Python 2 no later than 2020. For instance, the IPython developers have
now released IPython 5.0, which will be the last major release to
support Python 2.

I have written an Astropy Proposal for Enhancement (APE) that outlines
a suggested roadmap for future Astropy releases:

  https://github.com/astrofrog/astropy-APEs/blob/ape-python-3/APE10.rst

[for the background, APEs are a mechanism within the Astropy project
inspired by the Python Enhancement Proposals (PEPs), with the aim of
describing possible big changes that users and developers can then
comment on]

In this suggested roadmap, we would release a long term support (LTS)
version compatible with Python 2 and 3 next spring, which would be
supported until the end of 2019, while the first major release not
compatible with Python 2 would be in December 2017.

It is important to note that ending support for Python 2 by the end of
2019 does not mean that Python2 users will no longer be able to
install or use Astropy from 2020 onwards. Existing Astropy releases
will remain available indefinitely. Instead, the proposed roadmap
means that only new features developed after mid-2017 will be
available only to Python 3 users, and that we will not fix bugs for
Python 2-compatible releases after 2019.

We welcome feedback on this APE for the next couple of weeks, and by
August 19th at the latest. You can leave feedback on this pull
request:

 https://github.com/astropy/astropy-APEs/pull/13

After this date, depending on the level of consensus, the Astropy
coordination committee will either accept the APE (if there are no or
almost no objections), or settle on a decision process or a way
forward (if the APE is controversial).

Thanks!
Tom



More information about the AstroPy mailing list