Warn about EOL Python versions

[kesara]

What's the problem this feature will solve?

Some times users are not aware about Python release cycle.
It would be nice if pip can warn the users if they are running a EOL Python version.

Describe the solution you'd like

If a user is running a EOL Python version like Python 3.6.
Add a notice similar to [notice] A new release of pip is available saying that user is running an EOL Python version and should consider upgrading.

Alternative Solutions

N/A

Additional context

Sometimes users are unaware that they are are running an EOL Python version¹.

Code of Conduct

• I agree to follow the PSF Code of Conduct.

Footnotes

1. https://mailarchive.ietf.org/arch/msg/tools-discuss/8PcBl45UNbA_xk8SnYUJa3lVn7U/ ↩

[notatallshaw]

The problem with this is Python 3.6 isn't EoL everywhere. I believe, though double check, it's the default version of Python on RHEL 8, which will receive maintenance support till May 2029.

In general it is up to the specific Python distribution of when to mark something as End of Life, and this can wildly vary between different distributors, and they can change their mind mid cycle if they so choose.

IMO for pip to add a feature like this the Python distribution would need to provide this information to pip via a standard, such a feature would need to be discussed on the Python discussion board: https://discuss.python.org/c/packaging/14

[kesara]

But pip itself has already dropped the support for Python 3.6¹.
So I assume pip can follow the Python lifecycle schedule².

Footnotes

1. https://github.com/pypa/pip/pull/10641 ↩

2. https://devguide.python.org/versions/ ↩

[notatallshaw]

Pip doesn't have a concrete lifecycle of Python support, it's based on how many downloads pip receives. For example pip still supports Python 3.8 even though the PSF doesn't.

[sirosen]

I see two additional problems with this idea.

1. If someone is running 3.5, they're probably running an old version of pip. Therefore, adding a warning is not effective as outreach to that user.
2. It is out of scope for pip. pip doesn't (attempt to) manage your Python installation.