-
Notifications
You must be signed in to change notification settings - Fork 3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Drop support for Python 2.6? #3955
Comments
Oh I forgot to mention, the response in that thread was entirely positive about the idea of deprecating and dropping support for Python 2.6 over the span of ~4 days. |
Oh and also, as far as benefits, besides being able to use more modern mechanisms that don't exist in Python 2.6 (but do exist in 2.7 and 3.3+) this will also free our dependencies from needing to continue to support 2.6. They may still elect to do this, but now they don't need to do it for our sake. If we go forward with it, I'll also immediately adjust |
To clarify, declare 9.0 as last major version that will support 2.6? |
Lovely, can we have a timeline? |
@Ivoz Yes. @RonnyPfannschmidt We average 6-8 months between releases, so likely ~8mo-1y for it to be actually gone. |
If I wanted to do this for bdist_wheel, could I rely on the requires-python tagging to get it done before pip? |
Well, requires-python doesn't exist in a released version of pip yet. |
@dstufft nice, so we can drop 2.6 for pytest 4.0 maybe before |
Is the issue still planned? |
@gukoff Not quite sure what you mean. The current pip 9 series deprecates Python 2.6 support, and pip 10 won't support 2.6. There's no confirmed release date for pip 10, but as noted above, 6-8 months after pip 9 was released is a good estimate. Is that what you wanted to know? |
The closing of the issue confuses me; I'd like to see the intention to drop Python2.6 support documented somewhere, but the only referenced PR which is merged has nothing to do with it. If the latest releases explicitly deprecate it in code, it is already OK, thanks. |
In a recent thread on distutils-sig I asked the larger packaging community when they thought it would be reasonable to kill Python 2.6 support.
I go into more detail on distutils-sig, but I feel like Python 2.6 is a bit of an albatross around our neck and it only accounts for ~3% of the traffic from PyPI when you look at "modern" versions of pip (i.e. 6+) and it itself has been end-of-life for 2013-10-29 or almost 3 years now.
I would like to drop support for Python 2.6 in pip, as of pip 8 we've had a pending deprecation/yellow warning that was intended to just stay there for as long as we felt appropriate and didn't nail us to any hard timeline except for "a future version of pip". In light of that, I'd say let's just bump that up to a deprecation/red warning in Pip 9 and remove it in pip 10.
Thoughts? Unless someone comes out against this I plan on moving forward with it.
The text was updated successfully, but these errors were encountered: