Skip to content
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 3.5 after 0.13? #3293

Closed
max-sixty opened this issue Sep 8, 2019 · 8 comments · Fixed by #3358
Closed

Drop 3.5 after 0.13? #3293

max-sixty opened this issue Sep 8, 2019 · 8 comments · Fixed by #3358

Comments

@max-sixty
Copy link
Collaborator

I just saw https://numpy.org/neps/nep-0029-deprecation_policy.html, which suggests dropping 3.5 support earlier this year (though NEP is dated to this July)

Ofc, existing xarray versions would still work with 3.5, only upgrades would require the new version (which I know everyone knows, but reinforces we don't need to wait for everyone to move off 3.5)

@crusaderky
Copy link
Contributor

As somebody who's maintained conda in a large corporate environment, I can confirm that upgrading major version of python is a couple day's work at worst.
I'm strongly in favour of a rolling policy for maintaining the last 2 versions of python supported by the official anaconda release package (as opposed to just the repo).

@shoyer
Copy link
Member

shoyer commented Sep 10, 2019

To clarify — the proposal is to drop support for Python 3.5 in the upcoming 0.13 release? Or later, in 0.14? I’d be happy to go ahead and do it now... it would be nice to be able to rely on dict iteration order and f-strings.

@crusaderky
Copy link
Contributor

Getting rid of all the ordereddicts is a colossal change, and IMHO this release is already way too large...

@crusaderky
Copy link
Contributor

(that said I'm perfectly happy to have 0.14 one week after 0.13!)

@max-sixty
Copy link
Collaborator Author

I wasn't initially suggesting for 0.13 but not opposed to it

What about releasing 0.13 and then dropping 3.5? That way any 3.5 users have an up-to-date version and we can have a master on 3.6 very soon

@max-sixty
Copy link
Collaborator Author

(that said I'm perfectly happy to have 0.14 one week after 0.13!)

What are the semver rules for underlying versions? If we drop 3.5 support, do we need a minor version increase to 0.14, or is a patch release to 0.13.1 OK?

@shoyer
Copy link
Member

shoyer commented Sep 10, 2019

Getting rid of all the ordereddicts is a colossal change, and IMHO this release is already way too large...

I agree, I was suggesting simply dropping support for Python 3.5 without other associated changes -- those can come later.

@shoyer
Copy link
Member

shoyer commented Sep 10, 2019

(that said I'm perfectly happy to have 0.14 one week after 0.13!)

What are the semver rules for underlying versions? If we drop 3.5 support, do we need a minor version increase to 0.14, or is a patch release to 0.13.1 OK?

We don't really have strict rules about this, but I think it's friendlier to keep major version bumps of xarray's dependencies to major xarray releases (i.e., 0.13 or 0.14).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

3 participants