Skip to content
This repository has been archived by the owner on Oct 24, 2024. It is now read-only.

Warning about repo being archived #344

Merged
merged 19 commits into from
Oct 24, 2024

Conversation

TomNicholas
Copy link
Member

No description provided.

@TomNicholas TomNicholas added documentation Improvements or additions to documentation packaging labels Oct 23, 2024
@TomNicholas
Copy link
Member Author

God damnit I don't want to have to fix sphinx and conda build issues just to archive this repo anyway. @keewis do you think I can just merge and release this regardless of the CI not working?

@keewis
Copy link
Contributor

keewis commented Oct 23, 2024

we have to fix sphinx so we can publish the changed docs, and if we want to release we should also check that the package still works. So looks like we can't skip that, unfortunately?

@TomNicholas
Copy link
Member Author

ImportError: cannot import name 'HybridMappingProxy' from 'xarray.core.utils'

This is interesting - this change was done by Stephan recently as part of the datatree refactor. I think I will just pin xarray to <=2024.09.0 (the last released version before that change). If that causes a problem for anyone then it's just a sign they should migrate.

@keewis
Copy link
Contributor

keewis commented Oct 23, 2024

I think pinning should be fine, nobody should need to have two versions of datatree at the same time.

If it's fine with you I'll try to get the docs to at least compile (by pushing to this branch).

@keewis
Copy link
Contributor

keewis commented Oct 23, 2024

looks like the issue was the deprecation / discontinuation of sphinx-panels

Edit: it passed on RTD, not just locally

@keewis
Copy link
Contributor

keewis commented Oct 23, 2024

I think we can ignore the failing dev CI (xarray does not support 3.9 anymore, so the last release that did is pulled in, which makes it the only "dev" CI to succeed), so this should be good to go?

Copy link
Contributor

@keewis keewis left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

actually, one comment about the links to the migration guide, but otherwise this looks good to me

warnings.warn(
"The repository `xarray-contrib/datatree` has been archived! It is no longer maintained as of October 2024. "
"Datatree functionality has been migrated upstream to `pydata/xarray`, so please use `xarray.DataTree` instead. "
"To help you migrate there is a guide at https://github.com/pydata/xarray/blob/main/DATATREE_MIGRATION_GUIDE.md",
Copy link
Contributor

@keewis keewis Oct 23, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this might be tricky to change since we do plan on moving it out of the xarray repository at some point. Should we instead point to a dedicated discussion here that we can more easily modify?


There will be no further bugfixes or feature additions to this respository.

Users of this repository should migrate to using `xarray.DataTree` instead, following the `Migration Guide <https://github.com/pydata/xarray/blob/main/DATATREE_MIGRATION_GUIDE.md>`_.
Copy link
Contributor

@keewis keewis Oct 23, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

same here: should we link to a discussion on the datatree repository?

@TomNicholas
Copy link
Member Author

should we link to a discussion on the datatree repository?

Perhaps I should raise one single issue on this repo and pin it, as another form of warning. That becomes read-only once archived though. That can have a multitude of links, and we can point to that issue from the warning. How does that sound?

@TomNicholas
Copy link
Member Author

looks like the issue was the deprecation / discontinuation of sphinx-panels

also thank you helping with that! Really appreciate it.

@keewis
Copy link
Contributor

keewis commented Oct 23, 2024

That becomes read-only once archived though. That can have a multitude of links, and we can point to that issue from the warning. How does that sound?

That's fine, you can temporarily un-archive a repository to change the link – which is much easier than having to re-release again with a changed link.

But otherwise this sounds good to me!

@TomNicholas
Copy link
Member Author

Raised #345, so I will merge this PR and release as soon as we release xarray.

@keewis
Copy link
Contributor

keewis commented Oct 23, 2024

after updating the links?

@TomNicholas
Copy link
Member Author

By updating the links you mean link to the new release of xarray? I can do that.

@keewis
Copy link
Contributor

keewis commented Oct 23, 2024

nah, I meant change the link to the migration guide to issue. Also, might be good to ask people to open discussions on the xarray repository if they need help with the migration.

Copy link
Contributor

@keewis keewis left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I was thinking of something like this, but doesn't as helpful as directly linking to the guide. However, just to really stress this, I do think we can save ourselves a lot of trouble by keeping the link to the migration guide mutable.

README.md Outdated
PLEASE RAISE NEW ISSUES ON THE XARRAY REPOSITORY UPSTREAM
There will be no further bugfixes or feature additions to this respository.

Users of this repository should migrate to using `xarray.DataTree` instead, following the [Migration Guide](https://github.com/pydata/xarray/blob/main/DATATREE_MIGRATION_GUIDE.md).
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
Users of this repository should migrate to using `xarray.DataTree` instead, following the [Migration Guide](https://github.com/pydata/xarray/blob/main/DATATREE_MIGRATION_GUIDE.md).
for more information and a link to the migration guide see [this issue](https://github.com/xarray-contrib/datatree/issues/345)

@TomNicholas
Copy link
Member Author

Maybe actually instead I should make a new discussion on pydata/xarray, pin it, and use that both to link to the migration guide and collect feedback / tips for datatree users. That link will be just as permanent as an issue, so the warnings can just link to that. What do you think?

@keewis
Copy link
Contributor

keewis commented Oct 24, 2024

that would be better, actually, people are less likely to try to open an issue here if we directly link to the xarray repository.

@TomNicholas TomNicholas enabled auto-merge (squash) October 24, 2024 22:06
@TomNicholas TomNicholas merged commit 5f3956f into xarray-contrib:main Oct 24, 2024
10 of 13 checks passed
@TomNicholas TomNicholas deleted the archive_repo branch October 24, 2024 22:10
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
documentation Improvements or additions to documentation packaging
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants