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

staging: patch r-sleuth NAMESPACE #11

Closed
wants to merge 1 commit into from

Conversation

mschilli87
Copy link
Contributor

This is a workaround for pachterlab/sleuth#259 (effectively applying pachterlab/sleuth#260 after unpacking the package sources).
Without it, r-sleuth cannot be installed with Guix.

(substitute* "NAMESPACE"
(("^importFrom\\(rhdf5,h5write\\.default\\)" line)
(string-append "#" line)))
#t)))))
Copy link
Member

Choose a reason for hiding this comment

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

FYI: we no longer need #T at the end of build phases. On the master branch it still prints a warning, but on core-updates-frozen (to be merged today, hopefully) it's all fine.

Long ago we used booleans to indicate success or failure for build phases. Then we switched to using exceptions for failures. With core-updates-frozen the build systems are finally fully aware of this change.

Copy link
Member

Choose a reason for hiding this comment

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

Could you please add a link to the upstream discussion above this build phase. It's always good to keep a record right in the code.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

Thanks for the feedback. Much appreciated. I tried to address both points. Please double-check though since I did not test actually building sleuth using this latest iteration. 😉

* Drop superfluous import of function not exported by rhdf5 anymore.
@rekado
Copy link
Member

rekado commented Dec 20, 2021

I'm confused by the discussion around r-sleuth. Is this change a workaround for an upstream problem or a fix? Will sleuth not only build fine but also be usable?

IIUC upstream essentially abandoned sleuth. If that is so, perhaps we should just remove the package. Or is it worth keeping?

Sorry for being so dense. (It's heavy bones.)

@mschilli87
Copy link
Contributor Author

r-sleuth imports a function from r-rhdf5 that used to be exported nut was depricated and later removed in newer versions. However, r-sleuth doesn't use this function at all. My guess is it used to be used (didn't check) and was replaced at some point (maybe even as a response to the deprication notice by upstream, just guessing) without removing the import.
So it definitely is not just a weird build-issue workaround but r-sleuth is fully functional with this test. I used it myself after submitting this PR. I think it has value and is worth keeping. The creators are scientists and I guess none of the original authors (but the PI) are working in that lab anymore. That said, at least for kallisto I have seen maintainers taking over and I am hopeful r-sleuth will receive more love again in the future as well. As long as it is not officially dead, I'd like to keep it around if possible.
I am not aware of an alternative offering the same features so it certainly remains useful for my research.
That said, I'd argue the issues should be fixed upstream (which is why I opened a PR there) so I'd consider this a (hopefully temporary) workaround for what one could call upstream packaging bug (since only metadata require patching).

Hope this helps, sorry for not being dense at all (as usual). 😉

@rekado
Copy link
Member

rekado commented Dec 20, 2021

Thank you, this sounds all reasonable.

I've applied and pushed your patch. Thanks!

@rekado rekado closed this Dec 20, 2021
@mschilli87 mschilli87 deleted the r-sleuth-fix branch December 21, 2021 10:29
@mschilli87
Copy link
Contributor Author

In the meantime the upstream bug has been fixed. So with the next version bump, these changes should be undone.

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 this pull request may close these issues.

2 participants