CRAN Archival

[wincowgerDEV]

Just got this email, will anyone from hyperspec be fixing this or need help with fixing it? Open Specy is very dependant on hyperspec but if the plan is to remove hyperspec from Cran do I have permission to put the functions im dependant on into the OpenSpecy package?

Dear maintainers,

This concerns the CRAN packages

OpenSpecy hyperSpec ir

maintained by one of you:

Claudia Beleites Claudia.Beleites@chemometrix.gmbh: hyperSpec Henning Teickner henning.teickner@uni-muenster.de: ir Win Cowger wincowger@gmail.com: OpenSpecy

We have repeatedly asked for an update fixing the check problems shown on https://cran.r-project.org/web/checks/check_results_hyperSpec.html with no reply from the maintainer thus far.

Thus, package hyperSpec is now scheduled for archival on 2024-05-04, and archiving this will necessitate also archiving its CRAN strong reverse dependencies.

Please negotiate the necessary actions.

Best

[henningte]

Just got this email, will anyone from hyperspec be fixing this or need help with fixing it? Open Specy is very dependant on hyperspec but if the plan is to remove hyperspec from Cran do I have permission to put the functions im dependant on into the OpenSpecy package?

I second this. The 'ir' package also has some core functions that depend on 'hyperSpec' functions and in case the issues are not fixed, I would also like to put these fuctions into 'ir' if you allow this.

[GegznaV]

Thank you, @wincowgerDEV and @henningte, for this notice. I was not aware of the issue.

@r-hyperspec/maintainers, @r-hyperspec/r-hyperspec, hyperSpec is on the way to be removed from CRAN. Less than 2 weeks are left. @cbeleites, can you look into this?

As I understand, the main issues are related to the vignettes (PDF) and Rd files.

[GegznaV]

Who knows what is the procedure to replace a maintainer? Several years ago we talked with the team that with a new major release of hyperSpec I would be the official maintainer.

[henningte]

@GegznaV ,

From the CRAN submission checklist:

If for some reason the submission has to be made by someone else (for example, a co-author) this needs to be explained, and the designated maintainer will need to confirm the submission.

Explain any change in the maintainer’s email address and if possible send confirmation from the previous address (by a separate email to CRAN-submissions@R-project.org) or explain why it is not possible.

I've never transfered a package myself, but here's a Stackoverflow thread with more hints: https://stackoverflow.com/questions/39223320/transferring-maintainership-of-an-r-package-on-cran

Thank you for taking care of this issue.

[bryanhanson]

Go Vilmantas!

On Apr 27, 2024, at 5:53 AM, Henning Teickner @.***> wrote:

@GegznaV https://github.com/GegznaV ,

From the CRAN submission checklist https://cran.r-project.org/web/packages/submission_checklist.html:

If for some reason the submission has to be made by someone else (for example, a co-author) this needs to be explained, and the designated maintainer will need to confirm the submission.

Explain any change in the maintainer’s email address and if possible send confirmation from the previous address (by a separate email to @. @.>) or explain why it is not possible.

I've never transfered a package myself, but here's a Stackoverflow thread with more hints that sound reasonable: https://stackoverflow.com/questions/39223320/transferring-maintainership-of-an-r-package-on-cran

Thank you for taking care of this issue.

— Reply to this email directly, view it on GitHub https://github.com/r-hyperspec/hyperSpec/issues/124#issuecomment-2080589545, or unsubscribe https://github.com/notifications/unsubscribe-auth/AABCIPT4VXON2456YHM5MKLY7ONWLAVCNFSM6AAAAABGQTSETOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAOBQGU4DSNJUGU. You are receiving this because you are on a team that was mentioned.

============= Bryan A. Hanson Prof. Emeritus, DePauw University Chemistry & Biochemistry @. @. Nerdy blog: ChemoSpec.org I’m at GMT -7 year round

[cbeleites]

Dear all,

@GegznaV : I'd be more than happy to see you the new official maintainer of hyperSpec. I don't (as you know since years) and I won't have time in the foreseeable future to take anything even close to proper care of hyperSpec. I will be able to accept the maintainer change, though. (I've changed email in the past and I think changing maintainer is basically the same procedure, plusminus things that changed in the last x years)

Wrt to the CRAN NOTEs, I've tried to submit fixes. This was not successful, even though everything I submitted did pass locally --as-cran tests, and all win-buider versions as required. But it did not not pass the actual CRAN incoming tests where they test the package without suggested packages installed. Which apparently is not simulated by --as-cran nor on win-builder. This made the submission extremely tedious, and I ran out of time.

@GegznaV, I can let you have a dirty fixed version (on Wed 1st) that takes care of the issues listed on the CRAN test page, and that also skips the plotmat() test which fails if (suggested) plotrix is not available. If you decide so, it may be an option to remove all vignettes and point people to https://r-hyperspec.github.io/hyperSpec/ instead.

[cbeleites]

@henningte, @wincowgerDEV

Just got this email, will anyone from hyperspec be fixing this or need help with fixing it? Open Specy is very dependant on hyperspec but if the plan is to remove hyperspec from Cran do I have permission to put the functions im dependant on into the OpenSpecy package?

I second this. The 'ir' package also has some core functions that depend on 'hyperSpec' functions and in case the issues are not fixed, I would also like to put these fuctions into 'ir' if you allow this.

It's all under a FOSS license, you're welcome to take whatever you need.

OpenSpecy AFAIK uses only few file import functions, so that should be fairly easy to disentangle. (And for the JCAMP-DX import you may anyways want to use @bryanhanson's readjdx directly).

ir is more complicated I guess since it imports the class.

[bryanhanson]

Re: the CRAN tests. A while back they started requiring the what Claudia noticed: it must pass both with and w/o the suggested packages installed, and --as-cran is not (currently) doing that. Here's how I test in my local make file; you have to use environmental variables.

checkLean: $(TGZ)
      export _R_CHECK_CRAN_INCOMING_=FALSE; \
      export _R_CHECK_DEPENDS_ONLY_=TRUE; \
      R --no-init-file CMD check --as-cran $(TGZ)
checkFull: $(TGZ)
      export _R_CHECK_CRAN_INCOMING_=FALSE; \
      R --no-init-file CMD check --as-cran $(TGZ)

where $(TGZ) is the tar ball path. You have to run both of these statements.

The entry in WRE (Writing R Extensions) about _R_CHECK_DEPENDS_ONLY_ and friends is a little hard to parse but these settings are working as of last Friday.

[GegznaV]

I can let you have a dirty fixed version (on Wed 1st) that takes care of the issues listed on the CRAN test page, and that also skips the plotmat() test which fails if (suggested) plotrix is not available. If you decide so, it may be an option to remove all vignettes and point people to https://r-hyperspec.github.io/hyperSpec/ instead.

1) Regarding the vignettes: I believe it may be reasonable to consider removing them if addressing the associated issues is time-consuming. We must avoid being expelled from CRAN and there is not much time left.

2) Regarding the release of the version of hyperSpec requiring make: I lack a functional setup to build the package via make. Consequently, I am uncertain as to who should undertake the submission of the current fixes. Should it fall to me, I would require a fully prepared .tar.gz containing the package for submission.

[cbeleites]

export _R_CHECK_DEPENDS_ONLY_=TRUE; \

Thanks @bryanhanson, that helps indeed.

Do you have any hints how to get an interactive session working with a NoSuggests scenario?

[cbeleites]

@GegznaV I submitted since that was a minor step after getting the .tar.gz to pass checks OK with (R-current + R-devel) x (NoSuggests + normal), plus 3x win-builder.
I'd be happy to make you the new official maintainer as soon as you wish (and from my side: the sooner the better, but as long as nothing happens at CRAN I'm also happy if you take your time). Since that is done with a new submission to CRAN which you should do, please let me now well in advance so I'm available to confirm the change.

It has only a stub vignette ("hyperSpec") left that points to the articles at https://r-hyperspec.github.io/hyperSpec/articles/index.html.

I made it available as release https://github.com/r-hyperspec/hyperSpec/releases/tag/v0.100.2.

---

CRAN says it passed checks there as well now.

[GegznaV]

Thank you, @cbeleites, for taking care of this issue. Now it is solved. Now I'm reviewing the code and trying to remember which issues are the most important and how to make the transition to the new version of hyperSpec even smoother. I'll present my ideas in a new issue.

Thank you @wincowgerDEV and @henningte, for announcing the issue here on GitHub. A fixed version of hyperSpec is on CRAN.

[bryanhanson]

@cbeleites As far as interactive tests go, I just do those manually before submitting. I have no magic for that.