CRAN Release 1.4.0

[m-jahn]

CRAN checklist for re-release:

• [x] reduce dependencies <= 20, move BioC deps to suggests
• [x] usethis::use_cran_comments()
• [x] Update (aspirational) install instructions in README
• [x] Proofread Title: and Description:
• [x] Check that all exported functions have @return and @examples
• [x] Check that Authors@R: includes a copyright holder (role 'cph')
• [x] Check licensing of included files
• [x] Review https://github.com/DavisVaughan/extrachecks

Prepare for release:

• [x] urlchecker::url_check()
• [x] devtools::build_readme()
• [x] devtools::check(remote = TRUE, manual = TRUE)
• [x] devtools::check_win_devel()

Submit to CRAN:

• [x] usethis::use_version('patch')
• [ ] devtools::submit_cran()
• [ ] Approve email

Wait for CRAN...

• [ ] Accepted :tada:
• [ ] usethis::use_github_release()
• [ ] usethis::use_dev_version(push = TRUE)

[m-jahn]

@showteeth I see that you have released the package initially on CRAN, version 0.7.0, but it got removed again. Now with some additional work it should be possible to re-release it on CRAN. Are you up for this and, most importantly, responsive to the CRAN comments, emails etc that appear during the submission?

[m-jahn]

@showteeth version 1.4.0 is now merged, can be submitted to CRAN soon. Please get back to me with your comments, otherwise I can also try to submit next week.

[m-jahn]

@showteeth any thoughts on CRAN release? Your participation is required as you as the maintainer gets the emails from CRAN.

[m-jahn]

@showteeth we can still get this back on CRAN. Is there any hope you will ever respond? You can say no of course.

[m-jahn]

@showteeth package should still be submitted to CRAN; I can ask CRAN to transfer maintainership if you are unable to respond.

[showteeth]

@m-jahn OK! You are very familiar with this package. Sorry for the late response, I just finished my PhD! Next, we can maintain this package together.

[m-jahn]

dear @showteeth, congratulations to your PhD! My apologies, I wasn't aware that you were occupied with that.

Regarding the package, you can either submit it to CRAN yourself, or I can do that, but it is important that you are available under the given email adress . If I submit the package to CRAN, we need to transfer the cre tag in the author list to me, and CRAN will ask you by email to confirm this change (that's why the above mentioned email address must be valid).

[showteeth]

OK! I will first try to submit it to CRAN by myself.

[m-jahn]

Go for it. There's one additional open PR #41 that could be merged before submission. It will fix some of the open issues.

[showteeth]

I've already done it. Please check the email.

[m-jahn]

Alright, I have confirmed the submission.

[m-jahn]

the automatic pretest from CRAN found some problems which I think can be easily fixed. I will make some changes and open a PR, and then we can try to submit again. Will get back with more details.

[m-jahn]

CRAN internal checks found: Status: 4 NOTEs

* checking CRAN incoming feasibility ... [12s] NOTE
Maintainer: 'Michael Jahn <[email protected]>'
New submission
Package was archived on CRAN

2. -- comments that are not a NOTE?

Found the following (possibly) invalid URLs:
  URL: https://academic.oup.com/bioinformatics/article/37/3/422/5879987?login=false
    From: inst/doc/ggcoverage.html
          README.md
    Status: 403
    Message: Forbidden
  URL: https://academic.oup.com/peds/article/10/7/743/1593029?login=false
    From: inst/doc/ggcoverage.html
          README.md
    Status: 403
    Message: Forbidden
  URL: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8006021/ (moved to https://pmc.ncbi.nlm.nih.gov/articles/PMC8006021/)
    From: inst/doc/ggcoverage.html
          README.md
    Status: 301
    Message: Moved Permanently
For content that is 'Moved Permanently', please change http to https,
add trailing slashes, or replace the old by the new URL.
Please use DOIs for the following publisher URLs:
  https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8006021/

* checking DESCRIPTION meta-information ... NOTE
Maintainer field differs from that derived from Authors@R
  Maintainer: 'Michael Jahn <[email protected]>'
  Authors@R:  'Yabing Song <[email protected]>'

* checking Rd cross-references ... NOTE
Found the following Rd file(s) with Rd \link{} targets missing package
anchors:
  FormatTrack.Rd: import.gff
  LoadTrackFile.Rd: import.gff
  geom_base.Rd: translate
  geom_gene.Rd: import.gff
  geom_transcript.Rd: import.gff
Please provide package anchors for all Rd \link{} targets not in the
package itself and the base packages.

* checking examples ... [84s] NOTE
Examples with CPU (user + system) or elapsed time > 10s
                 user system elapsed
geom_tad        29.09   2.28   31.22
geom_transcript 10.83   0.19   11.00

Some of those are standard and usually nothing is done about them, particularly 1 and 5. The others should be fixed.

[m-jahn]

@showteeth now that you have merged the latest PR you can resubmit the package. The confirmation email should be sent to your email address now because I changed the maintainer email.

[showteeth]

@m-jahn I have tried again, but some new issues have arisen. I emailed the maintainer, but the solutions are hard to meet. Why not abandon to submit to CRAN, install from GitHub is pretty good.  BTW, I forward the email to you.

[m-jahn]

Thanks for trying! I think many people rely on CRAN and the safety and high standards it comes with. The regular checks including checks on the package's dependencies are something useful. Without that , upstream changes that make the package unusable can go unnoticed.

I dont think that these issues are particularly difficult to fix, just need to find out what exactly the CRAN NOTEs mean. No need to give up already, now that we came so far

[m-jahn]

one issue I was not aware is actually the thing with package suggests:. Apparently all suggested packages should only be used when loaded with a check that tests if the package is available like requireNamespace().

See this answer: https://stackoverflow.com/questions/79067855/r-cmd-check-cannot-find-packages-during-package-build-with-r-check-depends-only

[m-jahn]

Let's list the main points here and try to address them:

1. package suggests: vignette building and/or examples fail when suggests libs are not present. Solution: load packages nad run tests only conditionally on test that packages are installed

2. package size: 30 Mb, should be 5 Mb. Solution: try to reduce size of some files, but this is usually not a show stopper if it can be justified why this package is a bit larger --> test files

3. test run time: geom_tad 14.651 0.666 15.322; geom_transcript 5.692 0.044 5.738. Solution: reduce test file size or add a dontrun statement for these examples -- then runnning them is omitted.

I think this is doable.

[m-jahn]

I added fixes for all of the problems and bumped the version to 1.4.1. Also added explanation to cran comments regarding the file size, which is now reduced from 30 to 6 MB. That should be fine, I maintain packages that are larger than that.

You are free to resubmit

[m-jahn]

Did you manage to re-submit and get feedback @showteeth ?

[m-jahn]

Any news from CRAN on the latest update @showteeth ?