Prep release for SISMID 2025

[martinamorris]

Prepare for release:

• git pull
• Check current CRAN check results
• Polish NEWS
• urlchecker::url_check()
• devtools::check(remote = TRUE, manual = TRUE)
• devtools::check_win_devel()
• revdepcheck::revdep_check(num_workers = 4) - package no longer maintained; no revdeps for this pkg anyway
• Update cran-comments.md
• git push
• Draft blog post

Submit to CRAN:

• usethis::use_version('minor')
• devtools::submit_cran()
• Approve email

Wait for CRAN...

• Accepted 🎉

[martinamorris]

@mbojan I'm getting the following Note when checking the package build for CRAN release:

❯ checking dependencies in R code ... NOTE
  Namespace in Imports field not imported from: 'DT'
    All declared Imports should be used.

0 errors ✔ | 0 warnings ✔ | 1 note ✖

I've checked the code, and we do have uses of DT::

Any idea why this NOTE is being generated?

[mbojan]

Yeah, DT is missing from NAMESPACE, but I now see it actually conflicts with shiny package. I'll sort it out.

[mbojan]

Should be silent now. It's all because the app code is in inst folder so technically not a "package code", so invisible to r cmd check. All this fooled R into inferring that DT is not used at all.

[martinamorris]

Thanks so much @mbojan.

I'd love to update this package to the current Shiny development standards. Will probably start with an AI assisted translation, just for fun. But I'm also open to paying someone to do the update (from the SISMID teaching money I get).

[martinamorris]

@mbojan not a show stopper and not urgent, but using `` I get this error at the end (that is not an RCMD error)

ERROR: Unknown command "TMPDIR=C:/Users/Martina". Did you mean command "serve"?
   Warning message:
   In system2("quarto", "-V", stdout = TRUE, env = paste0("TMPDIR=",  :
     running command '"quarto" TMPDIR=C:/Users/Martina Morris/AppData/Local/Temp/RtmpwLDaVU/file71f07ea25d05 -V' had status 1

I don't see the command TMPDIR=C:/Users/Martina in findinfiles, so this is an upstream error I'm not sure how to trace.

[mbojan]

Hmm, strange. When does it happen?

[martinamorris]

Sorry, not sure how that sentence got munged, but I meant "but using devtools::check(remote = TRUE, manual = TRUE)"

[mbojan]

Even stranger. This error looks as if R wants to run Quarto , but it should not happen when running the package check