Resurrect package on CRAN

[zeileis]

Mikis @mstasinopoulos and Fernanda @fdebastiani, the package is currently archived on CRAN because the last version 2.1-2 had issues with examples in \donttest{...} that failed and were not corrected in time, see:

https://www.stats.ox.ac.uk/pub/bdr/donttest/gamlss.ggplots.out

I'm not sure if these problems were fixed already but it looks to me that loading/importing all packages needed for the examples should hopefully be straightforward. Let me know if I can help with this.

Additionally, I have updated the Authors@R in the DESCRIPTION and the README.md and added a NEWS.md.

[zeileis]

The package is still archived on CRAN but I think it would be good to have it back in order to use it for examples, etc. as Mikis did in his Function.qmd.

[fdebastiani]

Thanks Achim,

Do you have any suggestion to make the example properly written for CRAN?

Em 13 de mar de 2024, à(s) 23:02, Achim Zeileis @.***> escreveu:

The package is still archived on CRAN but I think it would be good to have it back in order to use it for examples, etc. as Mikis did in his Function.qmd https://github.com/gamlss-dev/admin/blob/main/Function.qmd.

— Reply to this email directly, view it on GitHub https://github.com/gamlss-dev/gamlss.ggplots/issues/1#issuecomment-1996256663, or unsubscribe https://github.com/notifications/unsubscribe-auth/AE7QMKZAFT2IWSZQQUH5C2LYYEAKXAVCNFSM6AAAAAA62Z6KECVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSOJWGI2TMNRWGM. You are receiving this because you were mentioned.

[zeileis]

As a general strategy:

• The code you put into \donttest{...} needs to run successfully when tested. The CRAN does not run these checks daily but occasionally.
• To check whether the \donttest{...} or \dontrun{...} examples actually run, you can use: R CMD check --run-donttest and/or --run-dontrun, respectively.

In this specific case:

• The error message in the output file says:

  pp+ geom_vline(xintercept = MM$pc, colour = "gray") Error in geom_vline(xintercept = MM$pc, colour = "gray") : could not find function "geom_vline" Execution halted

  This indicates that your example assumes that the ggplot2 package (where geom_vline resides) was attached to the user's search path but it actually isn't. So maybe just adding library("ggplot2") in the example is already sufficient.

• Finally, there is a message saying: executing %dopar% sequentially: no parallel backend registered. So maybe gamlss.foreach is not needed for this example?

[mstasinopoulos]

Dear Achim

I just pass the gamlss.data to CRAN

Next the gamlss.ggplots with correction as suggested

Thanks Mikis

On 14 Mar 2024, at 22:06, Achim Zeileis @.***> wrote:

As a general strategy:

The code you put into \donttest{...} needs to run successfully when tested. The CRAN does not run these checks daily but occasionally. To check whether the \donttest{...} or \dontrun{...} examples actually run, you can use: R CMD check --run-donttest and/or --run-dontrun, respectively. In this specific case:

The error message in the output file https://www.stats.ox.ac.uk/pub/bdr/donttest/gamlss.ggplots.out says:

pp+ geom_vline(xintercept = MM$pc, colour = "gray") Error in geom_vline(xintercept = MM$pc, colour = "gray") : could not find function "geom_vline" Execution halted This indicates that your example assumes that the ggplot2 package (where geom_vline resides) was attached to the user's search path but it actually isn't. So maybe just adding library("ggplot2") in the example is already sufficient.

Finally, there is a message saying: executing %dopar% sequentially: no parallel backend registered. So maybe gamlss.foreach is not needed for this example?

— Reply to this email directly, view it on GitHub https://github.com/gamlss-dev/gamlss.ggplots/issues/1#issuecomment-1998332464, or unsubscribe https://github.com/notifications/unsubscribe-auth/AE7QN7W6LK7FA36WYCLIWRTYYH7MNAVCNFSM6AAAAAA62Z6KECVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSOJYGMZTENBWGQ. You are receiving this because you were mentioned.

[zeileis]

Thanks, Mikis.

When you have updated gamlss.ggplots here on GitHub, let me know and I'm happy to test it as well before submitting it to CRAN.

Re: gamlss.data. We need the other direction, I think. The CRAN version was much more recent and needs to be imported into the GitHub repository (see https://github.com/gamlss-dev/gamlss.data/issues/1).

[zeileis]

Just for the record: Closing this issue because gamlss.ggplots is back on CRAN.