[R-pkg-devel] Deprecating apparent S3 method, changing name
A package of mine on CRAN has some old function names (not S3 methods) that include "." (e.g., "esa.plot"). In a new version I want to simultaneously 1. rename these functions to e.g. "esaPlot" 2. export a new S3 generic with the base name (e.g. "esa") and methods for different model types (e.g. "esa.secr"). When I deprecate the old functions (by exporting a shell function that calls .Deprecated and the new function), I get a package check note for e.g. esa.plot "* checking S3 generic/method consistency ... NOTE Apparent methods for exported generics not registered: etc." This is understandable, but how can I avoid it while properly deprecating the old functions? I seem to remember a recent question like this, but I'm sorry I cannot find it. Murray [[alternative HTML version deleted]] __ R-package-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel
Re: [R-pkg-devel] Deprecating apparent S3 method, changing name
Thanks for your excellent exposition of the problem, Ivan. The set-aside option would be attractive in the long term and make it easier to fix historical errors like mine. Presumably there is also a downside to creating a safe space for badly named functions. I am tossing up between (i) appealing to CRAN to allow the submission through with NOTEs until it feels safe to remove the functions with misleading names altogether, and (ii) removing them now and warning users about the new names in an onLoad message. Murray From: Ivan Krylov Sent: Friday, 27 September 2024 08:48 To: Murray Efford via R-package-devel Cc: Murray Efford Subject: Re: [R-pkg-devel] Deprecating apparent S3 method, changing name � Wed, 25 Sep 2024 05:13:31 + Murray Efford via R-package-devel �: > When I deprecate the old functions (by exporting a shell function > that calls .Deprecated and the new function), I get a package check > note for e.g. esa.plot "* checking S3 generic/method consistency ... > NOTE Apparent methods for exported generics not registered: etc." > > This is understandable, but how can I avoid it while properly > deprecating the old functions? R CMD check has a hard-coded list of functions that should not be considered S3 methods, in tools:::nonS3methods(). All other functions that start with the name of an exported generic followed by a dot will be considered unregistered S3 methods. Theoretically speaking, if there was a place in your package where you called the esa() generic with a user-supplied object, esa.plot() would be a reachable S3 method, because the user would be able to construct structure(list(), class = 'plot') and give it to your package code to call esa(plot_object) and have it dispatched to esa.plot(). (Which wouldn't be useful to the user, but that's how S3 works.) I'm afraid that the NOTE is unavoidable in a package that has both an esa() generic and an esa.plot() non-method function. Perhaps it could be explained in the submission comment? (Especially since esa.plot() currently seems unreachable as an S3 method to normal users of the package.) I wonder if it's possible to set aside a "please don't consider me for S3 dispatch" bit in the CLOSXP objects. Then package authors would be able to set it on their unfortunately named functions, R CMD check would look it up before considering them to be unexported methods, and the function findFunWithBaseEnvAfterGlobalEnv() that's responsible for looking up non-registered S3 methods during UseMethod() would skip over such closures. -- Best regards, Ivan [[alternative HTML version deleted]] __ R-package-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel
Re: [R-pkg-devel] Deprecating apparent S3 method, changing name
Check results vary. With win-builder both R-release (R 4.4.1) and R-devel I get
* checking S3 generic/method consistency ... NOTE
Apparent methods for exported generics not registered:
esa.plot
See section 'Registering S3 methods' in the 'Writing R Extensions'
manual.
But there's no such Note on mac-builder R 4.4.0 or my own R CMD check --as-cran
on Windows with R 4.3.2, nor apparently on R 4.4.1 linux.
Thanks for checking, but I think I do have a problem.
Murray
From: R-package-devel on behalf of Jan
van der Laan
Sent: Monday, 30 September 2024 00:46
To: r-package-devel@r-project.org
Subject: Re: [R-pkg-devel] Deprecating apparent S3 method, changing name
Hmm, the following doesn't seem to trigger warnings or notes in `R CMD
check --as-cran` (R 4.4.1; linux)
esa <- function(x) {
UseMethod("esa")
}
esa.plot <- function(y) {
.Deprecated("esaPlot")
esaPlot(y)
}
esaPlot <- function(y) {
message("esaPlot")
}
esa.numeric <- function(x) {
message("esa.numeric")
}
With NAMESPACE:
S3method(esa, numeric)
export(esa)
export(esa.plot)
export(esaPlot)
Am I doing something different than you are?
Jan
On 9/25/24 07:13, Murray Efford via R-package-devel wrote:
> A package of mine on CRAN has some old function names (not S3 methods) that
> include "." (e.g., "esa.plot"). In a new version I want to simultaneously
> 1. rename these functions to e.g. "esaPlot"
> 2. export a new S3 generic with the base name (e.g. "esa") and methods for
> different model types (e.g. "esa.secr").
>
> When I deprecate the old functions (by exporting a shell function that calls
> .Deprecated and the new function), I get a package check note for e.g.
> esa.plot
> "* checking S3 generic/method consistency ... NOTE
> Apparent methods for exported generics not registered: etc."
>
> This is understandable, but how can I avoid it while properly deprecating the
> old functions? I seem to remember a recent question like this, but I'm sorry
> I cannot find it.
>
> Murray
>
>[[alternative HTML version deleted]]
>
> __
> R-package-devel@r-project.org mailing list
> https://apc01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-package-devel&data=05%7C02%7Cmurray.efford%40otago.ac.nz%7Ce0eaedee90af46dc200e08dce07c68f8%7C0225efc578fe4928b1579ef24809e9ba%7C0%7C0%7C638632072195001025%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=G%2BO6NtGHVy9xvfu2yPGy%2BBKbnOrC6hNxFXTVxLxu3wQ%3D&reserved=0<https://stat.ethz.ch/mailman/listinfo/r-package-devel>
__
R-package-devel@r-project.org mailing list
https://apc01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-package-devel&data=05%7C02%7Cmurray.efford%40otago.ac.nz%7Ce0eaedee90af46dc200e08dce07c68f8%7C0225efc578fe4928b1579ef24809e9ba%7C0%7C0%7C638632072195023045%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=6hIjTj6E2HF8TGGwDb1JkVIjibztEgf10dSaxvvc8sg%3D&reserved=0<https://stat.ethz.ch/mailman/listinfo/r-package-devel>
[[alternative HTML version deleted]]
__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel
Re: [R-pkg-devel] Deprecating apparent S3 method, changing name
Almost too good to be true: with this in my NAMESPACE
export (esa.plot)
S3method(esa,plot)
it seems I can deprecate esa.plot by the usual route (calling .Deprecated and
passing call through to the new function esaPlot).
Thanks!
Murray
From: Ivan Krylov
Sent: Wednesday, 2 October 2024 00:45
To: Jan van der Laan
Cc: Murray Efford ; Murray Efford via
R-package-devel
Subject: Re: [R-pkg-devel] Deprecating apparent S3 method, changing name
� Tue, 1 Oct 2024 09:00:24 +0200
Jan van der Laan �:
> S3method(esa, plot, esaplotmethod)
>
> To register `esaplotmethod` as the method that is called when using
> `esa` with an object of type `plot`. This would prevent this issue.
> However, this probably (can't test as the note doesn't happen on my
> system) doesn't resolve the issue with the NOTE.
Excellent idea, thank you! It does work for R CMD check as well:
Sys.setenv("_R_CHECK_S3_METHODS_SHOW_POSSIBLE_ISSUES_" = 'TRUE')
tools::checkS3methods(dir = '.') # used to show esa.plot as well
# Apparent methods for exported generics not registered:
# fxi.contour fxi.mode
No need for extra bits or attributes on the esa.plot function, just
register a method for "esa" generic, "plot" class to prevent S3
dispatch from reaching the function in question.
--
Best regards,
Ivan
[[alternative HTML version deleted]]
__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel
[R-pkg-devel] Pre-test failure with RcppParallel on Windows R-devel
Submitting secr 5.2.0 I get this pre-test installation ERROR which is opaque to me. I haven't changed the relevant code, and the package passes checks on Winbuilder (below) and other systems (with a false positive URL fail NOTE). * checking whether package 'secr' can be installed ... ERROR Installation failed. See 'd:/RCompile/CRANincoming/R-devel/secr.Rcheck/00install.out' for details. which includes undefined reference to `RcppParallel::tbbParallelFor(unsigned long long, unsigned long long, RcppParallel::Worker&, unsigned long long, int)' Winbuilder and the CRAN precheck appear to use identical settings: * using log directory 'd:/RCompile/CRANguest/R-devel/secr.Rcheck' * using R Under development (unstable) (2025-01-22 r87618 ucrt) * using platform: x86_64-w64-mingw32 * R was compiled by gcc.exe (GCC) 13.3.0 GNU Fortran (GCC) 13.3.0 * running under: Windows Server 2022 x64 (build 20348) Any ideas? I'm guessing it's a transient CRAN problem. [[alternative HTML version deleted]] __ R-package-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel
Re: [R-pkg-devel] Pre-test failure with RcppParallel on Windows R-devel
Yes, that fixed it. My package was quickly re-tested and is now on CRAN. Thanks to all. Murray From: Kevin Ushey Sent: Friday, 24 January 2025 15:10 To: Murray Efford Cc: R-package-devel@r-project.org Subject: Re: [R-pkg-devel] Pre-test failure with RcppParallel on Windows R-devel Hi Murray, (repeating some of the conversation we had off-list) Sorry for the trouble -- I was just in the process of submitting RcppParallel to CRAN, and the first submission I made had this same issue. That should now be fixed up with RcppParallel 5.1.10, which is now on CRAN. Hopefully your next submission should go smoothly. Thanks, Kevin On Thu, Jan 23, 2025 at 5:14 PM Murray Efford via R-package-devel wrote: > > Submitting secr 5.2.0 I get this pre-test installation ERROR which is opaque > to me. I haven't changed the relevant code, and the package passes checks on > Winbuilder (below) and other systems (with a false positive URL fail NOTE). > > * checking whether package 'secr' can be installed ... ERROR Installation > failed. See 'd:/RCompile/CRANincoming/R-devel/secr.Rcheck/00install.out' for > details. > which includes > undefined reference to `RcppParallel::tbbParallelFor(unsigned long long, > unsigned long long, RcppParallel::Worker&, unsigned long long, int)' > > Winbuilder and the CRAN precheck appear to use identical settings: > > * using log directory 'd:/RCompile/CRANguest/R-devel/secr.Rcheck' > * using R Under development (unstable) (2025-01-22 r87618 ucrt) > * using platform: x86_64-w64-mingw32 > * R was compiled by > gcc.exe (GCC) 13.3.0 > GNU Fortran (GCC) 13.3.0 > * running under: Windows Server 2022 x64 (build 20348) > > Any ideas? I'm guessing it's a transient CRAN problem. > > > > > [[alternative HTML version deleted]] > > __ > R-package-devel@r-project.org mailing list > https://apc01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-package-devel&data=05%7C02%7Cmurray.efford%40otago.ac.nz%7Cec031787d425485d25ad08dd3c1c6162%7C0225efc578fe4928b1579ef24809e9ba%7C0%7C0%7C638732814784925323%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=wtlopPdXVxrQorfWcztJkHz%2BRCuMpZ7o5zIRZ8Zj3G8%3D&reserved=0<https://stat.ethz.ch/mailman/listinfo/r-package-devel> [[alternative HTML version deleted]] __ R-package-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel
