[R-pkg-devel] \linkS4class anchoring
Killick, Rebecca
r@k||||ck @end|ng |rom |@nc@@ter@@c@uk
Fri Apr 11 18:00:16 CEST 2025
Hi Fellow Package Maintainers,
I received the following helpful message from the CRAN maintainers about a new check that my EnvCpt package is failing:
"CRAN would like to move its package web pages to providing (static) HTML refmans instead of PDF refmans, which needs Rd cross-references to Rd \link{} targets not in the package itself nor in the base packages to use package anchors, i.e., use \link[PKG]{FOO} (see section "Cross-references" in "Writing R Extensions")."
The CRAN package checks here:
https://cran.r-project.org/web/checks/check_results_EnvCpt.html
note that:
"Found the following Rd file(s) with Rd \link{} targets missing package anchors: envcpt.Rd: cpt-class"
All of the \link uses within my package are anchored except for the \linkS4class{cpt} call so I think this is what it is referring to. I've checked the writing R extensions "Cross-references" section and it just says to use \linkS4class{<class>} but does so in reference to a class within the current package - which is what has been working up until now.
The class is within the changepoint package so, I tried the obvious inference, \linkS4class[changepoint]{cpt} but this creates the following warning on check:
"prepare_Rd: ./man/envcpt.Rd:63: unexpected TEXT '[changepoint]', expecting '{'
checkRd: (7) envcpt.Rd:63: Tag LIST is invalid in a \code block"
Has anyone else come across this (needing to reference a class in another package) and knows how to anchor it?
I could be lazy and just delete the reference to solve the problem but, as the class is given as outputs, I think it is helpful to users who haven't come across the class before to have a direct link to the structure of it.
Thanks,
Rebecca
Pronouns: They/Them/Theirs
Please note: Whilst I may choose to correspond by email during evenings and weekends as it fits best with my preferred working pattern, I do not expect responses outside your usual working pattern.
More information about the R-package-devel
mailing list