On 23 April 2020 at 19:06, Stuart Lacy via R-package-devel wrote: | I've been told to fix the problems that my package 'multistateutils' has on | OSX: | https://cran.r-project.org/web/checks/check_results_multistateutils.html | | This is seemingly something to do with my Rcpp code, but I can't debug it | on my local (Linux) machine, as when I check my code using Rhub it passes | the checks: | https://builder.r-hub.io/status/multistateutils_1.2.4.tar.gz-9295de3b93d746eb93983d33269cf08d | | Does anyone have any tips for testing their packages on OSX, or any | OSX-specific issues with Rcpp?
Yes -- Rcpp 1.0.4 passed on macOS at CRAN was released there. And only post-release, many macOS users (who could have tested pre-release) made a lot of noise concerning issues on their machines. Eventually, and thanks to careful digging by Winston, it was seen as being caused by a newer (relative to what CRAN has) macOS SDK issued by Apple. And Rcpp 1.0.4.6 was prepared almost three weeks ago, admitted on CRAN about two weeks ago ... but still only has one out of three binaries for macOS built on CRAN as you see at https://cloud.r-project.org/web/packages/Rcpp/index.html Given that the _binary_ package for macOS may be useless for you, you could try to force installation of the _source_ package for your testing. It should work just fine. If you want to see more details, see issue #1060 at the Rcpp GitHub repo. Dirk -- http://dirk.eddelbuettel.com | @eddelbuettel | e...@debian.org ______________________________________________ R-package-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel
