On 8 July 2021 at 10:41, Uwe Ligges wrote: | It was likely an interim version of Rcpp (which never made it to CRAN) | that was causing trouble.
Yup, our bad. While we test against 2300+ reverse depends, we only test against them _under the most recent versions we have seen_ which affects the Rcpp under test, and we do not have an (explicit) test vector for these 2300+ packages tested against everything built under the _CRAN_ version of Rcpp. So the need for one small adjustment slipped til we made it to CRAN, which caused this issue, and one for broom.mixed (as we communicated off-list to Ben). In short, no matter how large you try to span your test coverage and surface, there will always be a residual chance for something slipping through. Dirk -- https://dirk.eddelbuettel.com | @eddelbuettel | e...@debian.org ______________________________________________ R-package-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel
