[Bioc-devel] [External] different generator for random integers in R-devel
|uke-t|erney @end|ng |rom u|ow@@edu
Tue Mar 19 14:50:44 CET 2019
Use RNGversion("3.5.0") instead.
On Tue, 19 Mar 2019, Stadler, Michael wrote:
> Dear maintainers of the BioC build system
> I have a shy questions regarding R-devel versions on the build machines:
> Currently, these use different versions of R-devel:
> malbec2 | Linux ... | R Under development (unstable) (2019-01-21 r75999)
> tokay2 | Windows ... | R Under development (unstable) (2019-03-09 r76216)
> merida2 | OS X ... | R Under development (unstable) (2018-11-27 r75683)
> celaya2 | OS X ... | R Under development (unstable) (2019-01-22 r76000)
> A recent change in R-devel
> (mostly in r76160, see https://github.com/wch/r-source/commit/e4acfd1f240a42b3fc1474a4d97017114e4eb053)
> has changed the behaviour of sample()
> (see also https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=17494 and https://stat.ethz.ch/pipermail/r-devel/2018-September/076817.html).
> In order to maintain backward compatibility, we are using RNGkind(..., sample.kind = "Rounding")
> to temporarily fall back to the behaviour of R-3.5. However, this will cause an error on R-devel < r76160,
> where RNGkind() does not yet understand the "sample.kind" argument (currently malbec2, merida2 and celaya2).
> This will obviously resolve itself once R-devel gets updated on the build machines, latest upon release of BioC 3.9.
> I am just curious whether R-devel will be updated on the builders before that, so this issue does not mask other
> potential problems in our package (QuasR).
> Bioc-devel using r-project.org mailing list
Ralph E. Wareham Professor of Mathematical Sciences
University of Iowa Phone: 319-335-3386
Department of Statistics and Fax: 319-335-3017
241 Schaeffer Hall email: luke-tierney using uiowa.edu
Iowa City, IA 52242 WWW: http://www.stat.uiowa.edu
More information about the Bioc-devel