Cran issues with Stan packages on r-devel-linux-x86_64-debian-clang + r-devel-linux-x86_64-debian-gcc

Hi!

I checked the CRAN results for a number of Rstan dependent packages and all do ERROR for the platforms

r-devel-linux-x86_64-debian-clang + r-devel-linux-x86_64-debian-gcc

which is essentially a Debian 9.21 with gcc 9.2.1 / clang 9.0.0.

Can we ignore this? Anyone an idea on this? The packages I checked are my own (RBesT, OncoBayes2), Rstan & rstanarm … probably the others are also affected.

Sebastian

1 Like

I don’t think it has anything to do with the compiler because the new error is from R code due to matrix not inheriting from array anymore. So, this internal bayesplot function returns FALSE, which results in an error

So this is something for @Jonah to look into? Bayesplot is also showing these errors.

Yeah, I imagine we have a few weeks because this just got pushed into r-devel.

I guess you folk have also received the ‘fix it soon to stay on cran’ mail? I’m traveling and can’t easily check but I don’t think I check for class matrix at all, so I assume it’s just the Stan dependency that triggered the mail to me.

It is probably fixed by the new bayesplot on CRAN, but if not, post.