I am maintaining the EpiILMCT package in CRAN that has been removed recently 
because of the non-producibility issue. My package passes all the tests without 
warnings and errors. The issue comes from the tests examples of two functions 
that are based on Fortran subroutines (comparing the test .R files to the 
corresponding .Rout.save files). These have an option of fixing the (Fortran) 
random number generator used for making reproducible results. The difference of 
results were found in platforms that used the new version of GCC 8 or GCC 9. 
This might be because the RNG method used in random_number function is 
different between the used version GCC 6 or less and the new ones GCC 8 or 
greater. Platforms used version of GCC less than 8 have not produced any 
different results. So, does anyone have an idea of how to solve this issue.

Many thanks,


Sent from Mail<https://go.microsoft.com/fwlink/?LinkId=550986> for Windows 10

        [[alternative HTML version deleted]]

R-package-devel@r-project.org mailing list

Reply via email to