HenrikBengtsson / CBI-software

A Scientific Software Stack for HPC (CentOS oriented)
https://wynton.ucsf.edu/hpc/software/software-repositories.html
5 stars 2 forks source link

R: Don't install 'recommended' packages #19

Closed HenrikBengtsson closed 2 years ago

HenrikBengtsson commented 3 years ago

Issue

Since we install 'recommended' packages to the system library, users get warnings each time they update their packages;

> update.packages()
Warning: package 'boot' in library '/usr/lib/R/library' will not be updated
Warning: package 'class' in library '/usr/lib/R/library' will not be updated
Warning: package 'KernSmooth' in library '/usr/lib/R/library' will not be updated
Warning: package 'lattice' in library '/usr/lib/R/library' will not be updated
Warning: package 'MASS' in library '/usr/lib/R/library' will not be updated
Warning: package 'Matrix' in library '/usr/lib/R/library' will not be updated
Warning: package 'mgcv' in library '/usr/lib/R/library' will not be updated
Warning: package 'nlme' in library '/usr/lib/R/library' will not be updated
Warning: package 'nnet' in library '/usr/lib/R/library' will not be updated
Warning: package 'survival' in library '/usr/lib/R/library' will not be updated
...

This adds quite a bit of noise but it's also confusing and drives users to ignore warnings.

Decision

Don't install above 'recommended' packages. Instead, let users install them if they need them. Also, if they need them because they're needed by another package the user installs, then they'll be automatically installed.

To not install above packages when building R, use:

./configure --without-recommended-packages ...
HenrikBengtsson commented 3 years ago

In order to test our R build, we need to patch make check. R Core acknowledges the problem and is open to patches

HenrikBengtsson commented 3 years ago

This has been fixed for R 4.2.0.

For R 4.1.0, we need to install them in order to run the checks, but i guess we could uninstall the before deploying. Maybe it's sufficient to install just some of them, e.g. MASS for sure.

HenrikBengtsson commented 2 years ago

Done already for R 4.1.1