search

epiverse-trace / epiparameter

R package with library of epidemiological parameters for infectious diseases and functions and classes for working with parameters
https://epiverse-trace.github.io/epiparameter
Other
33 stars 11 forks source link

Decrease R mininum version to 4.0.0 #404

Closed joshwlambert closed 1 month ago

joshwlambert commented 1 month ago

This PR changes the minimum R version required by {epiparameter} to 4.0.0, after it was incremented to 4.1.0 in PR #384 (due to mitchelloharawild/distributional#128).

This PR also adds a R-CMD-check CI run for R 4.0 to check that the R package passes the check on the version stated in the DESCRIPTION.

This PR initially checked to see if the minimum R version could be reverted to 3.6.0 as it was before #384, however, due to {epireview} depending on {flextable} which in turn depends on {gdtools} which requires R 4.0, then this is required for the R-CMD-check workflow to install {epireview}.

Usage of the base pipe (|>) has been replaced in a vignette by the pipe from {magrittr} (%>%), with {magrittr} added as a suggested dependency.

The note in the design_principles.Rmd vignette on breaking the Epiverse R version policy has been removed as this will not longer be the case.

joshwlambert commented 1 month ago

This PR fails the R-CMD-check CI for both R v3.6.0 and v4.0.0. Due to {epireview} depending on {gdtools} which requires R 4.0 and then {distributional} which released v0.5.0 using the base pipe (|>). The base pipe has since been removed from {distributional} but this has not yet been released to CRAN.

Therefore, I will close this PR without merging, and retain the minimum R version of 4.1.0 for {epiparameter} and add a explicit R-CMD-check CI for R 4.1 on a separate PR.

The alternative would be to wait to for a new {distributional} release to enable the minimum R version to be 4.0, but this seems like a marginal gain and would be a blocker to releasing {epiparameter}.

joshwlambert commented 1 month ago

Closing, superseded by #405.