You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Some packages on CRAN follow x.y-z semantic versioning convention, for example sf, abind.
riskmetric::pkg_ref changes the semantics to be x.y.z due to the usage of base::package_version() specifically when working with the source packages.
I think it is conceptually not ideal, because later it is not possible to join assessment data with the original package DESCRIPTION meta info.
I am opening this issue to understand better why the line of code linked above is necessary, and why can't we use a string representation of the version. I am definitely missing some broader context here.
My initial thought was version comparisons, but it works fine with the standard library function:
Some packages on CRAN follow
x.y-z
semantic versioning convention, for example sf, abind.riskmetric::pkg_ref
changes the semantics to bex.y.z
due to the usage ofbase::package_version()
specifically when working with the source packages.I think it is conceptually not ideal, because later it is not possible to join assessment data with the original package DESCRIPTION meta info.
I am opening this issue to understand better why the line of code linked above is necessary, and why can't we use a string representation of the version. I am definitely missing some broader context here.
My initial thought was version comparisons, but it works fine with the standard library function:
cc @llrs-roche
The text was updated successfully, but these errors were encountered: