[R-pkg-devel] size of package and splitting into 2

Neal Fultz nfultz @ending from gm@il@com
Mon Jan 14 18:47:11 CET 2019


If you re-exported all the psychTools functions from psych, I doubt a
typical user would notice or care. For example, this is how dplyr split out
some of it's functions in to another package:
https://github.com/tidyverse/dplyr/blob/master/R/reexport-tidyselect.R

On Mon, Jan 14, 2019 at 9:03 AM William Revelle <lists using revelle.net> wrote:

> Dear friends,
>
> For the last several iterations of psych I have received the Note that the
> installed size is 6.7 MB with two subdirectories greater than 1 MB.
>
> In the most recent submission to CRAN this led to an automatic rejection
> which then Uwe kindly overrode to allow the installation on CRAN.
>
> In order to make life easier for the CRAN folk, I would like to try to
> avoid this problem in the future.
>
> The only  way I can see to get around this note is to split the package
> into 2:  psych and psychTools where psych would require psychTools but
> psychTools would not require psych.   The presumption is that with
> appropriate splitting, both packages would meet the 1 MB suggestion.
>
> However, this implies that loading psych still takes up just as much space
> as before (since it would also load psychTools) and so it is not clear if
> this is a benefit. It might even be detrimental to the naive user who
> doesn’t load them both.  It would be useful, perhaps, for those who use
> psych for the extra features rather than core psychometric ones.  They
> would just need to load psychTools.
>
> I am not sure how well the splitting would work in terms of solving the
> problem and before embarking on this exercise,  I would appreciate
> suggestions as to whether it is necessary (or at least useful).
>
> Thanks,
>
> Bill
>
>
> William Revelle            personality-project.org/revelle.html
> Professor                                 personality-project.org
> Department of Psychology www.wcas.northwestern.edu/psych/
> Northwestern University    www.northwestern.edu/
> Use R for psychology         personality-project.org/r
> It is 2   minutes to midnight   www.thebulletin.org
>
> ______________________________________________
> R-package-devel using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-package-devel
>

	[[alternative HTML version deleted]]



More information about the R-package-devel mailing list