Thanks to all for your replies. So far as I can see, there was
nothing wrong with my original approach, but I've decided to stuff all
the relevant definitions into a function (or functions), as this seems
to make "devtools::check()" happier.
-- Mike
On Fri, Jul 13, 2018 at 6:54 PM, Jeff Newmill
Avoiding rda files because they don't track well with version control seems
weak to me, since you should be creating the rda with an R file in the tools
directory.
On July 13, 2018 6:50:31 PM PDT, William Dunlap wrote:
>What the OP is doing looks fine to me.
>
>The environment holding the data
What the OP is doing looks fine to me.
The environment holding the data vectors is not necessary, but it helps
organize things - you know where to look for this sort of data vector.
I would avoid the *.rda file, since it is not text, hence not readily
editable
or trackable with most source contro
a) There is a mailing list for package development questions: R-package-devel.
b) This seems like a job for the sysdata.rda file... no explicit environments
needed. See the Writing R Extensions manual.
On July 13, 2018 5:51:06 PM PDT, Michael Hannon
wrote:
>Greetings. I'm putting together a s
I would usually use a function for this. It may not be more R like, but it is
more readable to me. If you want, to keep the columns in a file, you could have
the function initialize itself on the first call.
Mark
R. Mark Sharp, Ph.D.
Data Scientist and Biomedical Statistical Consultant
7526 Mea
5 matches
Mail list logo
