[Rd] Empty package skeleton
maechler at stat.math.ethz.ch
Sun Apr 14 21:11:38 CEST 2013
>>>>> Dirk Eddelbuettel <edd at debian.org>
>>>>> on Sun, 14 Apr 2013 06:51:38 -0500 writes:
> On 14 April 2013 at 14:10, Renaud Gaujoux wrote:
> | Hi,
> | I know this has been reported/asked before (
> | http://tolstoy.newcastle.edu.au/R/e15/devel/11/10/0831.html) but it would
> | still be just nice to have a fix for the not user-friendly fact that one
> | cannot create a completely empty package skeleton (see previous post for a
> | suggested patch).
> | In the same vein, I would add to the original suggestion from Herve Pages
> | that it would be nice to have a quick way to create an empty package
> | skeleton, skipping all objects possibly defined in .GlobalEnv, e.g.
> | "package.skeleton('pkg', NA)", although I believe Herve's patch would
> | automatically allow/handle a call like "package.skeleton('pkg',
> | character())".
> | Conclusion, nothing urgent, which could be done in 1min spare time, but
> | could save 1min to many friendly future package developers :)
> I am also willing to help here as I do not like the fact that the generated
> Rd file itself is broken. A hint from R Core that clean patches would be
> taken would be welcomed...
yes, such patches to package.skeleton would be welcomed.
I think we had never thought to make package.skeleton() into a
basic, fully-logic, high-standard, ..da-da-da..,.... R function
but rather as a simple way to get beginning package writers
But I have found (in teaching) as well the need for more
"logical" behavior of package.skeleton().
> [ And I often call package.skeleton() from a fresh session, that is pretty
> close to creating an empty package. ]
> Dirk Eddelbuettel | edd at debian.org | http://dirk.eddelbuettel.com
> R-devel at r-project.org mailing list
More information about the R-devel