On 13-03-22 4:40 PM, Berend Hasselman wrote:

On 22-03-2013, at 21:26, Dirk Eddelbuettel <e...@debian.org> wrote:


On 22 March 2013 at 21:03, Berend Hasselman wrote:
| I am running R CMD check on my package nleqslv with R-3.0.0 beta (2013-03-33 
r62364) on Mac OS X 10.6.8
|
| In contrast with R-2.15.3 R CMD check now issues a note for
|
| - Mercurial version control files .hgignore, .hgtags and directory .hg. These 
are however included in .hidden_file_exclusions.
| - typical Mac Finder files .DS_Store
|
| All of these are ignored when executing R CMD check --as-cran.
|
| I can't remove the .hg* files and directories. I could remove the .DS_Store 
files to avoid the note.

A few years ago the "Right Way To Do Things" changed.  You are really
supposed to do 'R CMD check' off the tar.gz, and you *do* have a toggle to
exclude what goes into the tar.gz:   .Rbuildignore


which I have. And making a .tar.gz will ignore the version control files. So 
why not ignore them when doing a check on a (allowed) package directory?

Because that makes the code messier. A directory is treated as if it was just extracted from the tarball. If the code had to distinguish two different ways of getting to that state it would be harder to maintain.

Duncan Murdoch





 From  R CMD check --help
Usage: R CMD check [options] pkgs

Check R packages from package sources, which can be directories or
package 'tar' archives with extension '.tar.gz', '.tar.bz2',
'.tar.xz' or '.tgz'.

So a directory is allowed.

| The note is harmless but annoying and strikes me as slightly overzealous.

So are are many other warnings.


It's a note and not a warning.

But in the long run I am happy they are there so that eg my system does not
get polluted with your .hg and .DS_Store files which are of zero use to me.


My .hg and .DS_Store files will not pollute your system.

Berend

Warnings are your friends. Be nice to them.

Dirk


--
Dirk Eddelbuettel | e...@debian.org | http://dirk.eddelbuettel.com

______________________________________________
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel


______________________________________________
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel

Reply via email to