Re: [Rd] --as-cran and need to ignore.svn directories

From: Paul Johnson <>
Date: Mon, 12 Mar 2012 13:43:06 -0500

On Mon, Mar 12, 2012 at 12:47 PM, Martin Maechler <> wrote:
>>>>>> Jari Oksanen <>
>    > Paul,
>    > I think the best solution is to 'svn export' svn directory to a
>    > temporary directory/folder:
> No, I don't think that should not be needed.
> When building the tarball,  'R CMD build' *does* ignore the .svn
> folders, at least for me.

My tarball was created by R CMD build, I don't know why the .svn folders were included. There are no .svn folders in the packages I made with 2.14.1, so I suppose that is the first problem I have to figure out. This was my first build with R-2.14.2. If I get an answer, I will write back.

> Paul's problem was using  --as-cran on his SVN *directory*
> but that's a slight contradiction in itself, as the CRAN maintainers
> always run the checks on a tarball...

I don't have an argument against what you say, and I'm glad to do it however you recommend. But I don't understand this. How do you get the tarball without first doing R CMD check and R CMD build on the working directory where the package lives? I just do what seems to be recommended in the tutorial by Friedrich Leisch "Creating R Packages: A Tutorial" and the R extensions document. Until now, I have done this in the shell to create the package.

R CMD check rockchalk
R CMD build rockchalk

You mean everybody doesn't do it that way?

>From now on, I'm adding --as-cran with check, but what else?

After that finishes, I should be running check on the tarball that was created by build?

While I'm asking basic packaging questions, I have a vignette packaging question. My vignette is in


The R build process runs latex to make sure the vignette can be built. But, as far as I can tell, it does not copy the pdf file into


So when building the package, I should still manually compile the pdf and move it over there?

> Martin
>    > svn export my-svn-directory tmp-pkg-directory
>    > R CMD build tmp-pkg-directory
>    > R CMD check --as-cran ...
>    > The two advantages of 'svn export' that it (1) strips the .svn specific
>    > files, and (2) it really exports only those files that really are under
>    > version control. More often than once I have had some non-svn files in
>    > my svn directory so that *my* version of the package works, but the one
>    > actually in subversion fails.
>    > Cheers, Jari Oksanen
>    > --
>    > Jari Oksanen, Dept Biology, Univ Oulu, 90014 Finland
>    >, Ph. +358 400 408593,

Paul E. Johnson
Professor, Political Science    Assoc. Director
1541 Lilac Lane, Room 504     Center for Research Methods
University of Kansas               University of Kansas  

______________________________________________ mailing list
Received on Mon 12 Mar 2012 - 18:45:16 GMT

This quarter's messages: by month, or sorted: [ by date ] [ by thread ] [ by subject ] [ by author ]

All messages

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Tue 13 Mar 2012 - 17:00:29 GMT.

Mailing list information is available at Please read the posting guide before posting to the list.

list of date sections of archive