[Rd] RE: [R] Building package on Linux and Windows

Liaw, Andy andy_liaw at merck.com
Fri Nov 19 15:23:02 CET 2004


Questions about packaging are better suited for R-devel (so I'm redirecting

My understanding is that R CMD build mypkg, w/o other flags, does a bit more
than tar/gzip:  it skips junk files (e.g., .o, .a, .so, etc.).  Also, the
--force flag will update the INDEX file.

On Windows, you'll probably want R CMD build --binary mypkg for building
binary package.  What this does is install the package in a temporary
location, then zip that installed package.  The installation step is
necessary, at least because the help files need to be generated from the Rd


> From:  michael watson (IAH-C)
> Hi
> OK, final stretch now, thank you to everyone who has helped!
> (hopefully) final question - what does R CMD build actually do that
> tar/gzip and WinZip do not?
> I have successfully used R CMD build to create a .tar.gz version of my
> package, which installs well and works under Linux - hurray!  
> I naively
> thought that all I had to do now was create a .zip version (using
> something like WinZip) and it would work on Windows too - wrong!
> On Windows I get:
> Error in file(file, "r") : unable to open connection
> In addition: Warning messages: 
> 1: error -1 in extracting from zip file 
> 2: cannot open file `coXpress/DESCRIPTION' 
> I didn't run R CMD build on Windows because I don't have "suitable
> versions of Unix tools including make, sh, rm, sed, 
> awk,mkdir, echo, cp
> and cat; we have packaged a set at
> http://www.murdoch-sutherland.com/Rtools/tools.zip.".  Do I 
> really have
> to install all of these on my Windows machine to build a package?  
> Asked another way - I have a linux .tar.gz file which works, how do I
> make a windows version?
> Thanks in advance!
> Mick
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! 
> http://www.R-project.org/posting-guide.html

More information about the R-devel mailing list