writing a simple package in R 2.0 under Windows XP
John Fox <jfox <at> mcmaster.ca> writes:
Building Simple Packages Under R for Windows
2. Make sure R is *not* installed under c:\Program Files (or in any location with spaces in the path) and that it is installed with package-building tools. I use c:\R for the installation; I'll assume this below -- make changes as necessary.
I just have been through building packages with XP this weekend. I used the default R installation which installs R 2.0.1 beta under C:\Program Files\R\rw2001beta and had no problems with using that pathname so I believe this warning is no longer needed, at least with XP and R 2.0.1. That would slightly simplify installation since it allows one to use the default installation. Uwe pointed out to me that one should use forward slashes rather than backward slashes on path names when using R CMD ... . I think you have avoided the need for pathnames in your script altogether but thought I would mention it since it was a significant gotcha for me.
5. Download fptex from <http://www.ctan.org/tex-archive/systems/win32/fptex/current/TeXSetup.exe> and install it. Note: This requires a fast internet connection during the installation. Don't install latex under c:\Program Files\!
I used MiKTeX which is a pain since there is some configuration to be done to use it with R although it works well once MiKTeX is so configured. The murdoch link you posted describes the configuration steps. Someone else also mentioned to me privately I should have tried fptex and would have avoided these configuration problems. (It occurred to me that if the Rd.sty file could be added to the MiKTeX repository online where MiKTeX gets its other .sty files then it could be pulled out from there and the MiKTeX configuration would not have to be modified to access the R directory tree.)
..\..\bin\R CMD build --force --binary --auto-zip %1 ..\..\bin\R CMD build --force %1 ..\..\bin\R CMD check %1
It was also pointed out to me that --force is an important switch; however, I did not use it and did not notice any problems. I also did not use --auto-zip. Any comments on the importance of these? Andy advised me to first use: R CMD install ... to get my software going and that advice seemed particularly useful since it allowed me to install, run and test my software as a package in R so that I could get it in better shape before submitting it to R CMD check ... and R CMD build ... --binary .