>>>>> On Sun, 27 Aug 2006 07:45:00 -0700, >>>>> Martin Morgan (MM) wrote:
> [A build system misconfiguration at Bioconductor lead to a discussion > about whether \usepackage{Sweave} should be included in .Rnw > documents. Leaving this line out causes a problem on Windows, as > indicated below.] > I think the problem is '~', which should, according to Donald > Arseneau's comment in this thread > http://groups.google.com/group/comp.text.tex/browse_thread/thread/640e89078d0697c4/5b6f9fc6b0da21bf?lnk=gst&q=windows+file+path&rnum=4#5b6f9fc6b0da21bf > be \string~. If this is the correct change, perhaps an upstream fix > could be introduced in src/library/util/R/Sweave.R. Sorry to join in late, I am currently at a conference with limited access to the Internet. I am not su sure how easy it is to make the conversion, i.e. if it is save to unconditionally change all tildes in the path as suggested above. Does anybody have any experience wether that is save on all platforms and TeX installations? Personally I think it would make a lot of sense if the Bioconductor check machines could include a version of RHOME/share/texmf in their TeX input path ... that would also allow to use Rd.sty etc. Otherwise most of what has been said in this thread is true: The full path is inserted into the document by Sweave such that you donnot have to copy something from RHOME to your tex installation every time you install a new version of R. On Unix it is easy: Put /usr/local/lib/R/share/texmf (or wherever you typically install your default R) into your tex path and you are done. On windows the normal R installation path contains the version number, so this will fail. Hence I did the path thing in R as a convenience for authors (and to make sure that R version and Sweave.sty are consistent, but the .sty is pretty stable anyway). Maybe the installer could/should create c:\programme\R\R-latest as a link to the newest installed version? Just a thought. Best, Fritz ______________________________________________ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel