Duncan wrote
> You can override this by setting the R_LIBS_USER variable in Cygwin to
> C:/Users/aldi/Documents/R/win-library/2.15 before you start R; then
> things should be fine.
If you use R_LIBS_USER then you have to be careful when later calling .libPaths
to add more libraries:
% env R_LIBS_USER=/homes/bill/packages/Rlib R --quiet
> .libPaths()
[1] "/homes/bill/packages/Rlib"
[2] "/opt/sw/R/R-2.15.1.atlas1/lib/R/library"
> print(.libPaths("/tmp")) # loses library specified by R_LIBS_USER
[1] "/tmp"
[2] "/opt/sw/R/R-2.15.1.atlas1/lib/R/library"
If you use R_LIBS_SITE instead of R_LIBS_USER then your chosen library
sticks around (it is saved in base::.Library.site, which is used by .libPaths())
% env R_LIBS_SITE=/homes/bill/packages/Rlib R --quiet
> .libPaths()
[1] "/homes/bill/packages/Rlib"
[2] "/opt/sw/R/R-2.15.1.atlas1/lib/R/library"
> print(.libPaths("/tmp")) # keeps library specified by R_LIBS_SITE
[1] "/tmp"
[2] "/homes/bill/packages/Rlib"
[3] "/opt/sw/R/R-2.15.1.atlas1/lib/R/library"
The difference between R_LIBS_USER and R_LIBS_SITE seems odd to me.
Bill Dunlap
Spotfire, TIBCO Software
wdunlap tibco.com
> -----Original Message-----
> From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On
> Behalf
> Of Duncan Murdoch
> Sent: Tuesday, October 16, 2012 12:39 PM
> To: Aldi Kraja
> Cc: r-help@r-project.org; Bert Gunter
> Subject: Re: [R] Windows 7 R (32/64bit) running under cygwin: package not
> found
>
> On 16/10/2012 3:02 PM, Aldi Kraja wrote:
> > Thank you all including Bert,
> >
> > In my opinion the library() when is working in cygwin env., does not
> > check the additional libraries, where R installed the additional packages.
> >
> > So the solution for this problem was to define manually in my programs
> > where is located the library of additional packages:
> > (although I was expecting library() itself to know about the additional
> > packages
> >
> > library('rgenoud',lib.loc='C:/Users/aldi/Documents/R/win-library/2.15')
> > This solved the problem. Now R works again in cygwin. :-)
>
> Windows defines various environment variables to set a user's home
> directory, and Cygwin sets different ones. R assumes the user library
> is in there. I forget which variable to look in, but for you, Windows
> thinks your home directory is C:/Users/aldi, and Cygwin probably thinks
> it is something like C:/cygwin/home/aldi.
>
> You can override this by setting the R_LIBS_USER variable in Cygwin to
> C:/Users/aldi/Documents/R/win-library/2.15 before you start R; then
> things should be fine.
>
> Duncan Murdoch
>
> >
> > Thanks,
> >
> > Aldi
> >
> > On 10/16/2012 1:19 PM, Bert Gunter wrote:
> > > I do not think that this has anything to do with the issue at hand,
> > > but just to clarify ...
> > >
> > > On Tue, Oct 16, 2012 at 10:18 AM, Jeff Newmiller
> > > <jdnew...@dcn.davis.ca.us> wrote:
> > >> Probably because when you run it from Cygwin the R_LIBS variable does
> > >> not point to
> the user and install library directories. I don't know how Rgui knows where
> they are
> (registry?)
> > > R for Windows need not use the registry at all, and does so optionally
> > > for just a couple of minor issues. See the R for WIndows FAQ 2.17 for
> > > details.
> > >
> > > ?library
> > > ?.libPaths
> > >
> > > describe how R sets/gets library paths (including defaults).
> > >
> > > Cheers,
> > > Bert
> > >
> > >
> > >> but you can look in the .Library and .Library.site variables to see the
> > >> results.
> > >
> > >> In a case like this, posting your sessionInfo() for each case is highly
> > >> recommended.
> > >> ---------------------------------------------------------------------------
> > >> Jeff Newmiller The ..... ..... Go
> > >> Live...
> > >> DCN:<jdnew...@dcn.davis.ca.us> Basics: ##.#. ##.#. Live
> > >> Go...
> > >> Live: OO#.. Dead: OO#.. Playing
> > >> Research Engineer (Solar/Batteries O.O#. #.O#. with
> > >> /Software/Embedded Controllers) .OO#. .OO#.
> > >> rocks...1k
> > >> ---------------------------------------------------------------------------
> > >> Sent from my phone. Please excuse my brevity.
> > >>
> > >> Aldi Kraja <a...@wustl.edu> wrote:
> > >>
> > >>> Hi,
> > >>> Using R 2.15.1 on Windows 7. Have installed both versions 32 and 64bit.
> > >>> In both of them among others I have installed a package rgenoud
> > >>> When I open R gui of 32bit and write library(rgenoud) it responds by
> > >>> showing a functional rgenoud version 5.7-8. The same it does on Rgui
> > >>> 64bit.
> > >>>
> > >>> Now I am working in cygwin (v. 1.12.4.0) with xwin. Normally before
> > >>> when
> > >>> I had installed a package, I only had to call the library with the name
> > >>>
> > >>> of the package and R will find the right one to load.
> > >>>
> > >>> Now when I apply R CMD BATCH script1.R out1.txt, under cygwin the first
> > >>>
> > >>> thing it reports:
> > >>>
> > >>> R version 2.15.1 (2012-06-22) -- "Roasted Marshmallows"
> > >>> After some generalities it reports
> > >>>> library(rgenoud)
> > >>> Error in library(rgenoud) : there is no package called 'rgenoud'
> > >>> Execution halted
> > >>>
> > >>> So my question is why under cygwin in a batch mode, it does not find
> > >>> the
> > >>> installed package, which is already installed in my laptop's R?
> > >>>
> > >>> Thank you in advance,
> > >>>
> > >>> Aldi
> > >>>
> > >>>
> > >>>
> >
> > ______________________________________________
> > R-help@r-project.org mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> > and provide commented, minimal, self-contained, reproducible code.
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
