[R] Loading a package shared lib works in 2.12 but not 2.11

Uwe Ligges ligges at statistik.tu-dortmund.de
Wed Apr 20 10:45:19 CEST 2011

On 13.04.2011 17:32, Von Der Hirschheydt, Juergen wrote:
> Hi,
> I have a problem with an external library on a previous R version.
> We've created our own package containing a mixture of C++ as well as R
> code which works fine under R 2.12.1. However, trying to install the
> very same package ZIP file on R 2.11.1 will issue an error when loading
> a library:

A zip file is a binary package that has been compiled for an almost 
fixed triple (R version, OS, hardware), where OS is Windows in 32-bit, 
64-bit or (exclusive) both of them in this case.

Therefore, you have to download a zip that was made for R-2.11.x which 
is available from yourCRANmirror/bin/windows/contrib/2.11 or more easily 
just via install.packages() that picks uop the correct repository. 
Alternatively, install from sources - and read the R Installation and 
Administration manual.

Uwe Ligges

>    >  utils:::menuInstallLocal()
>    package 'quasar' successfully unpacked and MD5 sums checked
>    >  require(quasar)
>    Loading required package: quasar
>    Error in library.dynam(lib, package, package.lib) :
>      shared library 'quasar' not found
>    In addition: Warning message:
>    package 'quasar' was built under R version 2.12.1
> (quasar being the name of the package). Does anyone have some pointers
> why it works under 2.12.1 but not 2.11.1 ? I've looked over the
> changelog but couldn't find any clue there.
> Has the mechanism for "useDynLib(quasar)" in the NAMESPACE file changed
> ?
> Thanks for your time,
> Cheers,
> Juergen
> ===============================================================================
> Please access the attached hyperlink for an important el...{{dropped:4}}
> ______________________________________________
> R-help at 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.

More information about the R-help mailing list