[BioC] unable to install qvalue..

[James W. MacDonald]

Hi Simon,

On 2/21/2011 10:40 AM, Simon Anders wrote:
> Hi Vince
>
> On 02/21/2011 04:05 PM, Vincent Carey wrote:
>> At least in R 2.12, tcltk package source is in src/library of the source
>> R distribution. There is no need to distribute it at CRAN AFAICT.
>
> Yes, this explains it. Sorry for missing this obvious point.
>
>
> I didn't have the 'tk-dev' package installed on my Ubuntu when I
> compiled my R, so it was skipped, and I guess this is also the original
> poster's problem.
>
> (Not wanting to compile all of R again, I just tried to install the
> tcltk package separately from my R source tree but that does not seem
> entirely straight-forward, because the Makefile in there need to be
> regenerated from the global configure script, and simply re-running does
> not do the job. Ok, I guess rebuilding everything is not a big deal and
> easiest.)
>
>
> The remaining issue is that the installation of 'qvalue' gives this
> misleading error message:
> Error : package 'tcltk' does not have a name space
>
> Other packages, e.g., 'tckltk2', properly report:
>
> Error in firstlib(which.lib.loc, package) :
> Tcl/Tk support is not available on this system
> Error : package 'tcltk' could not be loaded
>
>
> But, I guess, the best would be if John dropped the dependency on
> 'qvalue'. I don't suppose that we should take for granted that the
> average user has Tk available. (Is it bundled to the R installer for
> Windows?)

The tcltk package comes by default with the R installer on Windows, and 
for that matter on *nix as well. It is used by several of the base 
functions like install.packages(), and IMO can be considered at least on 
the level of a recommended package.

John can do with his package as he likes, but I don't see a strong 
argument for dropping a dependency on a package that has such a high 
expectation of being installed.

Best,

Jim


>
> Simon
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor

-- 
James W. MacDonald, M.S.
Biostatistician
Douglas Lab
University of Michigan
Department of Human Genetics
5912 Buhl
1241 E. Catherine St.
Ann Arbor MI 48109-5618
734-615-7826
**********************************************************
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues