[Bioc-devel] Universal mac OS X packages for BioC 1.8 ready for testing

[Kasper Daniel Hansen]

I am unable to access the repository using the Mac OS pull-down menu  
where I choose "Bioconductor (binaries)". I know this is not the way  
you specific below, but I gues it is the most natural way for most  
users. Perhaps one can do it using a "Other repository". Perhaps this  
needs changes in R.app - in that case it should be updated for R-2.3.1

Also: I tried installing affxparser which I know does not compile on  
Mac OS X with the GCC shipped with R-2.3.0 (we have a fix I will  
merge into Bioc 1.8 today). It looked like it installed, and the .tgz  
file contains .so libs both for i386 and ppc. Both I cannot load the  
package, dyn.load complains about unresolved Symbols (which makes  
sense, the lib does not fully compile). Anyway, it looks like the  
binary have been build even though the compilation fails. That is a  
bit strange.

/Kasper

On May 19, 2006, at 10:23 AM, Herve Pages wrote:

> Hi all,
>
> Thanks to Simon Urbanek, the universal Mac OS X packages for BioC 1.8
> are finally ready for testing. As always, the easiest way to install
> is to use the biocLite() function:
>
>> source("http://bioconductor.org/biocLite.R")
>> biocLite("name_of_the_package")
>
> You _must_ use the universal R 2.3.0 built from CRAN for this to work.
> Also don't forget to pass 'type="mac.binary"' to the biocLite()  
> function
> in case you have the default type set to "source" (you can check this
> with 'getOption("pkgType")').
> You can also install a default set of Bioconductor 1.8 packages by
> simply doing:
>
>> biocLite()
>
> If you are looking for a particular package but you are not sure about
> its name or even its existence, you can browse the Bioconductor 1.8
> package repositories by going to:
>
>   http://bioconductor.org/packages/1.8/BiocViews.html
>
> When available, a link to the Mac OS X universal package (.tgz  
> file) will
> show up on the package homepages. You can use this link for individual
> package download.
>
> Please try and give us some feedback (these universal binaries are  
> still
> experimental). If things are not too bad, we plan to announce on  
> the BioC
> list next tuesday.
>
> Thanks,
>
> H.
>
> -- 
> ------------------------
> Hervé Pagès
> E-mail: hpages at fhcrc.org
> Phone: (206) 667-5791
> Fax: (206) 667-1319
>
> _______________________________________________
> Bioc-devel at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel