[Bioc-devel] sma package moved to archived status (was Re: [BioC] failure to install: aCGH depends on unavailable package)

I've been informed by Ben Bolstad (package maintainer) that sma has been
retired/archived and so BioC developers will need to remove dependency on
sma for BioC 2.5. The following 6 packages will be directly affected:

aCGH, BiocCaseStudies, GeneSelector, limma, limmaGUI, MLInterfaces

For those looking for alternatives for particular functionality, we can open
up a discussion on the BioC-devel mailing list.


Cheers,
Patrick



Ramon Diaz-Uriarte wrote:

Dear Patrick,

Thanks a lot!


Best,

R.


On Friday 25 September 2009 19:10:03 Patrick Aboyoun wrote:

Ramon,
I am not sure what happened to the sma package on CRAN. It must have
been removed from the repository over the past week since our build
system had fresh installations last week and sma was still available. I
e-mailed the package author and CRAN maintainer to find out why this
change happened. I'll e-mail back my findings and what if any actions we
in BioC will have to take.


Patrick

Ramon Diaz-Uriarte wrote:

Dear All,

I just noticed (failures installing a bunch of BioC packages in new R
installations) that aCGH depends on "sma", but sma is no longer
available. Thus, the recommended procedure

source("http://bioconductor.org/biocLite.R")
biocLite("aCGH")
will fail. And any package which depends on aCGH (e.g., snapCGH) will
fail to install too.

I guess this is a bug in sma but, if I understand correctly, it should
not be possible for a package to fail the standard installation
procedure
for the stable release?



Best,

R.

P.D. sma used to be available in CRAN, so it is possible to get an old
version from
http://cran.r-project.org/src/contrib/Archive/sma/
install that, and then install aCGH.


### this is one example; it fails on at least two other different
systems
too.

version

? ? ? ? ? ? ? _
platform ? ? ? x86_64-pc-linux-gnu
arch ? ? ? ? ? x86_64
os ? ? ? ? ? ? linux-gnu
system ? ? ? ? x86_64, linux-gnu
status
major ? ? ? ? ?2
minor ? ? ? ? ?9.2
year ? ? ? ? ? 2009
month ? ? ? ? ?08
day ? ? ? ? ? ?24
svn rev ? ? ? ?49384
language ? ? ? R
version.string R version 2.9.2 (2009-08-24)


------------------------------------------------------------------------

MLInterfaces used sma::stat.diag.da
As the package was GPL2 I believe I can just move the relevant sources to
MLInterfaces for now; will include attribution in the man page.

However, the package does not seem to be too far from being checkable
with R 2.10, so perhaps biocore can take on the maintenance, and just distribute
via Bioconductor.  I would volunteer to do the first set of patches to
get it past
check, if we want to go that route.

On Fri, Sep 25, 2009 at 11:52 PM, Patrick Aboyoun <paboyoun at fhcrc.org> wrote:

I've been informed by Ben Bolstad (package maintainer) that sma has been
retired/archived and so BioC developers will need to remove dependency on
sma for BioC 2.5. The following 6 packages will be directly affected:

aCGH, BiocCaseStudies, GeneSelector, limma, limmaGUI, MLInterfaces

For those looking for alternatives for particular functionality, we can open
up a discussion on the BioC-devel mailing list.


Cheers,
Patrick



Ramon Diaz-Uriarte wrote:

Dear Patrick,

Thanks a lot!


Best,

R.


On Friday 25 September 2009 19:10:03 Patrick Aboyoun wrote:

Ramon,
I am not sure what happened to the sma package on CRAN. It must have
been removed from the repository over the past week since our build
system had fresh installations last week and sma was still available. I
e-mailed the package author and CRAN maintainer to find out why this
change happened. I'll e-mail back my findings and what if any actions we
in BioC will have to take.


Patrick

Ramon Diaz-Uriarte wrote:

Dear All,

I just noticed (failures installing a bunch of BioC packages in new R
installations) that aCGH depends on "sma", but sma is no longer
available. Thus, the recommended procedure

source("http://bioconductor.org/biocLite.R")
biocLite("aCGH")
will fail. And any package which depends on aCGH (e.g., snapCGH) will
fail to install too.

I guess this is a bug in sma but, if I understand correctly, it should
not be possible for a package to fail the standard installation
procedure
for the stable release?



Best,

R.

P.D. sma used to be available in CRAN, so it is possible to get an old
version from
http://cran.r-project.org/src/contrib/Archive/sma/
install that, and then install aCGH.


### this is one example; it fails on at least two other different
systems
too.

version

              _
platform       x86_64-pc-linux-gnu
arch           x86_64
os             linux-gnu
system         x86_64, linux-gnu
status
major          2
minor          9.2
year           2009
month          08
day            24
svn rev        49384
language       R
version.string R version 2.9.2 (2009-08-24)


------------------------------------------------------------------------