[pkg-bioc] Status report on the BioC packages

[Johannes Ranke]

Hi Stefan,

* Steffen Moeller <steffen_moeller at gmx.de> [070219 12:40]:
> Hi all,
> 
> the cran2deb does not perform an automated installation of packages. With a 
> regular installation of CRAN and BioC packages though the process works 
> fairly smoothly.

Sounds like you really pushed us a great step forward!
 
> The download script was modified to have virtually all the of the 1.9 bioc 
> release downloaded now. This comprises the data/annotation and the 
> data/experiment directories, summing up to 3.6GB of upstream source.
> Some data is rather ... er ... huge. WIth 7 Gig free I was rather surprised to 
> experience a disc full error for a single package build, not so much though 
> after seeing the GGdata's orig.tar.gz to occupy 235 MB already - SNP data, so 
> I presume. I kicked that one out *=|:o/
> 
> Of the 1142 libraries in bioc there are now 1062 compiled as Debian packages. 

Wow.

> At the moment I am going manually through the remaining once to decide if 
> they should be in at all and if so, what prevented them from building. The 
> cran2deb script needs some further automation for the addition of packages 
> and a resource-optimising build order. I think I also speak for Dirk when I 
> say that prior to getting that fixed, there is no way for us to step into the 
> limelight. Another issue is the integration of multiple platforms. Are there 
> any PowerPC and Sparc users among us? Are there any visionary thoughts on how 
> to present the 3.6 Gigabyte + CRAN Debian packages to anybody? On Alioth?

As far as CRAN packages are concerned, one could argue that the natural
place for them to appear would be CRAN, that is the bin/linux/debian
directory which I am currently maintaining. However, this is split in
two sections (sarge and etch) at the moment. I could build the packages
for etch amd64 and i386, as soon as it is stable. It would have to be
investigated how much disk space this would take and if CRAN maintainers
would be prepared to cope with such a sudden increase of space to be
mirrored. But I don't really know if CRAN package autobuilding is in as
good a state as BIOC.

Greetings,

Johannes

> 
> Many greetings
> 
> Steffen



-- 
Dr. Johannes Ranke                 jranke at uni-bremen.de     Key ID: F649AF90
UFT Bremen, Leobenerstr. 1         +49 421 218 8971
D-28359 Bremen                     http://www.uft.uni-bremen.de/chemie/ranke