[Bioc-devel] BiocUpgrades updates to devel instead of stable

[Julian Gehring]

Related to this, running 'biocLite("BiocUpgrade")' with an bioc-3.0 
triggers an upgrade to bioc-3.1 which is non-existing.

Best wishes
Julian


On 22.04.2014 14:38, Julian Gehring wrote:
> Hi,
>
> Alejandro and I digged deeper into this:
>
> The bug I described before as caused by 'BiocInstaller:::.onLoad' which
> sets 'BiocInstaller:::IS_UPGRADEABLE' to TRUE for every bioc release
> with an even minor (e.g. 2.12, 2.14).  Within the load function,
>
>    (BIOC_VERSION$minor %% 2L) == 0L
>
> checks against the minor of the bioc number, which looks
> non-informative.  Rather, checking if the minor of the 'BiocInstaller'
> package is even or odd could be more useful here
>
>    (packageVersion("BiocInstaller")$minor %% 2L) == 0L
>
> which is TRUE for stable and FALSE for devel.
>
> Best wishes
> Julian
>
>
> On 17.04.2014 18:50, Steve Lianoglou wrote:
>> HI,
>>
>> On Thu, Apr 17, 2014 at 7:09 AM, Julian Gehring
>> <julian.gehring at embl.de> wrote:
>>> Hi,
>>>
>>> When I run
>>>
>>>    source("http://bioconductor.org/biocLite.R")
>>>    biocLite("BiocUpgrade")
>>>
>>> on a stable R-3.1.0 with bioc 2.14 already installed, bioc wants to
>>> upgrade to bioc 3.0 (devel):
>>>
>>>    Bioconductor version 2.14 (BiocInstaller 1.14.1), ?biocLite for help
>>>    Upgrade all packages to Bioconductor version 3.0? [y/n]
>>>
>>> Is this intentional?  The documentation on the website and ?BiocUpgrade
>>> suggests that it upgrades to the latest bioc 'release'.
>>
>> Is this some potential weirdness with ?BiocInstaller::useDevel maybe?
>>
>> -steve
>>
>
> _______________________________________________
> Bioc-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel