[Bioc-devel] questions about windows build execute permissions and R version dependency

[Alicia Schep]

Thanks for the clarification! I will ignore these warnings.

-Alicia


On Thu, Mar 23, 2017 at 7:36 AM Leonardo Collado Torres <lcollado at jhu.edu>
wrote:

You can also test your package on R Travis against R 3.4 as shown at
https://travis-ci.org/leekgroup/recount#L518. Just use this on your
.travis.yml file
https://github.com/leekgroup/recount/blob/master/.travis.yml#L3

On Thu, Mar 23, 2017 at 4:26 AM, Martin Morgan
<martin.morgan at roswellpark.org> wrote:
> On 03/22/2017 11:08 PM, Alicia Schep wrote:
>>
>> Hi,
>>
>> I have a two questions about the bioconductor build and BiocCheck process
>> when submitting packages.
>>
>> 1) My submitted package ('motifmatchr') returns the following warning on
>> the Windows computer:
>>
>> Warning: file 'motifmatchr/cleanup' did not have execute permissions:
>> corrected
>>
>> How can this warning be averted?  How to give the file permission? I
tried
>> fixing this issue by making the file executable on my computer-- on
GitHub
>> it shows up as marked as "Executable File". This did not prevent the
>> warning though.
>
>
> It seems like you've done the right thing and R is mistaken. It's also
> irrelevant, since you have a cleanup.win file that supersedes cleanup.
> Ignore the warning.
>
>> 2) The same package also gives a warning in BiocCheck related to the
>> version of R required in DESCRIPTION.  As the package is currently hosted
>> on GitHub and tested against the current release of R using travis, I
>> would
>> like the github version to have a dependency of only R 3.3 rather than R
>> 3.4. Is it possible to have a different dependency for the R version on
>> GitHub versus on Bioconductor?
>
>
> Certainly anything is possible, e.g., creating a release branch on github
> with a different dependency, but again I'd suggest evaluating whether the
> warning is relevant in your situation, and if not ignoring it.
>
> From the Bioconductor perspective, the fact that your package builds on
> travis under an earlier version of R is unknown, so the assertion that it
> only requires R-3.3 is not supportable -- the warning tries to bring your
> version requirements into line with what we know to be true. Of course the
> assertion that it depends on a version of R > 3.4 is also not supportable.
>
> Martin
>
>>
>> Appreciate any guidance on these question.  Thanks,
>>
>> Alicia
>>
>>         [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> Bioc-devel at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>>
>
>
> This email message may contain legally privileged and/or...{{dropped:2}}
>
>
> _______________________________________________
> Bioc-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel

	[[alternative HTML version deleted]]