Generally the guidelines are at
https://bioconductor.org/developers/how-to/git/
and you'll access 'devel' as the master branch and release as the
RELEASE_3_10 branch of
git clone git at git.bioconductor.org:packages/netDx
'We' do not offer a faster way to iterate on the nightly builds; the
traditional approach, usually 'good enough' for most packages, is to
maintain an installation of the relevant R / Bioconductor platforms, e.g.,
R-devel / Bioc 3-11 for 'devel and R-3.6.2 / Bioc 3-10 for release for the
current cycle, following instructions and links at
https://bioconductor.org/developers/how-to/useDevel/
You could also try managing the two versions using the
Bioconductor/Bioconductor_docker image
http://bioconductor.org/help/docker/
and Travis etc images (I don't have expertise on this, maybe someone else
on the list can provide detail) but my overall experience is that these
layers introduce the risk that one ends up debugging the tool rather than
the package.
Regardless, the basic process is to make and commit changes locally, then
on a clean local checkout build and check
cd /tmp
git clone path/to/updated/repo
R CMD build repo
R CMD check repo_x.y.z.tar.gz
or similar, depending on your development system.
If there are obscure errors then one might visit
http://bioconductor.org/developers/how-to/troubleshoot-build-report/
for inspiration.
Martin
From: Shraddha Pai <shraddhapai.neuro at gmail.com>
Reply-To: "shraddha.pai at utoronto.ca" <shraddha.pai at utoronto.ca>
Date: Monday, February 24, 2020 at 10:31 PM
To: Martin Morgan <mtmorgan.bioc at gmail.com>
Cc: Bioc-devel <bioc-devel at r-project.org>
Subject: Re: [Bioc-devel] Deleting netDx.examples package from BioC
Hi Martin,
Great, thank you. I will set the number of cores to one for the examples
(must have had something running in parallel).
I was unable to find the BioC github repo for netDx in the devel branch -
what is the URL for that? It would help me ascertain if my push indeed
synchronized with BioC github repo.
Also is there another build process I could use to get quicker feedback on
issues, so I can resolve them sooner? If I understand correctly, waiting
for the nightly build means having the turnaround time of a day to know
whether or not a fix worked.
Best,Shraddha
On Sat, Feb 22, 2020 at 7:49 AM Martin Morgan <mailto:
mtmorgan.bioc at gmail.com> wrote:
We've removed netDx.examples from the build manifest and from our git
repository; these changes will propagate through the system over the next
couple of days. Thanks for pointing this out!
The builds are nightly, and summarized from this page
http://bioconductor.org/checkResults/
This is the report for your package
http://bioconductor.org/checkResults/3.11/bioc-LATEST/netDx/tokay2-checksrc.html
Note the 'snapshot' date (the time the nightly builds started) and the
time the page was generated (after the completion of nightly builds of all
packages). Any commit after the snapshot date won't be included in the
build report.
Looks like there is a (obscure!) problem on the Linux builder -- parallel
evaluation is often challenging to make robust, because the builder is
heavily used, and a better strategy is usually to tackle more modest
analyses in examples.
Martin
?On 2/21/20, 2:43 PM, "Bioc-devel on behalf of Shraddha Pai" <mailto:
bioc-devel-bounces at r-project.org on behalf of mailto:
shraddhapai.neuro at gmail.com> wrote:
Hello BioC team,
My package netDx was recently accepted to BioC (yay!). Its original
companion package netDx.examples was, as well. However, the latter is
now
outdated because I'm pulling the data for the examples directly from
BioConductor.
How can I delete the netDx.examples package from BioConductor?
https://bioconductor.org/packages/devel/data/experiment/html/netDx.examples.html
Separately, I am also refreshing my memory about how feedback from the
build process works after a package is submitted, so I may have
questions
after reading the instructions...
Thank you,
Shraddha
--
--
Shraddha Pai, PhD
http://shraddhapai.com ; @spaiglass on Twitter
Post-doctoral Fellow, http://baderlab.org
The Donnelly Centre for Cellular and Biomolecular Research
University of Toronto
[[alternative HTML version deleted]]