Thanks very much.
Ouch, well clearly it would have been a help had I known to do that.
Got it and you are correct: whipping long quiescent neurons into action
finally sorted this out. It's edge or corner case stuff here I guess
but may be useful to have in the archives for someone else some day.
What I had forgotten was that I had used the pkgdown package to invoke
the gh-pages branch (done automatically I think which contributed to my
forgetting it). That, with a CNAME entry in my DNS four years back, maps
cpsyctc.github.io to https://cecpfuns.psyctc.org/ (the URL is given to
github via the first line of _pkgdown.yml). That means that github then
magically does the pkgdown thing updating the gh-pages branch and that
is then accessible at https://cecpfuns.psyctc.org/. As far as I can
see, the ahead:behind count for the gh-pages branch in github is pretty
meaningless here. Someone please correct me if I am wrong about that.
What threw me off was another bit of forgetting: though I was updating
the package version number in NEWS.md, I was forgetting to update it in
DESCRIPTION so when I looked at https://cecpfuns.psyctc.org/ I saw the
old version number and thought the process wasn't working.
Solved! I have added some notes, largely to myself to make up for
crumbling memory, to my Rblog (https://www.psyctc.org/Rblog/) at
https://www.psyctc.org/Rblog/posts/2021-02-10-making-my-first-usable-package/#using-github-pages-and-pkgdown-to-create-a-page-about-the-package
Apologies for bandwidth and thanks again for the list,
Chris
On 06/08/2025 17:00, Michael Chirico wrote:
Hey Chris,
Have a look at the actual list of commits in your gh-pages vs. main
branches, I suspect you'll spot what's going on pretty quickly :)
https://github.com/cpsyctc/CECPfuns/commits/gh-pages/
vs
https://github.com/cpsyctc/CECPfuns/commits/main/
Mike C
On Wed, Aug 6, 2025 at 8:56 AM Chris Evans <chrish...@psyctc.org> wrote:
Sorry for a very trivial issue at one edge of R package
management: use
of github.
My package (https://github.com/cpsyctc/CECPfuns) is mostly very
simple
as I am entirely self-taught as a statistics user and R programmer
and
as I am ageing! (I was a fairly good psychiatrist and
psychotherapist
and still am a pretty productive researcher I think!)
The package is mostly aimed at people like me who find R amazingly
helpful but who sometimes struggle with the documentation,
warnings and
error messages. Hence a lot of it is trivial but is trying to wrap
things up to give end users reasonable help, the odd vignette (not
enough currently) and good input sanity checking and non-geek worded
warnings and errors.
OK enough of the apologies. I have always found git and github in
the
"dark arts" side of things but I thought that pushing my commits
(from
within a package worked on locally in Rstudio) was working for
updating
the package and mostly I thought the evidence was that it was.
Clearly
I wasn't checking well enough.
At some point I clearly created a gh-pages branch and I think that
was
to get github to create the user friendly html page(s) about the
package. Reading the web now and trying to separate up to date
from old
advice from out of date advice I think this is no longer necessary
and I
could get the html from the main branch and ditch the gh-pages branch.
However, as the subject line says, I now see that the two branches
are
badly out of synch. Github says that main is my default branch and
gh-pages an active branch and at the moment both say they were last
updated six hours ago which was when I last pushed changes up.
However,
gh-pages says its behind:ahead count is 100:31 which sounds nasty
and I
don't think my use of pkgdown and things in /doc and /docs are
updating
correctly.
Having scoured the web for help on this I am rattled and reaching out
for hel. As I say above, I think using gh-pages is no longer
necessary
and I would like to synchronise the two branches if that is possible
then delete the gh-pages branch and then get my courage up and try to
get back to understanding how best to document things with github
pages
and with pkgdown.
I know this is very small beer against the fascinating things I read
here but I hope someone can look at my mess and point me to safe ways
forward.
Thanks in advance and more generally, thanks for the list which is,
though often way beyond what I need, always educational for me.
Chris
--
Chris Evans (he/him)
Visiting Professor, UDLA, Quito, Ecuador
CORE site: http://www.coresystemtrust.org.uk
<https://www.coresystemtrust.org.uk/>
Other work web site: https://www.psyctc.org/psyctc/
Personal site: https://www.psyctc.org/pelerinage2016/
______________________________________________
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel
--
Chris Evans (he/him)
Visiting Professor, UDLA, Quito, Ecuador
CORE site: http://www.coresystemtrust.org.uk
<https://www.coresystemtrust.org.uk/>
Other work web site: https://www.psyctc.org/psyctc/
Personal site: https://www.psyctc.org/pelerinage2016/
<https://ombook.psyctc.org/book>
______________________________________________
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel
