CRAN issues July 2024

[chainsawriot]

Dear maintainer,
Please see the problems shown on
https://cran.r-project.org/web/checks/check_results_rio.html.
Please correct before 2024-08-18 to safely retain your package on CRAN.
Best,
-k

I think I can safely ignore the arrow one on r-devel-linux-x86_64-debian-clang because it's due to the installation issue of arrow.

But the second one is a mystery. I think I didn't see it yesterday. Is it transient?

[chainsawriot]

Testing Fedora GCC https://github.com/r-hub2/educated-yogic-starfish-rio/actions/runs/10175030046

[chainsawriot]

Cannot reproduce the same problem on rhub2. Let's wait one day to see if that's transient CRAN problems.

[TimTaylor]

I can reproduce on current R-devel. Deleteing the following in .Rbuildignore will make it go away

rio/.Rbuildignore

Line 24 in 44439eb

^inst/doc/.+\.Rmd$

The CRAN commit that causes the issue is wch/r-source@a053afc.

Unsure whether this will change but may be worth tightening up the .Rbuildignore file anyway.

[chainsawriot]

@TimTaylor Thank you very much for the pointer! Super helpful information. I will amend as you suggested and perhaps test it with an offline Fedora with r-devel.

[TimTaylor]

No worries. I wonder if this could catch out a few maturer packages that have accumulated bits in .Rbuildignore. Reading item 20 in https://cran.r-project.org/doc/manuals/R-exts.html#Checking-packages we have (emphasis mine):

... (If the vignette sources are in the deprecated location inst/doc, do mark such target output files to not be installed in .Rinstignore.)

So I guess historically (pre-dates my R usage), the vignette source was often kept there and hence rio's .Rbuildignore.

Hope this isn't teaching you to suck eggs so to speak. I just wanted to note it in this issue for my own reference.