[R-pkg-devel] Inconsistent R CMD Check results

[Tomas Kalibera]

On 1/23/23 17:51, Ying Li via R-package-devel wrote:
> Dear all,
>
> Hope you are well! Recently, in the check before a re-submission, I got an unexpected note when doing R CMD Check using rhub::check_for_cran(), saying that "Examples with CPU (user + system) or elapsed time > 5s".
>
> I didn't expect this note to appear because the same examples didn't cause this note in the previous submission last December. Compared with the previous submission, we only edited documentation, there was no change in the R code for functions of our package.
>
> To double-check, I downloaded our previous submission from CRAN and checked it again. It's strange that this current CRAN version also has this note now. However, when submitting it last December, there were no such notes. Do you have any idea why the R CMD Check results are inconsistent?

You could check on Winbuilder as that is a concrete setup used for 
incoming checks. But in principle, the computation may take longer 
because of changes in dependencies, external software and R. Which can 
be due to a performance regression, a bugfix, etc. Only 
debugging/profiling could reveal the true cause.

> Since the examples caused no note in the last submission and were successfully submitted to CRAN, I'm not sure whether I should change the examples in this re-submission to eliminate this sudden note. Any suggestions will be appreciated!
If you were close to the limit last time, it may be a coincidence and 
you might simply reduce the examples to be on the safe side.

Tomas

>
> Thank you!
> Ying
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-package-devel using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-package-devel