[Rd] validspamobject?
Duncan Murdoch
murdoch@dunc@n @ending from gm@il@com
Sun Aug 19 00:22:16 CEST 2018
On 18/08/2018 3:47 PM, Ronald Barry wrote:
> Hi,
> I have submitted an updated R package to CRAN, the only problem seems to
> be the examples take too long to run on Windows and Debian. Several
> examples take over 10 seconds and the entire set takes around 110 seconds.
> However, I don't see what the 'target' amount of time is. How much will
> this need to be sped up to pass the test? Also, are there some tricks to
> speeding up examples that you have seen? For instance, is it possible to
> have the identical example in two help files but only have them run once
> (other than the obvious, which is putting a 'do not run' on one of them,
> which would seem to invite possible bad code)? Thanks for any information.
Doesn't the error message say something like "Examples with CPU or
elapsed time > 5s"? So to avoid that message, reduce the time to less
than the stated limit.
You can't share examples between different help pages, but you can
combine topics in a single page, so there's only one example to run.
The usual way to speed up examples is to make up trivial datasets for
them, rather than running them on real data.
Duncan Murdoch
>
> Ron Barry
>
> On Tue, Aug 14, 2018 at 4:56 PM, Ronald Barry <rpbarry using alaska.edu> wrote:
>
>> Greetings,
>> My R package has been showing warnings of the form:
>>
>> `validspamobject()` is deprecated. Use `validate_spam()` directly
>>
>> None of my code uses the function validspamobject, so it must be a problem
>> in another package I'm calling, possibly spam or spdep. Has this problem
>> occurred with other people? It doesn't have any deleterious effect, but
>> it's annoying. In particular, how do I determine which package is causing
>> this warning? Thanks.
>>
>> Ron B.
>>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-devel using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>
More information about the R-devel
mailing list