[R-pkg-devel] advice on a rejected package (I think, because time was 13min, in excess of 10min limit)

[Uwe Ligges]

Actually, oce is one of the very few packages where you cannot do much 
to reduce this:

* checking R code for possible problems ... [177s] OK
* checking examples ... [128s] OK
* checking tests ... [70s] OK
* checking re-building of vignette outputs ... [52s] OK

This shows the runtime checks only consume 4 minutes, but the package is 
rather complex and needs a lot of time in the other parts of the checks.

Therefore reverse dependency checks have been triggered on CRAN.

Best,
Uwe Ligges




On 12.05.2022 11:34, Tomas Kalibera wrote:
> 
> On 5/12/22 10:55, Daniel Kelley wrote:
>>
>>
>>> On May 12, 2022, at 4:34 AM, Tomas Kalibera
>>> <tomas.kalibera using gmail.com> wrote:
>>>
>>> CAUTION: The Sender of this email is not from within Dalhousie.
>>>
>>> On 5/11/22 22:26, Daniel Kelley wrote:
>>>> Thanks, Dirk. On my machine, the test suite takes 85 seconds.
>>>> Assuming the CRAN machines are about half as fast, eliminating all
>>>> the tests might get oce under the 10 minute limit.
>>>
>>> Out of curiosity, how did you find out it is because of a time limit? In
>>> the outputs shown in your original post I only see a failed test on
>>> Linux. If it takes only 85 seconds on your machine, using reasonable
>>> amount of memory, but several times longer elsewhere, I would recommend
>>> first looking into the cause of the slowdown.
>>
>> Tomas, thanks for your comment.  Perhaps I ought to provide a bit more
>> information.
>>
>> I found out the time near the end of the email, where it said
>>
>> Flavor: r-devel-windows-x86_64
>> Check: Overall checktime, Result: NOTE
>> Overall checktime 13 min > 10 min
>>
>> The 85 seconds to which I refer is on my local machine (a laptop) and
>> it is only for the tests part. I don't think I saw time broken-down in
>> the links provided in the email I received (pasted below ... but note
>> that none of these links works for me today, although they worked 15
>> hours ago).
> 
> I see. On my check server for ucrt3, 1.7-2 also took ~13 minutes to
> check, so I assume this is reproducible and you should be able to debug
> it elsewhere if you arrange to get some timings. Or maybe you could
> deduce something from my timestamps:
> 
> May 11 04:12 00_pkg_src
> May 11 04:16 oce
> May 11 04:16 00install.out
> May 11 04:20 oce-Ex.R
> May 11 04:21 oce-Ex.Rout
> May 11 04:21 oce-Ex.pdf
> May 11 04:21 tests
> May 11 04:25 oce-manual.pdf
> May 11 04:25 Rdlatex.log
> May 11 04:25 oce-manual.log
> May 11 04:25 00check.log
> 
> and for tests:
> 
> May 11 04:21 testthat.R
> May 11 04:21 startup.Rs
> May 11 04:23 testthat
> May 11 04:23 testthat.Rout
> 
>>
>> Importantly, I am quite curious about your reference to a failed linux
>> test.  I don't recall seeing any errors for my new version (1.7-3) on
>> the following links (from the original email I received from CRAN).  I
>> did, however, see errors for the previous version (1.7-2); indeed
>> these were the reason why I was told to make a new CRAN release.
> 
> Sorry, I read the log incorrectly, that was for 1.7-2.
> 
> Best
> Tomas
> 
>>
>> Windows:
>> <https://win-builder.r-project.org/incoming_pretest/oce_1.7-3_20220511_192652/Windows/00check.log>
>> Status: OK
>>
>> Debian:
>> <https://win-builder.r-project.org/incoming_pretest/oce_1.7-3_20220511_192652/Debian/00check.log>
>> Status: OK
>>
>> More details are given in the directory:
>> <https://win-builder.r-project.org/incoming_pretest/oce_1.7-3_20220511_192652/>
>>
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-package-devel using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-package-devel