[R-pkg-devel] CRAN submission struggle

Tue Jan 2 13:59:13 CET 2024

On 29.12.2023 09:13, Greg Hunt wrote:
> Christaan,
> The elapsed time note is because CRAN expects that examples will be
> configured to run single threaded and some package that you use, or a
> package used by a package that you use is multi-threading by default and
> using more CPU time than clock time. If you cannot figure out how to
> reconfigure the multi-threaded package, a number of people have found that
> the simplest thing to do is disable running the example (which reduces the
> effective test coverage provided by the example).

No and no:

1.
       user system elapsed
IOPS 10.06   3.35   35.04

suggests that either the machines this ran on is heavily loaded (so that 
elapsed >> user) or something is waiting like some internet access. [For 
multithreading it would be user > elapsed.]

2. The solution is not to exclude examples ebtirely, as we need runtime 
checks.

For internet access: Set a timeout and let the exampe fail gracefully in 
case web access takes more than, e.g., 2. sec.

In general: Please reduce each example to less than 5 sec.

So use small toy examples. If you really want to add rel world examples 
that may take longer, then add them in addition to toy examples and wrap 
in \donttest{}.

Best,
Uwe Ligges

> 
> I haven’t encountered the miktex exception file before but i suspect its a
> side effect of a miktex error. Packages should not leave files behind in
> the temp directory. If you expect a miktex error you need to remove the
> file. If you don’t you need to track down and fix or work around the bug.
> The build process is really a quality check on your package.
> 
> Greg
> 
> On Fri, 29 Dec 2023 at 3:01 am, Christiaan Pieterse <
> pietie.cjp.1908 using gmail.com> wrote:
> 
>> Hi,
>>
>> Thank you for showing the difference in the ExampleTradeData. I've fixed
>> this by adding a .Gitignore file and a "data-raw" folder to load the
>> ExampleTradeData. I hope I did this correctly. When I check the package (
>> https://github.com/WoutersResearchGroup/R-IO-PS/tree/CRAN-prep) in
>> RStudio.
>> I only get 3 notes (see below), and if I run it in PositCloud, it crashes
>> or yields the same 1 ERROR and 2 NOTES result as before. Why might this be?
>> Is it a problem or is it fine if I continue working in RStudio since I
>> cannot increase the specs in PositCloud because I'm working on a research
>> group account?
>>
>> Here are the 3 notes I receive in RStudio:
>>
>> The first is the expected New Submission Note.
>>
>> The second is the runtime that is too long:
>> * checking examples ... [43s] NOTE
>> Examples with CPU (user + system) or elapsed time > 5s
>>        user system elapsed
>> IOPS 10.06   3.35   35.04
>> How can I reduce this time? I'm not sure how to reduce the size of my
>> ExampleTradeData without the check giving errors when running the example.
>>
>> The third note I am unsure what it means:
>> * checking for detritus in the temp directory ... NOTE
>> Found the following files/directories:
>>    'lastMiKTeXException'
>>
>> Kind regards
>> Christiaan
>>
>> On Thu, 28 Dec 2023 at 15:55, Ivan Krylov <krylov.r00t using gmail.com> wrote:
>>
>>> Hi Christiaan,
>>>
>>> В Thu, 28 Dec 2023 14:57:55 +0200
>>> Christiaan Pieterse <pietie.cjp.1908 using gmail.com> пишет:
>>>
>>>> Still, I couldn't figure out why I ran into this problem, so I
>>>> created a test file called "Test Example.R" (available at the same
>>>> GitHub repository:
>>>> https://github.com/WoutersResearchGroup/R-IO-PS/tree/CRAN-prep).
>>>
>>> I see you're always adding or updating files to the GitHub repo by
>>> means of uploading. While that's certainly one way to use GitHub, it's
>>> combines the least convenient aspects of two approaches to using GitHub.
>>>
>>> With GitHub purely in the browser, GitHub is just a website where you
>>> keep and edit code, running nothing else on the local computer. Code
>>> can be run in Codespaces or using GitHub Actions. Microsoft will want
>>> to be paid money to run code on their computers.
>>>
>>> With GitHub as a Git remote, there is a local checkout [*] that's kept
>>> in sync with GitHub by means of commits [**] and pushes [***], letting
>>> you create meaningful, describable snapshots of changes in your code
>>> spanning multiple files at the same time.
>>>
>>> Right now, it probably feels like Dropbox but worse.
>>>
>>>> This file creates the function in the global environment (note that
>>>> this is the same function code as available in the package
>>>> "R/iopspackage2.0.R" file), and then runs this function with the same
>>>> example as in the package (If you want to try this yourself, just
>>>> load the data/ExampleTradeData.rda in before running the Test Example
>>>> file). This test file yields no errors when I run it and produces the
>>>> correct results. When I then proceed to build and check the package,
>>>> it yields the same example error as before. I do not understand why
>>>> or what could cause this issue.
>>>
>>> The difference is in the ExampleTradeData variable, which "Test
>>> Example.R" doesn't define.
>>>
>>> With data(ExampleTradeData), the script works.
>>>
>>> With ExampleTradeData <-
>>>
>>>
>> read.csv(system.file("extdata","ExampleTradeData.csv",package="iopspackage")),
>>> the script fails exactly the same way as example(IOPS) does.
>>>
>>>> I'm not sure if I should send out another email to the developers to
>>>> see if someone else spots something I'm not seeing.
>>>
>>> It may help to keep Cc: r-package-devel using r-project.org in the e-mails
>>> for the search engines to index the potential solutions in the mailing
>>> list archives.
>>>
>>> --
>>> Best regards,
>>> Ivan
>>>
>>> [*]
>>> https://git-scm.com/book/en/v2/Git-Basics-Getting-a-Git-Repository
>>>
>>> [**]
>>>
>>>
>> https://git-scm.com/book/en/v2/Git-Basics-Recording-Changes-to-the-Repository
>>>
>>> [***]
>>> https://git-scm.com/book/en/v2/Git-Basics-Working-with-Remotes
>>>
>>
>>          [[alternative HTML version deleted]]
>>
>> ______________________________________________
>> R-package-devel using r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-package-devel
>>
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-package-devel using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-package-devel