[R-pkg-devel] some notes

Uwe Ligges ligges at statistik.tu-dortmund.de
Tue Apr 24 19:44:45 CEST 2018



On 24.04.2018 19:40, Duncan Murdoch wrote:
> On 24/04/2018 1:07 PM, Xubo Yue wrote:
>> Everyone, I appreciate for your help!
>>
>> Hey Uwe Ligges:
>>
>> There are four examples in my package: read data, calculate 
>> correlation, plot, and summarize connection network. The plot is the 
>> slowest one and I have to use donttest{} to avoid testing it on CRAN. 
>> All other examples are fast and take less than 10s, so I can let CRAN 
>> test it.
>>
>> Also Swetlana suggest me to write a vignette to demonstrate plot. 
>> However, my plot is in rgl device (an rgl window separated from 
>> default R plot window) and cannot be integrated into rmarkdown.
> 
> That's not true -- see the rgl vignettes, and the rglwidget() function. 
> The main limitation is that the output needs to be HTML rather than 
> LaTeX/PDF.  (There are ways to incorporate interactive rgl graphics into 
> PDF documents, but they are limited and fragile, so I don't recommend 
> them.)

Same from here, note that Swetlana generally suggested to move things 
that are slightly longer than the 5sec threshhold for examples to tests 
*or* perhaps in vignettes, this was not intended as a specific advice 
for this particular example.

Best,
Uwe

> Duncan Murdoch
> 
>>
>> So in a nutshell: can I just use donttest{} to my plot example and 
>> comment that it takes a long time to run?
>>
>> It is my first time to submit r package, sorry for any inconvenience!
>>
>> Thank you very much for your help!
>>
>> On Tue, Apr 24, 2018 at 1:01 PM, Duncan Murdoch 
>> <murdoch.duncan at gmail.com <mailto:murdoch.duncan at gmail.com>> wrote:
>>
>>     On 24/04/2018 12:48 PM, Uwe Ligges wrote:
>>
>>
>>
>>         On 24.04.2018 18:45, Spencer Graves wrote:
>>
>>     [... deleting irrelevant stuff...]
>>
>>
>>
>>                      If it becomes infeasible to find smaller datasets,
>>             etc., you can
>>             cut out certain tests with a construct like the following:
>>
>>
>>             if(!fda::CRAN()){
>>             # ... tests to run on other computers but to skip on 
>> CRAN.  ...
>>             }
>>
>>
>>         Ummm, how should fda::CRAN() now that it runs on CRAN?
>>
>>         CRAN prefers to ship a set of examples/tests etc. that meet our
>>         requirements. If you want to ru addtional tests and/or examples,
>>         then
>>         run them conditionally on something that is true for your local
>>         machine,
>>         e.g. some defined environment variable.
>>
>>
>>     For Xubo Yue, who may not be aware of all of the players:
>>
>>     Uwe Ligges is one of the individuals who runs CRAN.  When you submit
>>     a package to CRAN, you are asking Uwe and others to distribute your
>>     work. You should listen to what he says.
>>
>>     Spencer Graves and I are not involved in running CRAN.  We will give
>>     you advice that we think is good advice, but if it conflicts with
>>     Uwe's, follow his.
>>
>>     Duncan Murdoch
>>
>>
>>
>>
>> -- 
>> *Xubo (Max) YUE, *PhD Student,
>> /The Industrial and Operation Engineering Department,/
>> /College of Engineering,/
>> /The University of Michigan, Ann Arbor, MI, USA/
>>
>>
>



More information about the R-package-devel mailing list