[R-pkg-devel] Warning in dir.create(vd2 <- "vign_test") : 'vign_test' already exists

Duncan Murdoch murdoch@dunc@n @end|ng |rom gm@||@com
Thu Jul 1 01:14:11 CEST 2021


On 30/06/2021 6:50 p.m., Rolf Turner wrote:
> 
> On Tue, 29 Jun 2021 14:09:34 -0500
> Spencer Graves <spencer.graves using effectivedefense.org> wrote:
> 
>> Hello, All:
>>
>>
>> 	  "R CMD check Ecfun_0.2-5.tar.gz" with R 4.1.0 under macOS
>> 11.4 gave "Warning in dir.create(vd2 <- "vign_test") : 'vign_test'
>> already exists".  It ultimately said, "Status: OK", so maybe I
>> shouldn't worry about it.  It's available at:
>>
>>
>> https://github.com/sbgraves237/Ecfun
>>
>>
>> 	  This passed all 4 GitHub Actions, so maybe I should not
>> worry about this.
>>
>>
>> 	  Comments?
>> 	  Thanks,
>> 	  Spencer Graves
> 
> It is rather presumptuous of me to comment, but for what it's worth I
> personally would not ignore such a warning.  There is (it seems to me)
> always a possibility that the warning is symptomatic of a problem that
> could at some stage whip around and bite you on the arse.  I would try
> to find out why the directory "vign_test" (unexpectedly?) exists.  Where
> does it come from? Who/what created it?
> 
> If its possible existence is "understandable" perhaps you should test
> for its existence (using file.exists()?) and if it does indeed exist,
> don't try to create it.
> 
> Feel free to ignore me (and I'll go away! :-) ).

I agree the warning shouldn't be ignored, but I believe it was created 
by R CMD check, not by Spencer directly.  So the real question is:  why 
was it already there?

Duncan Murdoch



More information about the R-package-devel mailing list