[R] tempdir() does not respect TMPDIR

[Jinsong Zhao]

Thanks a lot for the confirmation and explanation.

On 2020/8/31 3:45, Henrik Bengtsson wrote:
> Sorry, I should retract my claim that it's too late to set TMPDIR in
> .Renviron.  It does indeed work on Linux and R 4.0.2, e.g.
> 
> $ cd
> $ mkdir test
> $ cd test
> $ echo "TMPDIR=$PWD" > ./.Renviron
> $ cat ./.Renviron
> TMPDIR=/home/hb/test
> Rscript --no-init-file -e "tempdir()"
> [1] "/home/hb/test/RtmpyH47tc"
> 
> Hmm... either this has changed "recently" or I've got it wrong all the
> time.  Eitherway, I need to revise the vignette in my 'startup'
> package.

This works just becuase .Renviron is in the same directory R launched. 
It confirmed what you stated that launched Rgui via double-clicking 
.RData. Generally, we put .Renviron in R_USER (on Windows).

> 
> Sorry for the misleading comment.
> 
> So, back to your comment about it does *not* work, that is,
> ~/.Renviron is not read, when you double-click on an .RData file.  I
> just tried with R 4.0.2 in a Windows 10 VM and I think I can reproduce
> what you're describing.
> 
> The problem seems to be that when one launches Rgui via
> double-clicking .RData, the Rgui will only read ./.Renviron, that is,
> the .Renviron file that is located in the same folder as the .RData
> file.  It will never load ~/.Renviron (e.g.
> C:/Users\alice/Documents/.Renviron) unless the .RData file is in that
> folder too.

I should guess to that you said. When I launched R from console (cmd on 
Windows) by command like:

C:\Users\zjs>"c:\Program Files\R\R-4.0.2\bin\R.exe"
or
C:\Users\zjs>"c:\Program Files\R\R-4.0.2\bin\x64\Rgui.exe"

The tempdir() does not return the value I set in .Renviron. It means 
when R launching, it search .Renvrion in the current directory. If it 
does not get the .Renviron, it does not search it in R_USER. Am I right?

So where I set TMPDIR, R would use it whenever it launched? The purpose 
that I tried to set TMPDIR is RStudio may refuse to work when user name 
contained non-latin characters.

Thanks again.

Best,
Jinsong

> 
> This looks odd to me but it could be that I made another mistake in my
> conclusions above.  I let someone else with a less mushy brain take
> over from here.
> 
> /Henrik
> 
> On Sat, Aug 29, 2020 at 4:31 PM Jinsong Zhao <jszhao using yeah.net> wrote:
>>
>> I read the help page, I don't understand it very well, since I set the
>> environmental variable TMPDIR in .Renviron. What confused me is when
>> double clicking the *.RData to launch R, the tempdir() does not respect
>> the environmental variable TMPDIR, but launch R by double clicking Rgui
>> icon does.
>>
>> Best,
>> Jinsong
>>
>> On 2020/8/30 0:36, Henrik Bengtsson wrote:
>>> It is too late to set TMPDIR in .Renviron.  It is one of the
>>> environment variables that has to be set prior to launching R.  From
>>> help("tempfile", package = "base"):
>>>
>>> The environment variables TMPDIR, TMP and TEMP are checked in turn and
>>> the first found which points to a writable directory is used: if none
>>> succeeds ‘/tmp’ is used. The path should not contain spaces. **Note
>>> that setting any of these environment variables in the R session has
>>> no effect on tempdir(): the per-session temporary directory is created
>>> before the interpreter is started.**
>>>
>>> /Henrik
>>>
>>> On Sat, Aug 29, 2020 at 6:40 AM Jinsong Zhao <jszhao using yeah.net> wrote:
>>>>
>>>> Hi there,
>>>>
>>>> When I started R by double clicking on Rgui icon (I am on Windows), the
>>>> tempdir() returned the tmpdir in the directory I set in .Renviron. If I
>>>> started R by double clicking on a *.RData file, the tempdir() return the
>>>> tmpdir in the directory setting by Windows system. I don't know whether
>>>> it's designed.
>>>>
>>>>    > sessionInfo()
>>>> R version 4.0.2 (2020-06-22)
>>>> Platform: x86_64-w64-mingw32/x64 (64-bit)
>>>> Running under: Windows 10 x64 (build 18363)
>>>> ...
>>>>
>>>> Best,
>>>> Jinsong