Skip to content

R doesn't recognize R_HOME value

9 messages · John Kane, Jeff Newmiller, Peter Dalgaard +2 more

#
Upgraded to 2.15.1 from 2.15.0 this morning, on Windows 7.
I'm setting R_HOME in Control Panel, and before the upgrade, 2.15.0
recognized the value with no problems and would use it to find
rprofile.site, etc.
Now, after upgrade, neither .0 nor .1 recognize it at all. From the cmd
line, Windows indicates it is definitely set to the value I've specified.
Anyway, as a result, R is not reading in renviron.site and not running
rprofile.site.

I don't know where to start troubleshooting this. I've restarted the
machine...because it's Windows...to no effect.



--
View this message in context: http://r.789695.n4.nabble.com/R-doesn-t-recognize-R-HOME-value-tp4637510.html
Sent from the R help mailing list archive at Nabble.com.
#
Can we have a Window subset of fortunes?

John Kane
Kingston ON Canada
____________________________________________________________
FREE 3D EARTH SCREENSAVER - Watch the Earth right on your desktop!
#
???

Windows 7 is where I'm seeing the problem.

The root problem: while I have R_HOME legitimately specified within Windows
7, and issuing 'set R_HOME' from the command line returns exactly the path
I've specified, doing a Sys.getenv('R_HOME') from R returns a completely
different path.

I ask, "What would cause a Sys.getenv() call to return a different value
than the OS does?"



--
View this message in context: http://r.789695.n4.nabble.com/R-doesn-t-recognize-R-HOME-value-tp4637510p4637672.html
Sent from the R help mailing list archive at Nabble.com.
#
You are asking us how your operating system works, but this list is about R.

The key thing you appear to not understand is that environment variables are local to each process, and default environment variables are received from the process that started any given process.

I ask you to go study this in a more appropriate place, and for our sanity in the meantime suggest that you set the environment variables in the Advanced area of the System control panel, and then log out and log in to make sure all your processes receive them.
---------------------------------------------------------------------------
Jeff Newmiller                        The     .....       .....  Go Live...
DCN:<jdnewmil at dcn.davis.ca.us>        Basics: ##.#.       ##.#.  Live Go...
                                      Live:   OO#.. Dead: OO#..  Playing
Research Engineer (Solar/Batteries            O.O#.       #.O#.  with
/Software/Embedded Controllers)               .OO#.       .OO#.  rocks...1k
--------------------------------------------------------------------------- 
Sent from my phone. Please excuse my brevity.
Kirk Fleming <kirkrfleming at hotmail.com> wrote:

            
#
On 24/07/2012 21:57, Kirk Fleming wrote:
A different process and hence a different environment ... this is why 
they are called 'environment variables' and are specific to a process.

'The command line' is actually a shell process, not the OS itself. 
Setting environment variables in a shell affects only that shell and 
those which inherit from it.

See ?R_HOME in R, which tells you that is the R process 'is normally set 
on startup'.   'Normally' because there are many ways to start R 
(embedded, for example: see 'Writing R Extensions'), and you have not 
actually told us how you started R.  But Rgui.exe and Rterm.exe do 
always set R_HOME.
#
On Jul 25, 2012, at 07:58 , Prof Brian Ripley wrote:

            
Notice also that R_HOME represents R's own knowledge of its installation directory, where it looks for various support files, the default package library, etc. 

R sets R_HOME for itself; even if you could, you really do not want to override it. Try looking at ?Startup for the correct ways to specify alternative locations of profile files, etc.
#
On 25/07/2012 09:14, peter dalgaard wrote:
alternative locations of profile files, etc.

Well, I was being careful, and the following may be too technical for 
R-help.

R's own front-ends set R_HOME for themselves, but embedded versions of R 
may well get it from somewhere else and for such uses the end-user may 
need to specify it as an environment variable in the process embedding 
R.  Alternative front-ends (e.g. RStudio) are examples of embedding and 
do need a way to find R's own installation directory (which may or may 
not be recorded in the Windows Registry on Windows).  Rgui.exe uses its 
own path to find R_HOME.
#
Panel, and never actually set it from within a shell. I'll take the advice
given, however, and not set it to any value at all. I did feel this was an R
question since things were working as expected until upgrading, then not
working as expected afterward. With my new enhanced understanding I'll start
from scratch and get it resolved.



--
View this message in context: http://r.789695.n4.nabble.com/R-doesn-t-recognize-R-HOME-value-tp4637510p4637795.html
Sent from the R help mailing list archive at Nabble.com.
1 day later
#
I wanted to wrap this up, since I feel it's been resolved.

R_HOME never did need to over-ridden by me, was not being over-ridden in
spite of my attempts, and was never a factor during the long period when
'everything worked'.

In fact, my entire ordeal was caused by removing a comment from the
rprofile.site file and in so doing, accidentally removing a closing brace
from a function definition. In using a quick-and-dirty editor, this error
wasn't highlighted. Having no output from rprofile whatsoever and no errors,
I concluded it was not being run, and not being found.



--
View this message in context: http://r.789695.n4.nabble.com/R-doesn-t-recognize-R-HOME-value-tp4637510p4638006.html
Sent from the R help mailing list archive at Nabble.com.