The way we solve this issue on our machines (and it is much more serious than you describe on a large, shared system) is by using lmod, which is basically a fancy and convenient way to manage your environmental variables. Basically you add anaconda to your path only when using it. Doing this by hand is extremely annoying, hence lmod. You may want to experiment with it, or at least look at what the user experience look like in our docs: https://www2.cisl.ucar.edu/resources/computational-systems/yellowstone/access-and-user-environment/environment-modules and maybe also: https://www2.cisl.ucar.edu/resources/computational-systems/yellowstone/software/python
On Thu, Mar 3, 2016 at 12:20 PM, Tiffany Timbers <[email protected]> wrote: > Hi all, > > I have been using Anaconda as package/environment manager for my Python > version and packages, but this has led to my Bash Shell using the Anaconda > version of R when I run Rscripts, and not the version of R I use in RStudio > when I develop. This is something I don't want to happen, I want the command > line to access the same version of R and R packages as I have installed via > RStudio. I prefer to manage R and R package versions and installation in R > because I can get more recent versions of R and R packages via RStudio > compared to using conda. > > I have tried several things to fix this. For example, specifying the PATH in > .bash_profile to be the R version I wanted to use didn't work unless I > removed the PATH to Anaconda, but then I can't use conda to manage Python. > The only thing I could do to fix this was to delete the R and Rscript that > Anaconda installed. I don't think this fix is optimal, is there a better way > to deal with this? > > Also, this may be something we need to think about when people who develop > in R routinely and use Rscripts, come to a Software Carpentry workshop to > learn Python and install Anaconda. Doing this will change the version of R > and R packages that they can now access from the command line (unless my > experience here is unique). So we in effect, in some cases our installation > might be breaking some of their working tools... > > Thanks, > Tiffany > > Thanks! > Tiffany > -- > Tiffany A. Timbers, PhD > Banting Postdoctoral fellow, Leroux Lab > Department of Molecular Biology and Biochemistry > Simon Fraser University > 8888 University Drive > Burnaby, BC, V5A 1S6 > > 604-803-4962 > > [email protected] > tiffanytimbers.com > @TiffanyTimbers > > _______________________________________________ > Discuss mailing list > [email protected] > http://lists.software-carpentry.org/mailman/listinfo/discuss_lists.software-carpentry.org _______________________________________________ Discuss mailing list [email protected] http://lists.software-carpentry.org/mailman/listinfo/discuss_lists.software-carpentry.org
