emacs-orgmode
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [O] HowTo? have R :sessions in org mode invoke a specific R


From: Cook, Malcolm
Subject: Re: [O] HowTo? have R :sessions in org mode invoke a specific R
Date: Fri, 16 Oct 2015 19:44:57 +0000

Replying on an old thread:

> >>On Mon, Nov 4, 2013 at 11:22 AM, Cook, Malcolm <address@hidden>
 > wrote:
 > >  >> I wish to lock an org buffer to a particular installation of R so that 
 > > all
 > code blocks use it.
 > >  >>
 > >  >> I found that I could not, for instance, write `:session R-3.0.1`.
 > >  >>
 > >  >> Is there some best way for me to accomplish this?
 > >  >>
 > >  >> Perhaps some way of setting inferior-R-program-name to a buffer local
 > value?
 > >  >
 > >  >I'm no good at in-buffer variables, but this one looks of interest:
 > >  >org-babel-R-command
 > >
 > > Indeed, John,  that is the variable I would like to have respected
 > > when I
 > >
 > > (setq-local inferior-R-program-name "R-3.0.1")
 > >
 > >
 > > But, alas it is not.
 > >
 > > Any other takers?
 > >
 > > Thanks!
 > 
 > I think you got it backwards: if you are trying to use org-babel to evaluate 
 > an R
 > code block, then the variable you should set is org-babel-R-command as John
 > indicated. Setting inferior-R-program-name to something does not tell
 > anything to org-babel. To use your term, org-babel does not respect inferior-
 > R-program-name; the only thing it knows is org-babel-R-command. Only ESS
 > respects inferior-R-program-name, but you are not using ESS here. So have
 > you tried
 > 
 >  (setq-local org-babel-R-command "R-3.0.1 --slave --no-save")

Yes!  Indeed!  And, Thanks!  

And... I don't even need to remember to eval the setq statement if I include in 
my org the following works everytime

# Local Variables:
# org-babel-R-command: "R-3.0.1 --slave --no-save"
# End:

Or put it as first line in the file:

# -*- org-babel-R-command: "R-3.0.1 --slave --no-save"-*-

Cheers,

Malcolm

 > 
 > in your org-mode buffer where you are evaluating your R code?
 > The options are guesses based on the current setting. Untested since I know
 > virtually nothing about R.
 > 
 > Nick
 > 




reply via email to

[Prev in Thread] Current Thread [Next in Thread]