Hi, "Loris Bennett" <loris.benn...@fu-berlin.de> writes:
> Hi John, > > John Hendy <jw.he...@gmail.com> writes: > >> On Mon, Mar 13, 2017 at 9:32 AM, Loris Bennett >> <loris.benn...@fu-berlin.de> wrote: >>> Hi, >>> >>> In a file that used to work, when I now try to run some R code which >>> generates an image, I get the error: >>> >>> cannot open file './usage_users_historical_facet.pdf' >>> >>> If I check the working directory for the R session I see >>> >>> getwd() >>> [1] "/usr/share/emacs/site-lisp/ess" >>> >>> instead of the directory in which my org file is located. >>> >>> I am now running Org 9.0.5 on Emacs 25.1.1. The last time things worked >>> I was definitely using an older version of Org and possibly Emacs 24. >>> >>> Any thoughts? >> >> I agree with Nick on an ECM. Just to start the ball rolling: >> [snip (54 lines)] >> >> How does this go for you? > > Thanks for doing most of the work for me. Unfortunately for me, the ECM > works fine, so I'm going to have to do some work myself and debug my > crufty old .emacs. I finally got round to bisecting my .emacs and worked out the sequence of events which lead me to shooting myself in the foot: 1. Installed Emacs 25 under /usr/local 2. Discovered that Emacs 25 couldn't find the ESS stuff 3. Defined a variable 'ess-directory' to be "/usr/share/emacs/site-lisp/ess/" 4. Used the variable in a function to add ESS to the load-path It turns out that variable 'ess-directory' already exists and is, according to the documentation, "the directory ESS is run from". Thus renaming the variable 'ess-directory' to 'non-in-the-foot-shooting-ess-directory' solves the problem. Cheers, Loris -- This signature is currently under construction.
