use a similar repl result method in swank-media.R for ggplot objects

[swankr.git] / README

diff --git a/README b/README
index cf5e37fc9e32309d72e06d61ceed8adfc1bdae8c..ce8005c3f3037eebf4bdb7d1b7c2d92b8d4fd1bc 100644 (file)
--- a/README
+++ b/README
@@ -23,6 +23,7 @@
 * Installation
 ** Running
    To begin using swankr:
+
    1. start R;
    2. load the swank.R file:
 #+BEGIN_SRC R
@@ -33,16 +34,24 @@
       swank();
 #+END_SRC
    4. within emacs, load and initialize slime;
-   5. run slime-connect, accepting the default host and port, and
-      acknowledging the protocol version mismatch.
-   At this point, a R REPL should appear.
+#+BEGIN_SRC emacs-lisp
+      (require 'slime)
+      (slime-setup '(slime-repl))
+#+END_SRC
+   5. run =M-x slime-connect=, accepting the default host and port,
+      and acknowledging the protocol version mismatch.
+
+   At this point, an R REPL should appear.
 ** Emacs customization
    At a minimum, slime needs to be set up to function.  I've
    used the following forms in my =~/.emacs=
 #+BEGIN_SRC emacs-lisp
    (require 'slime)
-   (slime-setup '(slime-repl slime-scratch))
-#+END_SRCA
+   (slime-setup '(slime-repl slime-scratch slime-media))
+#+END_SRC
+   The =slime-media= contrib is new and (at present) R-specific,
+   allowing for image results to be embedded in the REPL.
+
    In addition, for keybindings like =C-c C-c= to work properly, emacs
    needs to be told how to guess where a function definition begins.
    This can be achieved with /e.g./
@@ -68,7 +77,8 @@
   accessible through
   <http://common-lisp.net/r/users/crhodes/swankr.git> and
   git://common-lisp.net/crhodes/swankr/swankr.git; a web view of the
-  development history is [[http://common-lisp.net/gitweb?p=users/crhodes/swankr.git][available through gitweb]].
+  development history is [[http://common-lisp.net/gitweb?p=users/crhodes/swankr.git][available through gitweb]].  You can also view
+  the current lists of [[file:BUGS.org]] and [[file:TODO.org]] items.
 * Acknowledgments
   Thanks to my colleagues at [[http://www.gold.ac.uk/][Goldsmiths, University of London]], for
   suggesting that I investigate R for numerical and graphical