R devel - Sep 1997 - R-alpha: R interaction within ESS-4.9-b11: two small issues

1) I installed the latest ESS (Emacs Speaks Statistics) beta release
on a machine running Linux.  Since there is no S-PLUS product for
Linux, I set the ess-site.el file for R, xlisp-stat, and S version 4.
I.e.,
 ;; (1.5) Require the needed dialects for your setup.

 ;;(require 'essd-s+3)
 (require 'essd-r)
 (require 'essd-xls)
 ;;(require 'essd-sas)
 ;;(require 'essd-s3)  ;; You might not have this
 (require 'essd-s4)  ;; or this one...

That does not seem to work.  I believe that under the present
structure you must require essd-s+3 to get definitions of some
functions used by the other dialects.  Perhaps that dependency could
be removed at some time.

2) I think the current definitions
essd-r.el:61:    (ess-retr-lastvalue-command    . ".Last.value <-
get(\"smode.lvsave\",envir=1)\n")
essd-r.el:62:    (ess-save-lastvalue-command    .
"assign(\"smode.lvsave\",.Last.value,envir=1)\n")
may be misusing the "envir" argument.  I am getting messages in the
interaction window like
 > Error in get(x, envir, mode, inherits) : variable "smode.lvsave"
was not found
If I try to use assign with envir = 1 at the top level I get an error message
 > assign("smode.lvsave", 12, envir = 1)
 Error: invalid envir argument

I haven't gone into debugging of the ESS code to find out if this what
is causing the problem.

Should the assign and get use "inherits = TRUE" instead of
"envir=1"?
The R documentation is a bit vague about the use of the envir argument.

-- 
Douglas Bates                            bates@stat.wisc.edu
Statistics Department                    608/262-2598
University of Wisconsin - Madison        http://www.stat.wisc.edu/~bates/
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To:
r-devel-request@stat.math.ethz.ch
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

>>>>> "Douglas" == Douglas Bates
<bates@stat.wisc.edu> writes:
    Douglas> 1) I installed the latest ESS (Emacs Speaks Statistics)
    Douglas>    beta release
    Douglas> on a machine running Linux.  Since there is no S-PLUS product
for
    Douglas> Linux, I set the ess-site.el file for R, xlisp-stat, and S
version 4.
    Douglas> I.e.,
    Douglas>  ;; (1.5) Require the needed dialects for your setup.

    Douglas>  ;;(require 'essd-s+3)
    Douglas>  (require 'essd-r)
    Douglas>  (require 'essd-xls)
    Douglas>  ;;(require 'essd-sas)
    Douglas>  ;;(require 'essd-s3)  ;; You might not have this
    Douglas>  (require 'essd-s4)  ;; or this one...

    Douglas> That does not seem to work.  I believe that under the
    Douglas> present structure you must require essd-s+3 to get
    Douglas> definitions of some functions used by the other dialects.
    Douglas> Perhaps that dependency could be removed at some time.

I just did.  Sorry (changed for beta-12, to release Friday).  I've
moved the basic formatting into essl-s.el (ESS, Languages in the S
family), which is required by essd-s3, essd-s4, essd-s+3,essd-s+3...

    Douglas> 2) I think the current definitions

    Douglas> essd-r.el:61:    (ess-retr-lastvalue-command
    Douglas> . ".Last.value <-
get(\"smode.lvsave\",envir=1)\n")
    Douglas> essd-r.el:62:    (ess-save-lastvalue-command
    Douglas> .
"assign(\"smode.lvsave\",.Last.value,envir=1)\n")

    Douglas> may be misusing the "envir" argument.  I am getting
    Douglas> messages in the interaction window like

    >> Error in get(x, envir, mode, inherits) : variable
    >> "smode.lvsave" was not found 

    Douglas> If I try to use assign with envir = 1 at the top level I
    Douglas> get an error message

    >> assign("smode.lvsave", 12, envir = 1)

    Douglas>  Error: invalid envir argument

    Douglas> I haven't gone into debugging of the ESS code to find out
    Douglas> if this what is causing the problem.

    Douglas> Should the assign and get use "inherits = TRUE"
instead
    Douglas> of "envir=1"?  The R documentation is a bit vague
about
    Douglas> the use of the envir argument.

This is a bit more important.  I've not yet found an acceptable work
around (well, Kurt H. said he could provide a hacked solution soon).

I'm still hunting for suggestions...

best,

-- 
-tony (Anthony Rossini)	           Statistics Dept., U South Carolina
rossini@stat.sc.edu	           Columbia, SC 29208
http://www.stat.sc.edu/~rossini/   803-777-3578(O) 803-777-4048 (fax)  
        "'Bother,' said Pooh, as Cthulu rose up and ate him."
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To:
r-devel-request@stat.math.ethz.ch
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-