S+6 setup for Unix

Rodney Sparapani rsparapa at post.its.mcw.edu
Thu May 1 17:20:24 CEST 2003


>
>I think you are looking in the wrong place.
>ess-S+6-startup-delay is defined in essdsp6w.el for windows.
>For unix you should be looking at essd-sp6.el
>
>On windows there is a distinction between gnuclient and gnuclientw.
>I don't think that distinction exists on unix.
>With gnuclient, the S process waits until the edit is complete.
>This is the correct choice for edit().
>
>With gnuclientw, the S process does not wait.  This is the correct choice
>for page()
>
>I believe inferior-S+6-print-command is normally send to lp or lpr on unix
>.machines.  That's not always available on windows so I sent it to the
>emacs process (with gnuclientw).  Sometimes the user will actually print it.
>I find that I want the print material in a searchable and scrollable buffer.
>
>Some unix systems use gnuclient, others emacsclient.  Make sure you use the
>one that you have.
>
>Rich
>

Rich:

Thanks that helps alot.  But, I wasn't looking in any files.  I was just looking 
at the variable settings with customize.  And, even though I'm on Unix, I see 
alot of Windows settings.  I think we need to have a more user-friendly model, 
i.e. ESS[SAS] has the same variables under Unix, Mac and Windows with different 
defaults, if necessary.  We can't expect that someone is going to go to the 
source code each time, either.  And, if they do need to, some way of minimizing 
the number of variants is necessary.  Perhaps I'm out of my mind, but looking 
over our notes from JSM, it would be nice if there were as few files as 
possible:

1: ess-s.el or ess-sp.el (basics)
2: and ess-sp6.el (S+6-specific for both Unix and Windows)

Rodney Sparapani              Medical College of Wisconsin
Sr. Biostatistician           Patient Care & Outcomes Research
rsparapa at mcw.edu              http://www.mcw.edu/pcor
Was 'Name That Tune' rigged?  WWLD -- What Would Lombardi Do



More information about the ESS-help mailing list