[ESS] S+ re-factoring ( some recent bugs/annoyances)

[Vitalie Spinu]

Dear Gerald and other S+ users, 

  >> gerald.jean at dgag.ca
  >> on Wed, 14 Nov 2012 10:34:46 -0500 wrote:

  > Thank you very much, I am a long time user of ESS (10+ years) and this
  > "feature" introduced with a more R-slanted ESS was for me an annoyance!!!

There have been a couple of consistency measures taken with respect to
naming conventions in ESS[S+]. From now on ESS does not keep track of S+
versions with one-emacs-package-per-S+-version as it did before, but
rather one common S+ package for all versions. All variables and
function names are now figured as *-S+-* instead of the old *-S+6-*. The
main entry function is S+ and the ess-dialect name is also S+. Program
location is in inferior-S+-program-name, and it defaults to Splus on
unix-like machines.

These changes do not mean that you have to change any of your
configuration files, the old names have been kept as aliases.

As we don't have too many S+ testers in ess-core (only one actually), we
would appreciate if some of you can test ESS[S+] from SVN before the
next update release.

  > Another annoyance is that I can't get a function's argument list anymore by
  > just pressing F3, and I think any R-Splus user will use this feature
  > several times a day, in R, no problems, works fine but not in Splus
  > anymore!!!


I would like to solve this, but I don't have S+. 

You mean eldoc arguments right? F3 is your custom shortcut. In order to
implement even a basic eldoc for S+ I need to know how all arguments of
a function could be retrieved.

Does the following work?
  
  > argsAnywhere("plot")
  function (x, y, ...) 
  NULL
  > args("plot")
  function (x, y, ...) 
  NULL
  > 

In addition to the above, R has utils:::functionArgs that returns all
the posible arguments of function's methods (S3 only). S+ is completely
S4 right? Is there an easy way to retrieve all posible arguments of all
the methods of a function?

    Thanks, 
    Vitalie.