R-beta: dataframes and functions

Jim Lindsey (jlindsey@luc.ac.be)
Wed, 18 Feb 1998 10:06:07 +0100 (MET)

From: Jim Lindsey <jlindsey@luc.ac.be>
Message-Id: <9802180906.AA09029@alpha.luc.ac.be>
Subject: R-beta: dataframes and functions
To: r-help@stat.math.ethz.ch (r-help)
Date: Wed, 18 Feb 1998 10:06:07 +0100 (MET)

As I guessed, globality of dataframes in functions is another one of
those wonderful S-Plus features. Is there any advantage  to this???
  Note that an undetached dataframe `disappears' when the function
exits. But the invisible variables it contained do not. I had about 30
invisible copies of the variables that ls() could not see. But one
called factor kept giving a warning message (outside the original
function) so I knew something fishy was going on. What possibly could
be the advantage of such behaviour?
  This is closely related to the reason .RData is so dangerous for
serious data analysis, as I described on the list about a year ago.
All sorts of invisible (and unknown, unless warnings appear such as
ignoring nonfunctions or incompatible vector lengths) conflicts can
occur with old dataframes in the memory image. That is why .RData is
banned on our site.
r-help 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-help-request@stat.math.ethz.ch