fUtilities::isS4

Yohan Chalabi chalabi at phys.ethz.ch
Fri Jan 30 09:17:14 CET 2009


>>>> "PS" == Phil Spector <spector at stat.berkeley.edu>
>>>> on Thu, 29 Jan 2009 10:32:49 -0800 (PST)

   PS> Hi!
   PS> I'm having some difficulties using the isS4 function in
   PS> the fUtilities package.  Specifically, I think that the
   PS> test for NULL slotNames may not always give the correct
   PS> result.
   PS>
   PS> For example, consider an lm object:
   PS>
   PS> > x = rnorm(10)
   PS> > y = rnorm(10)
   PS> > z = lm(y~x)
   PS> > base::isS4(z)
   PS> [1] FALSE
   PS> > fUtilities::isS4(z)
   PS> [1] TRUE
   PS> > slotNames(z)
   PS> [1] .S3Class
   PS>
   PS> Since showMethods(class='lm') doesn't return anything, I don't
   PS> think lm objects should be identified as S4 objects by isS4.
   PS>
   PS> Thanks!


Hi Phil,

Thanks to point this out!

I will remove 'isS4' from the 'fUtilities' package because it is
not needed anymore.

BTW, have you noticed that the new Rmetrics packages should not
depend on 'fUtilities' which will be considered as deprecated in the
future.

If you are using 'fCalendar' and 'fSeries', please move to their new
versions named timeDate and timeSeries respectively.

HTH
Yohan

-- 
PhD student
Swiss Federal Institute of Technology
Zurich

www.ethz.ch



More information about the Rmetrics-core mailing list