R help - Apr 2002 - rbind(NULL,NULL)

In the time honored spirit of wishing to do nothing well, could I suggest
that the Splus (versions 5 and 6)  response to:
> rbind(NULL,NULL)
NULL

is preferable to the R response:
> rbind(NULL,NULL)
Error in rbind(NULL, NULL) : attempt to set an attribute on NULL

This is on:

platform sparc-sun-solaris2.8
arch     sparc
os       solaris2.8
system   sparc, solaris2.8
status
major    1
minor    4.1
year     2002
month    01
day      30
language R

It is tempting to conclude from this that Splus does nothing better than R,
but it is always dangerous to generalize from one example.


url:	http://www.econ.uiuc.edu		Roger Koenker
email	roger at ysidro.econ.uiuc.edu		Department of Economics
vox: 	217-333-4558				University of Illinois
fax:   	217-244-6678				Champaign, IL 61820

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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 at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

>>>>> "Roger" == Roger Koenker <roger at
ysidro.econ.uiuc.edu> writes:
    Roger> In the time honored spirit of wishing to do nothing
    Roger> well, could I suggest that the Splus (versions 5 and 6) response
to:

    >> rbind(NULL,NULL)
    Roger> NULL

    Roger> is preferable to the R response:

    >> rbind(NULL,NULL)
    Roger> Error in rbind(NULL, NULL) : attempt to set an
    Roger> attribute on NULL

a simpler example is

   rbind(NULL)
or cbind(NULL).

However, I don't see why allowing the above should be better than
giving an error.  In R, we have made a distinction between
0-length vectors and NULL for quite a while, and AFAIR, S-plus
has done so, too, in other contexts.

Note that   rbind(numeric(0))  {as in rbind(x[x!=x])} *does* work
(i.e. give 0 x 0 matrix, not "NULL"!).


    <...>

    Roger> It is tempting to conclude from this that Splus does
    Roger> nothing better than R, but it is always dangerous to
    Roger> generalize from one example.

[ big ;-) ]

Note that S-plus has a bug here: It does not work according to
documentation, since its  help(cbind) {or (rbind)} says
>>                        Build Matrix from Columns or Rows
>>                                        
>> DESCRIPTION:
>> 
>>    Returns a matrix that is pieced together from several vectors and/or
>>    matrices. The functions cbind and rbind are generic. In particular,
>>    there are cbind and rbind methods for data frames and classes in the
>>    Matrix library.
whereas in your example it returns NULL which is not a matrix...
I hope this doesn't sound too nitpicking, but you started to
talk about the nits.. :-)

With high regards,
Martin

Martin Maechler <maechler at stat.math.ethz.ch>
http://stat.ethz.ch/~maechler/
Seminar fuer Statistik, ETH-Zentrum  LEO C16	Leonhardstr. 27
ETH (Federal Inst. Technology)	8092 Zurich	SWITZERLAND
phone: x-41-1-632-3408		fax: ...-1228			<><
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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 at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._