[R] Library com.braju.oo provides reference variables
Henrik Bengtsson
henrikb at braju.com
Wed Mar 28 00:11:23 CEST 2001
After posting (Mar 22 2001) a suggestion on how to call by reference using
.Alias several people replied an said that .Alias should NOT be used for
different reasons. I also found out that problems can occur when ".Alias
pointers" is passed between different container objects (lists etc). The
behavior of .Alias turned out to be unpredictable in some cases. In other
words, ".Alias pointers" are out of the question.
However, pointers, or preferable references, are important and needed. One
reason is that reference variables are memory efficient. Another reason is
that, the concept of reference variables allows you to have several
container objects referring to the same object. For instance, you might want
to have one list containing all red cars and another list containing all 4
cylinder cars. It might be that the two lists contains the same car(s). For
memory reasons you do not want to keep copies of the common cars in the two
lists. Also, you might want to do modifications to car objects, e.g.
changing the license number. Such changes should be reflected in the objects
contained in the two lists, without having to create new lists etc. I hope
these arguments convince some that references are necessary.
The library com.braju.oo provides the notion of references and objects. The
library is written in 100% [R] so it should be possible to use the library
on any system (I know for sure that it works on both Windows 98 and Sun
Solaris). I have tested it myself for a couple of weeks and would now like
to get other peoples opinion and feedback on it. The library seems to work
as expected. However, it should still be considered as a beta version since
the API might be changed. The com.braju.oo library can be downloaded at
http://www.braju.com/R/com.braju.oo/
There you will also find [R] help files and some example code. Please, give
me feedback on this one.
Have a nice day
Henrik Bengtsson
Lund University & UC Berkeley
henrikb at braju.com
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
More information about the R-help
mailing list