[Rd] Reorganization of packages in the R distribution

[Henrik Bengtsson]

> -----Original Message-----
> From: r-devel-bounces at stat.math.ethz.ch 
> [mailto:r-devel-bounces at stat.math.ethz.ch] On Behalf Of Paul Gilbert
> Sent: den 12 december 2003 16:20
> To: Prof Brian Ripley
> Cc: Paul Gilbert; R-devel at stat.math.ethz.ch
> Subject: Re: [Rd] Reorganization of packages in the R distribution
> 
> 
> Prof Brian Ripley wrote:
> 
> >On Fri, 12 Dec 2003, Paul Gilbert wrote:
> >
> >  
> >
> >>Prof Brian Ripley wrote:
> >>
> >>    
> >>
> >>>There are a small number of CRAN packages that attempt to modify 
> >>>system functions and so will need updating.  (Known 
> examples are in 
> >>>dse:tframe, gregmisc and mclust and some testing code elsewhere.)
> >>>
> >>>      
> >>>
> >>Brian
> >>
> >>What do you mean by "updating?"  In tframe I modify a few
functions 
> >>like
> >>
> >>     start <- function (x, ...) if (is.Ttframed(x))
start(tframe(x),
> >>...) else UseMethod("start")
> >>
> >>If that can no longer be done then this is a serious fundamental 
> >>change
> >>that breaks all my code. I hope that is not what you mean. I'm
just 
> >>going away for a week, but will follow up when I return.
> >>    
> >>
> >
> >It's always been incorrect code, and it no longer works.  You
should 
> >not be masking system generics, as the namespace 
> registration mechanism 
> >does not work on your version.
> >  
> >
> So what would be the proper mechanism for doing something like this?

> Basically I need to do something that I believe is like multiple 
> inheritance. I need to dispatch on the class of a certain part of an

> object rather than on the object itself.  (BTW, this code has 
> worked for 
> about  ten years and  I believe predates your definition of  
> "corrrect".)
> 
> Paul

Hi, not knowing yours details, but here's an idea that might work.
Under S3/UseMethod() you can "emulate" multiple inheritance by adding
one superclass in front of the other to the class attribute. For
instance, if ClassAA inherits from ClassA and ClassBB inherits from
ClassB and MyClass inherits from ClassAA and ClassBB, then you'll get

 class(x) <- c("MyClass", "ClassAA", "ClassBB", "ClassA", "ClassB")

Of course, it is important to get the order correct, and things can
get messy if you're not careful. So, whatever function makes your
object 'x' "Ttframed", couldn't that function at the same time add the
superclass "TtframedClass" in front of the class vector before
returning the object?! 

However, I see a real problem if you do not control/own the functions
that makes an object "Ttframed".

Best wishes

Henrik Bengtsson
Lund University