R help - Sep 2004 - S4 method selection based on second argument

I'm translating some Matlab code and need some help figuring
out how to change this call into an S4 generic method.

In matlab, there's a function called 'repmat' with three
calling sequences (all I have to deal with anyway):
    1) B = repmat(A, m, n)
    2) B = repmat(A, [m n])
    3) B = repmat(A, n)
In all cases, A is the fill value, m is number of rows,
and n is number of columns.

As separate functions, the translations would roughly be:

repmat1 <- function(A, m, n) {
    kronecker(matrix(1, n, m), A)
}

repmat2 <- function(A, rc) {
    repmat1(A, rc[1], rc[2])
}

repmat3 <- function(A, n) {
    repmat1(A, n, n)
}

Suggestions?


----------------------------------------------------------
SIGSIG -- signature too long (core dumped)

Hi,
If it's the only reason why you think you have to use S4 then take a 
look at ?missing.
If you are interested in S4:
The first version gives an overview of setMethods. Then I give a simpler 
example. And finally a pure ?missing solution.

A) Declare function repmat as generic
if(!isGeneric("repmat"))
setGeneric("repmat"
           ,function(obx,oby,obz,...)
           standardGeneric("repmat"))

#B = repmat(A, m, n)

setMethod("repmat",signature(obx="numeric",oby="numeric",obz="numeric")
    ,function(obx,oby,obz){print("thre arg call")
    })

#B = repmat(A, [m n])
#B = repmat(A, n)

setMethod("repmat",signature(obx="numeric",oby="numeric",obz="missing")
    ,function(obx,oby)
    {
       print("two arg")
    #handle the difference in oby.
        if(length(oby)==1)
         oby
       else
         obx
    })

##END A ###
#simpler

#if(!isGeneric("repmat")) #commented out because you have already
declared it.
setGeneric("repmat"
           ,function(object,...)
           standardGeneric("repmat"))

setMethod("repmat",signature(object="numeric")
,function(object,oby,obz)
{
	#using ?missing handle the differences.
	if(missing(obz))
	print("new version")
})

Note that you can have the same in S3 using ?missing
something like this.

repmat <- function(A, m, n) {
	if(missing(n)
	{
		if(length(m)==2)
			{	
				repmat1(A, m[1], m[2])
			}
		else
		kronecker(matrix(1, m, m), A)	
	}
	kronecker(matrix(1, n, m), A)}
}
/E



Paul Roebuck wrote:
>I'm translating some Matlab code and need some help figuring
>out how to change this call into an S4 generic method.
>
>In matlab, there's a function called 'repmat' with three
>calling sequences (all I have to deal with anyway):
>    1) B = repmat(A, m, n)
>    2) B = repmat(A, [m n])
>    3) B = repmat(A, n)
>In all cases, A is the fill value, m is number of rows,
>and n is number of columns.
>
>As separate functions, the translations would roughly be:
>
>repmat1 <- function(A, m, n) {
>    kronecker(matrix(1, n, m), A)
>}
>  
>
>repmat2 <- function(A, rc) {
>    repmat1(A, rc[1], rc[2])
>}
>
>repmat3 <- function(A, n) {
>    repmat1(A, n, n)
>}
>
>Suggestions?
>
>
>----------------------------------------------------------
>SIGSIG -- signature too long (core dumped)
>
>______________________________________________
>R-help at stat.math.ethz.ch mailing list
>https://stat.ethz.ch/mailman/listinfo/r-help
>PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
>
>  
>

-- 
Dipl. bio-chem. Witold Eryk Wolski         
MPI-Moleculare Genetic
Ihnestrasse 63-73 14195 Berlin           _
tel: 0049-30-83875219                   'v'
http://www.molgen.mpg.de/~wolski       /   \
mail: witek96 at users.sourceforge.net  ---W-W----
      wolski at molgen.mpg.de

Paul Roebuck <roebuck <at> odin.mdacc.tmc.edu> writes:

: 
: I'm translating some Matlab code and need some help figuring
: out how to change this call into an S4 generic method.
: 
: In matlab, there's a function called 'repmat' with three
: calling sequences (all I have to deal with anyway):
:     1) B = repmat(A, m, n)
:     2) B = repmat(A, [m n])
:     3) B = repmat(A, n)
: In all cases, A is the fill value, m is number of rows,
: and n is number of columns.
: 
: As separate functions, the translations would roughly be:
: 
: repmat1 <- function(A, m, n) {
:     kronecker(matrix(1, n, m), A)
: }
: 
: repmat2 <- function(A, rc) {
:     repmat1(A, rc[1], rc[2])
: }
: 
: repmat3 <- function(A, n) {
:     repmat1(A, n, n)
: }
: 
: Suggestions?

This can be done easily without any oo dispatch like this:

repmat <- function(A, m, n = if (length(m) == 1)m) {
	m <- c(m, n)
	stopifnot(length(m) == 2)
	kronecker(matrix(1,m[1],m[2]),A)
}