Skip to content

R-alpha: R-0.50-a3(+) Method despatching bug ?

2 messages · Martin Maechler, Thomas Lumley

Original
Martin Maechler
# 
It is very wierd...  Can some of you confirm the following behavior ?
It is a new bug (feature ?)  which was not yet in 0.49 ...




noquote <- function(obj) {
	## constructor for a useful "minor" class
	if(!inherits(obj,"noquote")) class(obj) <- c(class(obj),"noquote")
	obj
}
"[.noquote" <- function (x, subs) structure(unclass(x)[subs], class = "noquote")

print.noquote <- function(obj,...) {
	## method for (character) objects of class 'noquote'
	cl <- class(obj)
	class(obj) <- cl[cl != "noquote"]
	NextMethod("print", obj, quote = FALSE, ...)
}

nq <- noquote(letters)

##-- the next two should be equivalent, but are not ...?..
print(nq)
nq
[1] a b c d e f g h i j k l m n o p q r s t u v w x y z
Error: couldn't find function "print.noquote"
[1] "a" "b" "c" "d" "e" "f" "g" "h" "i" "j" "k" "l" "m" "n" "o" "p" "q" "r" "s"
[20] "t" "u" "v" "w" "x" "y" "z"
Error: couldn't find function "print.noquote"
[1] "[.noquote"     "noquote"       "nq"            "print.noquote"
function (obj, ...) 
{
        ## method for (character) objects of class 'noquote'
        cl <- class(obj)
        class(obj) <- cl[cl != "noquote"]
        NextMethod("print", obj, quote = FALSE, ...)
}
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
r-devel 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-devel-request@stat.math.ethz.ch
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Thomas Lumley
#
On Wed, 20 Aug 1997, Martin Maechler wrote:

            

      
      
      
    

        
I'm not sure that you can expect NextMethod to work if you fiddle with the
class list...

However, there is still something wrong with method dispatching, since
print.noquote <- function (obj, ...)  NextMethod(quote = FALSE,...)
doesn't work either.  The error occurs when R tries for some reason to
make a recursive call to print.noquote.

Using debug() also shows the interesting fact that
R> nq 
calls print() with its arguments evaluated
------
debugging in: print(structure(c("a", "b", "c", "d", "e", "f", "g", "h",
"i", 
"j", "k", "l", "m", "n", "o", "p", "q", "r", "s", "t", "u", "v", 
"w", "x", "y", "z"), class = "noquote"))
debugging in: print.noquote(structure(c("a", "b", "c", "d", "e", "f", "g", 
"h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r", "s", "t", 
"u", "v", "w", "x", "y", "z"), class = "noquote"))
Error: couldn't find function "print.noquote"
-------
but
----------
R> print(nq)
 print(nq)
debugging in: print(nq)
debugging in: print.noquote(nq)
 [1] a b c d e f g h i j k l m n o p q r s t u v w x y z
attr(,"class")
[1] noquote
exiting from: print.noquote(nq)
exiting from: print(nq)
----------
doesn't evaluate the argument first.




Thomas Lumley
------------------------------------------------------+------
Biostatistics		: "Never attribute to malice what  :
Uni of Washington	:  can be adequately explained by  :
Box 357232		:  incompetence" - Hanlon's Razor  :
Seattle WA 98195-7232	:				   :
------------------------------------------------------------



=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
r-devel 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-devel-request@stat.math.ethz.ch
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-