The R Project SVN R

Rev

Details | Last modification | View Log | RSS feed

Rev Author Line No. Line
25384 ripley 1
\name{codes-deprecated}
24604 ripley 2
\alias{codes-deprecated}
15168 pd 3
\title{Factor Codes}
6098 pd 4
\description{
15168 pd 5
  This (generic) function returns a numeric coding of a factor.  It can
6
  also be used to assign to a factor using the coded form.
24604 ripley 7
 
8
  It is now \code{\link{Deprecated}}.
6098 pd 9
}
15168 pd 10
\usage{
11
codes(x, \dots)
24158 hornik 12
codes(x, \dots) <- value
15168 pd 13
}
15518 ripley 14
\arguments{
15
  \item{x}{an object from which to extract or set the codes.}
16
  \item{\dots}{further arguments passed to or from other methods.}
24604 ripley 17
  \item{value}{replacement value.}
15518 ripley 18
}
6098 pd 19
\value{
15168 pd 20
  For an ordered factor, it returns the internal coding (1 for the
21
  lowest group, 2 for the second lowest, etc.).
2016 pd 22
 
15168 pd 23
  For an unordered factor, an alphabetical ordering of the levels is
24
  assumed, i.e., the level that is coded 1 is the one whose name is
25
  sorted first according to the prevailing collating sequence.
26
  \bold{Warning:} the sort order may well depend on the locale, and
27
  should not be assumed to be ASCII.
6098 pd 28
}
29
\note{
15168 pd 30
  Normally \code{codes} is not the appropriate function to use with an
31
  unordered factor.  Use \code{\link{unclass}} or
32
  \code{\link{as.numeric}} to extract the codes used in the internal
33
  representation of the factor, as these do not assume that the codes
34
  are sorted.
35
 
36
  The behaviour for unordered factors is dubious, but compatible with S
37
  version 3.  To get the internal coding of a factor, use
38
  \code{\link{as.integer}}.  Note in particular that the codes may not
39
  be the same in different language locales because of collating
40
  differences.
6098 pd 41
}
2 r 42
\seealso{
2016 pd 43
  \code{\link{factor}}, \code{\link{levels}}, \code{\link{nlevels}}.
2 r 44
}
24604 ripley 45
\examples{\dontrun{
2016 pd 46
codes(rep(factor(c(20,10)),3))
47
 
1055 maechler 48
x <- gl(3,5)
2 r 49
codes(x)[3] <- 2
2016 pd 50
x
51
 
52
data(esoph)
53
( ag <- esoph$alcgp[12:1] )
54
codes(ag)
55
 
56
codes(factor(1:10)) # BEWARE!
24604 ripley 57
}}
286 maechler 58
\keyword{category}
59
\keyword{classes}
2016 pd 60
 
61