1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
|
\name{data.set manipulation}
\alias{subset.data.set}
\alias{unique,data.set-method}
\alias{merge,data.set,data.set-method}
\alias{merge,data.set,data.frame-method}
\alias{merge,data.frame,data.set-method}
\alias{cbind.data.set}
\alias{rbind.data.set}
\title{Manipulation of Data Sets}
\description{
Like data frames, \code{data.set} objects have
\code{\link[base]{subset}}, \code{\link[base]{unique}},
\code{\link[base]{cbind}}, \code{\link[base:cbind]{rbind}},
\code{\link[base]{merge}} methods defined for them.
The semantics are basically the same as the methods defined
for data frames in the \code{base} package, with the only difference
that the return values are \code{data.set} objects.
In fact, the methods described here are front-ends to the
corresponding methods for data frames, which are constructed
such that the "extra" information attached to variables within
\code{data.set} objects, that is, to \code{item} objects.
}
\usage{
\method{subset}{data.set}(x, subset, select, drop = FALSE, \dots)
\S4method{unique}{data.set}(x, incomparables = FALSE, \dots)
\S3method{cbind}{data.set}(\dots, deparse.level = 1)
\S3method{rbind}{data.set}(\dots, deparse.level = 1)
\S4method{merge}{data.set,data.set}(x,y, \dots)
\S4method{merge}{data.set,data.frame}(x,y, \dots)
\S4method{merge}{data.frame,data.set}(x,y, \dots)
}
\arguments{
\item{x,y}{\code{data.set} objects. On of the arguments to
\code{merge} may also be an object coercable into a data frame
and the result still is a \code{data.set} object.
}
\item{subset}{a logical expression, used to select observations from
the data set.}
\item{select}{a vector with variablen names, which are retained in the
data subset.}
\item{drop}{logical; if \code{TRUE} and the result has only one
column, the result is an item and not a data set.}
\item{\dots}{for \code{subset}: a logical vector
of the same length as the number of rows of the \code{data.set}
and, optionally, a vector of variable names (tagged as \code{select});
for \code{unique}: further arguments, ignored;
for \code{cbind}, \code{rbind}: objects coercable
into data frames, with at least one being a \code{data.set}
object;
for \code{merge}: further arguments
such as arguments tagged with \code{by}, \code{by.x}, \code{by.y},
etc. that specify the variables by which to merge
the data sets of data frames \code{x} and \code{y}.
}
\item{incomparables}{a vector of values that cannot be compared. See
\code{\link[base]{unique}}.
}
\item{deparse.level}{an argument retained for
reasons of compatibility of the default methods
of \code{\link[base]{cbind}} and \code{\link[base:cbind]{rbind}}.
}
}
\examples{
ds1 <- data.set(
a = rep(1:3,5),
b = rep(1:5,each=3)
)
ds2 <- data.set(
a = c(3:1,3,3),
b = 1:5
)
ds1 <- within(ds1,{
description(a) <- "Example variable 'a'"
description(b) <- "Example variable 'b'"
})
ds2 <- within(ds2,{
description(a) <- "Example variable 'a'"
description(b) <- "Example variable 'b'"
})
str(ds3 <- rbind(ds1,ds2))
description(ds3)
ds3 <- within(ds1,{
c <- a
d <- b
description(c) <- "Copy of variable 'a'"
description(d) <- "Copy of variable 'b'"
rm(a,b)
})
str(ds4 <- cbind(ds1,ds3))
description(ds4)
ds5 <- data.set(
c = 1:3,
d = c(1,1,2)
)
ds5 <- within(ds5,{
description(c) <- "Example variable 'c'"
description(d) <- "Example variable 'd'"
})
str(ds6 <- merge(ds1,ds5,by.x="a",by.y="c"))
# Note that the attributes of the left-hand variables
# have priority.
description(ds6)
}
|
