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
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
|
\name{extract_array}
\alias{extract_array}
\alias{extract_array,ANY-method}
\alias{extract_array,array-method}
\alias{extract_array,data.frame-method}
\alias{extract_array,DataFrame-method}
\alias{as.array.Array}
\alias{as.array,Array-method}
\alias{as.data.frame.Array}
\alias{as.data.frame,Array-method}
\alias{as.vector.Array}
\alias{as.vector,Array-method}
\alias{as.logical.Array}
\alias{as.logical,Array-method}
\alias{as.integer.Array}
\alias{as.integer,Array-method}
\alias{as.numeric.Array}
\alias{as.numeric,Array-method}
\alias{as.complex.Array}
\alias{as.complex,Array-method}
\alias{as.character.Array}
\alias{as.character,Array-method}
\alias{as.raw.Array}
\alias{as.raw,Array-method}
\title{extract_array}
\description{
\code{extract_array} is an internal generic function not intended to be
used directly by the end user. It has methods defined for array, data.frame,
\link[S4Vectors]{DataFrame} objects, and other array-like objects.
Note that \code{extract_array} is part of the \emph{seed contract} as
defined in the \emph{Implementing A DelayedArray Backend} vignette from
the \pkg{DelayedArray} package.
}
\usage{
## The extract_array() S4 generic:
extract_array(x, index)
## extract_array() methods defined in the S4Arrays package:
\S4method{extract_array}{ANY}(x, index)
\S4method{extract_array}{array}(x, index)
\S4method{extract_array}{data.frame}(x, index)
\S4method{extract_array}{DataFrame}(x, index)
}
\arguments{
\item{x}{
An array-like object.
This can be an ordinary array, a \link[SparseArray]{SparseArray}
object from the \pkg{SparseArray} package, a \linkS4class{dgCMatrix}
object from the \pkg{Matrix} package, a \link[DelayedArray]{DelayedArray}
object from the \pkg{DelayedArray} package, or any object with an array
semantic (i.e. an object for which \code{dim(x)} is not NULL).
Note that data.frame and \link[S4Vectors]{DataFrame} objects are also
supported.
}
\item{index}{
An unnamed list of integer vectors, one per dimension in \code{x}.
Each vector is called a \emph{subscript} and can only contain
positive integers that are valid 1-based indices along the corresponding
dimension in \code{x}.
Empty or missing subscripts are allowed. They must be represented
by list elements set to \code{integer(0)} or \code{NULL}, respectively.
The subscripts cannot contain NAs or non-positive values.
Individual subscripts are allowed to contain duplicated indices.
}
}
\details{
\code{extract_array()} methods need to support empty or missing subscripts.
For example, if \code{x} is an M x N matrix-like object, then
\code{extract_array(x, list(NULL, integer(0)))} must return an M x 0
ordinary matrix, and \code{extract_array(x, list(integer(0), integer(0)))}
a 0 x 0 ordinary matrix.
Also subscripts are allowed to contain duplicated indices so things like
\code{extract_array(x, list(c(1:3, 3:1), 2L))} need to be supported.
Finally, for maximum efficiency, \code{extract_array()} methods
should not try to do anything with the dimnames on \code{x}.
}
\value{
An \emph{ordinary} array of the same \code{type()} as \code{x}.
For example, if \code{x} is an object representing an M x N matrix
of complex numbers (i.e. \code{type(x) == "complex"}), then
\code{extract_array(x, list(NULL, 2L))} must return the 2nd column
in \code{x} as an M x 1 \emph{ordinary} matrix of \code{type()}
\code{"complex"}.
}
\seealso{
\itemize{
\item \code{S4Arrays::\link[S4Arrays]{type}} to get the type of the
elements of an array-like object.
\item \link[base]{array} and \link[base]{data.frame} objects in base R.
\item \link[SparseArray]{SparseArray} objects implemented in the
\pkg{SparseArray} package.
\item \link[DelayedArray]{DelayedArray} objects implemented in the
\pkg{DelayedArray} package.
\item \link[S4Vectors]{DataFrame} objects implemented in the
\pkg{S4Vectors} package.
}
}
\examples{
extract_array
showMethods("extract_array")
## extract_array() works on array-like objects like SparseArray objects,
## dgCMatrix objects, DataFrame objects, etc...
## --- On a SparseArray object ---
library(SparseArray)
a <- array(0L, 5:3)
a[c(1:2, 8, 10, 15:17, 20, 24, 40, 56:60)] <- (1:15)*10L
svt <- as(a, "SparseArray")
svt
extract_array(svt, list(NULL, c(4L,2L,4L), 1L))
extract_array(svt, list(NULL, c(4L,2L,4L), 2:3))
extract_array(svt, list(NULL, c(4L,2L,4L), integer(0)))
## Sanity checks:
stopifnot(
identical(extract_array(svt, list(NULL, c(4L,2L,4L), 1L)),
as.array(svt)[ , c(4L,2L,4L), 1L, drop=FALSE]),
identical(extract_array(svt, list(NULL, c(4L,2L,4L), 2:3)),
as.array(svt)[ , c(4L,2L,4L), 2:3]),
identical(extract_array(svt, list(NULL, c(4L,2L,4L), integer(0))),
as.array(svt)[ , c(4L,2L,4L), integer(0)])
)
## --- On a dgCMatrix object ---
library(Matrix)
m <- a[ , , 1]
dgcm <- as(m, "dgCMatrix")
dgcm
extract_array(dgcm, list(NULL, c(4L,2L,4L)))
## Sanity check:
stopifnot(
identical(extract_array(dgcm, list(NULL, c(4L,2L,4L))),
as.matrix(dgcm)[ , c(4L,2L,4L)])
)
## --- On a data.frame or DataFrame object ---
df <- data.frame(a=44:49, b=letters[1:6], c=c(TRUE, FALSE))
DF <- as(df, "DataFrame")
extract_array(df, list(4:2, c(1L,3L)))
extract_array(DF, list(4:2, c(1L,3L)))
## Sanity check:
target <- as.matrix(df)[4:2, c(1L,3L)]
dimnames(target) <- NULL
stopifnot(
identical(extract_array(df, list(4:2, c(1L,3L))), target),
identical(extract_array(DF, list(4:2, c(1L,3L))), target)
)
}
\keyword{internal}
\keyword{array}
|
