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
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
|
\name{G_axis.default}
\alias{axis.default}
\alias{xscale.components.default}
\alias{yscale.components.default}
\title{Default axis annotation utilities}
\description{
Lattice funtions provide control over how the plot axes are annotated
through a common interface. There are two levels of control. The
\code{xscale.components} and \code{yscale.components} arguments can be
functions that determine tick mark locations and labels given a
packet. For more direct control, the \code{axis} argument can be a
function that actually draws the axes. The functions documented here
are the defaults for these arguments. They can additonally be used as
components of user written replacements.
}
\usage{
xscale.components.default(lim,
packet.number = 0,
packet.list = NULL,
top = TRUE,
\dots)
yscale.components.default(lim,
packet.number = 0,
packet.list = NULL,
right = TRUE,
\dots)
axis.default(side = c("top", "bottom", "left", "right"),
scales, components, as.table,
labels = c("default", "yes", "no"),
ticks = c("default", "yes", "no"),
\dots, prefix)
}
\arguments{
\item{lim}{ the range of the data in that packet (data subset
corresponding to a combination of levels of the conditioning
variable). The range is not necessarily numeric; e.g. for factors,
they could be character vectors representing levels, and for the
various date-time representations, they could be vectors of length 2
with the corresponding class. }
\item{packet.number}{ which packet (counted according to the packet
order, described in \code{\link{print.trellis}}) is being
processed. In cases where all panels have the same limits, this
function is called only once (rather than once for each packet), in
which case this argument will have the value \code{0}.
}
\item{packet.list}{ list, as long as the number of packets, giving
all the actual packets. Specifically, each component is the
list of arguments given to the panel function when and if that
packet is drawn in a panel. (This has not yet been implemented.) }
\item{top, right}{ the value of the \code{top} and \code{right}
components of the result, as appropriate. See below for
interpretation.}
\item{side}{ on which side the axis is to be drawn. The usual partial
matching rules apply. }
\item{scales}{ the appropriate component of the \code{scales} argument
supplied to the high level function, suitably standardized. }
\item{components}{ list, similar to those produced by
\code{xscale.components.default} and
\code{yscale.components.default}. }
\item{as.table}{ the \code{as.table} argument in the high level
function. }
\item{labels}{ whether labels are to be drawn. By default, the rules
determined by \code{scales} are used. }
\item{ticks}{ whether labels are to be drawn. By default, the rules
determined by \code{scales} are used. }
\item{\dots}{ many other arguments may be supplied, and are passed on
to other internal functions. }
\item{prefix}{
A character string identifying the plot being drawn (see
\code{\link{print.trellis}}). Used to retrieve location of current
panel in the overall layout, so that axes can be drawn
appropriately.
}
}
\value{
\code{xscale.components.default} and \code{yscale.components.default}
return a list of the form suitable as the \code{components} argument
of \code{axis.default}. Valid components in the return value of
\code{xscale.components.default} are:
\item{\code{num.limit}}{ A numeric limit for the box. }
\item{\code{bottom}}{ A list with two elements, \code{ticks} and
\code{labels}. \code{ticks} must be a list with components
\code{at} and \code{tck} which give the location and lengths of tick
marks. \code{tck} can be a vector, and will be recycled to be as
long as \code{at}. \code{labels} must be a list with components
\code{at}, \code{labels}, and \code{check.overlap}. \code{at} and
\code{labels} give the location and labels of the tick labels; this
is usually the same as the location of the ticks, but is not
required to be so. \code{check.overlap} is a logical flag
indicating whether overlapping of labels should be avoided by
omitting some of the labels while rendering. }
\item{\code{top}}{ This can be a logical flag; if \code{TRUE},
\code{top} is treated as being the same as \code{bottom}; if
\code{FALSE}, axis annotation for the top axis is omitted.
Alternatively, \code{top} can be a list like \code{bottom}.
}
Valid components in the return value of
\code{yscale.components.default} are \code{left} and \code{right}.
Their interpretations are analogous to (respectively) the
\code{bottom} and \code{top} components described above.
}
\details{
These functions are part of a new API introduced in lattice 0.14 to
provide the user more control over how axis annotation is done. While
the API has been designed in anticipation of use that was previously
unsupported, the implementation has initially focused on reproducing
existing capabilities, rather than test new features. At the time of
writing, several features are unimplemented. If you require them,
please contact the maintainer.
}
\examples{
str(xscale.components.default(c(0, 1)))
set.seed(36872)
rln <- rlnorm(100)
densityplot(rln,
scales = list(x = list(log = 2), alternating = 3),
xlab = "Simulated lognormal variates",
xscale.components = function(...) {
ans <- xscale.components.default(...)
ans$top <- ans$bottom
ans$bottom$labels$labels <- parse(text = ans$bottom$labels$labels)
ans$top$labels$labels <-
if (require(MASS))
fractions(2^(ans$top$labels$at))
else
2^(ans$top$labels$at)
ans
})
## Direct use of axis to show two temperature scales (Celcius and
## Fahrenheit). This does not work for multi-row plots, and doesn't
## do automatic allocation of space
F2C <- function(f) 5 * (f - 32) / 9
C2F <- function(c) 32 + 9 * c / 5
axis.CF <-
function(side, ...)
{
ylim <- current.panel.limits()$ylim
switch(side,
left = {
prettyF <- pretty(ylim)
labF <- parse(text = sprintf("\%s ~ degree * F", prettyF))
panel.axis(side = side, outside = TRUE,
at = prettyF, labels = labF)
},
right = {
prettyC <- pretty(F2C(ylim))
labC <- parse(text = sprintf("\%s ~ degree * C", prettyC))
panel.axis(side = side, outside = TRUE,
at = C2F(prettyC), labels = labC)
},
axis.default(side = side, ...))
}
xyplot(nhtemp ~ time(nhtemp), aspect = "xy", type = "o",
scales = list(y = list(alternating = 3)),
axis = axis.CF, xlab = "Year", ylab = "Temperature",
main = "Yearly temperature in New Haven, CT")
## version using yscale.components
yscale.components.CF <-
function(...)
{
ans <- yscale.components.default(...)
ans$right <- ans$left
ans$left$labels$labels <-
parse(text = sprintf("\%s ~ degree * F", ans$left$labels$at))
prettyC <- pretty(F2C(ans$num.limit))
ans$right$ticks$at <- C2F(prettyC)
ans$right$labels$at <- C2F(prettyC)
ans$right$labels$labels <-
parse(text = sprintf("\%s ~ degree * C", prettyC))
ans
}
xyplot(nhtemp ~ time(nhtemp), aspect = "xy", type = "o",
scales = list(y = list(alternating = 3)),
yscale.components = yscale.components.CF,
xlab = "Year", ylab = "Temperature",
main = "Yearly temperature in New Haven, CT")
}
\seealso{
\code{\link{Lattice}},
\code{\link{xyplot}},
\code{\link{print.trellis}}
}
\author{ Deepayan Sarkar \email{Deepayan.Sarkar@R-project.org}}
\keyword{dplot}
|
