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
|
\name{contents}
\alias{contents}
\alias{contents.data.frame}
\alias{print.contents.data.frame}
\alias{html.contents.data.frame}
\alias{contents.list}
\alias{print.contents.list}
\title{Metadata for a Data Frame}
\description{
\code{contents} is a generic method for which \code{contents.data.frame}
is currently the only method. \code{contents.data.frame} creates an
object containing the following attributes of the variables
from a data frame: names, labels (if any), units (if any), number of
factor levels (if any), factor levels,
class, storage mode, and number of NAs. \code{print.contents.data.frame}
will print the results, with options for sorting the variables.
\code{html.contents.data.frame} creates HTML code for displaying the
results. This code has hyperlinks so that if the user clicks on the
number of levels the browser jumps to the correct part of a table of
factor levels for all the \code{factor} variables. If long labels are
present (\code{"longlabel"} attributes on variables), these are printed
at the bottom and the \code{html} method links to them through the
regular labels. Variables having the same \code{levels} in the same
order have the levels factored out for brevity.
\code{contents.list} prints a directory of datasets when
\code{\link{sasxport.get}} imported more than one SAS dataset.
If \code{options(prType='html')} is in effect, calling \code{print} on
an object that is the contents of a data frame will result in
rendering the HTML version. If run from the console a browser window
will open.
}
\usage{
contents(object, \dots)
\method{contents}{data.frame}(object, sortlevels=FALSE, id=NULL,
range=NULL, values=NULL, \dots)
\method{print}{contents.data.frame}(x,
sort=c('none','names','labels','NAs'), prlevels=TRUE, maxlevels=Inf,
number=FALSE, \dots)
\method{html}{contents.data.frame}(object,
sort=c('none','names','labels','NAs'), prlevels=TRUE, maxlevels=Inf,
levelType=c('list','table'),
number=FALSE, nshow=TRUE, \dots)
\method{contents}{list}(object, dslabels, \dots)
\method{print}{contents.list}(x,
sort=c('none','names','labels','NAs','vars'), \dots)
}
\arguments{
\item{object}{
a data frame. For \code{html} is an object created by
\code{contents}. For \code{contents.list} is a list of data frames.
}
\item{sortlevels}{set to \code{TRUE} to sort levels of all factor
variables into alphabetic order. This is especially useful when two
variables use the same levels but in different orders. They will
still be recognized by the \code{html} method as having identical
levels if sorted.}
\item{id}{an optional subject ID variable name that if present in
\code{object} will cause the number of unique IDs to be printed in
the contents header}
\item{range}{an optional variable name that if present in \code{object}
will cause its range to be printed in the contents header}
\item{values}{an optional variable name that if present in
\code{object} will cause its unique values to be printed in the
contents header}
\item{x}{
an object created by \code{contents}
}
\item{sort}{
Default is to print the variables in their original order in the
data frame. Specify one of
\code{"names"}, \code{"labels"}, or \code{"NAs"} to sort the variables by,
respectively, alphabetically by names, alphabetically by labels, or by
increaseing order of number of missing values. For
\code{contents.list}, \code{sort} may also be the value
\code{"vars"} to cause sorting by the number of variables in the dataset.
}
\item{prlevels}{
set to \code{FALSE} to not print all levels of \code{factor} variables
}
\item{maxlevels}{maximum number of levels to print for a \code{factor} variable}
\item{number}{
set to \code{TRUE} to have the \code{print} and \code{latex} methods
number the variables by their order in the data frame
}
\item{nshow}{set to \code{FALSE} to suppress outputting number of
observations and number of \code{NA}s; useful when these counts
would unblind information to blinded reviewers}
\item{levelType}{
By default, bullet lists of category levels are
constructed in html. Set \code{levelType='table'} to put levels in
html table format.
}
\item{\dots}{
arguments passed from \code{html} to \code{format.df},
unused otherwise
}
\item{dslabels}{
named vector of SAS dataset labels, created for
example by \code{\link{sasdsLabels}}
}
}
\value{
an object of class \code{"contents.data.frame"} or
\code{"contents.list"}. For the \code{html} method is an \code{html}
character vector object.
}
\author{
Frank Harrell
\cr
Vanderbilt University
\cr
\email{fh@fharrell.com}
}
\seealso{
\code{\link{describe}}, \code{\link{html}}, \code{\link{upData}},
\code{\link{extractlabs}}, \code{\link{hlab}}
}
\examples{
set.seed(1)
dfr <- data.frame(x=rnorm(400),y=sample(c('male','female'),400,TRUE),
stringsAsFactors=TRUE)
contents(dfr)
dfr <- upData(dfr, labels=c(x='Label for x', y='Label for y'))
attr(dfr$x, 'longlabel') <-
'A very long label for x that can continue onto multiple long lines of text'
k <- contents(dfr)
print(k, sort='names', prlevels=FALSE)
\dontrun{
html(k)
html(contents(dfr)) # same result
latex(k$contents) # latex.default just the main information
}
}
\keyword{data}
\keyword{interface}
\concept{html}
|
