Is there an easier way to access attributes of a class in R, can I use dot notation?

Question

I have created an object in R that contains several attributes. How can I easily access them?

I can do:

attr(x, attributeName)

or:

attributes(x)$attributeName

but none of them is really convenient.

Is there a quicker way (like the dot in C++ or Java)?

Solution 1

Don't use attributes for your object, use a list:

myobj <- structure(list(a = 1, b = 2), class = "myclass")
print.myclass <- function(x, ...) cat("A: ", x$a, " B: ", x$b, "\n", sep = "")
myobj

Of course, this might not work if you're extending an existing object (e.g. vector), but in my experience it's a generally better way to build objects.

Solution 2

attributes() returns a named list. I'd call it once and store them, then access via names. There is no point repeatedly calling either attr() or attributes() if you don't have to.

x <- 1:10
attr(x, "foo") <- "a"
attr(x, "bar") <- "b"
(features <- attributes(x))

which gives:

R> (features <- attributes(x))
$foo
[1] "a"

$bar
[1] "b"

then access in the usual way

R> features["foo"]
$foo
[1] "a"

R> features$foo
[1] "a"

Solution 3

probably there is no built-in function that is counter part of . in C++, but you can define it like this:

> `%.%` <- function(o, a) attr(o, as.character(substitute(a)))
> x <- 1
> attr(x, "orz") <- 2
> x%.%orz
[1] 2

Solution 4

Example of using the match.length attribute that is returned from regexpr:

Three strings in a vector, first and third include an embedded string:

data=c("<a href=\"ch4.html\">Chapter 1</a>",
       "no quoted string is embedded in this string",
       "<a   href=\"appendix.html\">Appendix</a>")

Use regexpr to locate the embedded strings:

> locations <- regexpr("\"(.*?)\"", data)

Matches are in the first string (at 9 with length 10) and third string (at 11 with length 15):

> locations
[1]  9 -1 11
attr(,"match.length")
[1] 10 -1 15
attr(,"useBytes")
[1] TRUE

Vector from the attribute:

> attr(locations,"match.length")
[1] 10 -1 15

Use substr and the attribute vector to extract the strings:

> quoted_strings=substr( data, 
                         locations, 
                         locations+attr(locations,"match.length")-1 )    
> quoted_strings
[1] "\"ch4.html\""      ""                  "\"appendix.html\""

Maybe you'd like to remove the embedded quote characters from your strings:

> gsub("\"", "", quoted_strings)
[1] "ch4.html"      ""              "appendix.html"

An alternative is to use regmatches:

> regmatches(data,locations)
[1] "\"ch4.html\""      "\"appendix.html\""

View more solutions