class() on substitute(...) output?
On Jan 3, 2014, at 9:46 AM, Hadley Wickham <h.wickham at gmail.com> wrote:
On Thu, Jan 2, 2014 at 7:16 PM, Henrik Bengtsson <hb at biostat.ucsf.edu> wrote:
Does it make sense to talk about the class of the output of
substitute(...)? I'm puzzled by the following outputs:
ee <- list(
A = substitute( a <- 1 ),
B = substitute({ a <- 1 }),
C = substitute(( a <- 1 )),
D = substitute( a == 1 )
)
t(sapply(ee, FUN=function(e) { c(typeof=typeof(e), mode=mode(e), class=class(e)) }))
typeof mode class
A "language" "call" "<-"
B "language" "call" "{"
C "language" "(" "("
D "language" "call" "call"
That the mode in C is "(", is motivated in help("mode"): "that some
calls have mode "(" which is S compatible." However, what's the
explanation for the different classes? Is that intended or just
"garbage" output?
?class has: "Many R objects have a class attribute, a character vector giving the names of the classes from which the object inherits. If the object does not have a class attribute, it has an implicit class, "matrix", "array" or the result of mode(x) (except that integer vectors have implicit class "integer"). (Functions oldClass and oldClass<- get and set the attribute, which can also be done directly.)" which suggests either a bug or some tweaks are needed to the documentation.
Definitely not the former. What happens is easy enough to see, if you follow the definition of the .Primitive. Routine R_data_class in attrib.c does the computation. When there is no class attribute, it supplies the definition for "matrix" and "array", turns the 3 types for functions into "function", and distinguishes among objects of type "language" to give specific classes to the syntactic functions called, such as `while`(), `if`(), etc. It does seem that the documentation should say something similar. John
Is there any point in ever using mode() except for S+ compatibility? It just adds some confusing aliases on top of typeof. Hadley -- http://had.co.nz/
______________________________________________ R-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel