section_fun: Section a function and set default values in function

In doBy: Groupwise Statistics, LSmeans, Linear Estimates, Utilities

Section a function and set default values in function

Description

Section a functions domain by fixing certain arguments of a function call.

Usage

set_default(fun, nms, vls = NULL)

section_fun(fun, nms, vls = NULL, method = "args")

section_fun_sub(fun, nms, vls = NULL, envir = parent.frame())

section_fun_env(fun, nms, vls = NULL)

get_section(object)

get_fun(object)

Arguments

fun	Function to be sectioned
nms	Either a named list of the form name=value where each name is the name of an argument of the function (in which case vls is ignored) or a character vector of names of arguments.
vls	A vector or list of values of the arguments
method	One of the following: 1) "args" (default); based on substituting fixed values into the function argument list as default values). For backward compatibility can also be "def". 2) "body" for substituting fixed values into the function body. For backward compatibility can also be "sub". 3) "env": (for environment); using an auxillary argument for storing sectioned values.
envir	Environment
object	An object from section_fun (a scaffold object).

Details

Let E be a subset of the cartesian product X x Y where X and Y are some sets. Consider a function f(x,y) defined on E. Then for any x in X, the section of E defined by x (denoted Ex) is the set of $y$s in Y such that (x, y) is in E. Correspondingly, the section of f(x,y) defined by x is the function $f_x$ defined on Ex given by $f_x(y)=f(x,y)$.

section_fun is a wrapper for calling set_default (default method), section_fun_env or section_fun_sub. Notice that creating a sectioned function with section_fun_sub can be time consuming.

Value

A new function: The input function fun but with certain arguments fixed at specific values.

Author(s)

Søren Højsgaard, [e-mail] based on code adapted from the curry package.

See Also

bquote_fun_list()

Examples

f  <- function(x, y){x + y}

f_ <- section_fun(f, list(y = 10),    method="args") ## "def"" is default
f_ <- section_fun(f, nms="y", vls=10, method="args") ## SAME AS ABOVE
f_
f_(x=1)

f_ <- section_fun(f, list(y = 10),    method="body") ## 
f_ <- section_fun(f, nms="y", vls=10, method="body") ## SAME AS ABOVE
f_
f_(x=1)

f_ <- section_fun(f, list(y = 10),    method="env")
f_ <- section_fun(f, nms="y", vls=10, method="env") ## SAME AS ABOVE
f_
f_(x=1)
get_section(f_)
get_fun(f_)

 
## With more complicated values:
g <- function(A, B) {
  A + B
}
g_ <- section_fun(g, list(A = matrix(1:4, nrow=2)))
g_ <- section_fun(g, "A", list(matrix(1:4, nrow=2)))
g_(diag(1, 2))

g_ <- section_fun(g, list(A = matrix(1:4, nrow=2)))

## Using built in function
set.seed(123)
rnorm5 <- section_fun(rnorm, list(n=5)) 
rnorm5(0, 1)

set.seed(123)
rnorm(5)

doBy documentation built on Oct. 8, 2024, 1:06 a.m.