qq: Quasiquotation. Perform template substitutions on a quoted R...

In crowding/vadr: Macros, metaprogramming, and pleasant programming idioms.

Description

This is an extended version of the bquote utility. 'qq' quotes its first argument, then scans for terms wrapped in .(), ..(), or names that match `.()` The wrapped expressions or names are evaluated in the given environment. Expressions wrapped in ..() will be interpolated into the argument list in which they occur. Names wrapped in '.()' will be substituted and coerced to name.

Usage

qq(...)

Arguments

expr	An expression, left unevaluated.

Details

Invocations of qq() can be nested within the .() sections and they should work as promised.

Value

For qq, A language object; for qe, evaluates the expression in the calling environment.

Author(s)

Peter Meilstrup

See Also

macro bquote substitute quoting.env

Examples

#Basic substitution
default <- 1
qq( function(x, y = .(default)) x+y )
#function(x, y = 1) x + y

# splicing substitution:
paste.before <- alist("hello", "cool")
paste.after <- alist("!", "Now is", date())
qq(cat(...(paste.before), "world", ...(paste.after), '\n'))
#cat("hello", "cool", "world", "!", "Now is", date(), "\n")

# Name substitution:
element_to_access <- "x"
qq(function(data) data$`.(element_to_access)`)
#function(data) data$x

argument.name <- "x"
qq(
  function(`.(argument.name)`)
  cat(.(argument.name), " is ", `.(argument.name)`, "\n")
)
#function(x) cat("x", " is ", x, "\n"))

# Note that in the argument list to a function, function argument
# names are names, and defaults are values; that is
function(x=1, y) x+y
# is equivalent to
function(.=...(alist(x=1, y=))) x+y
# or
function(.=...(list(x=1, y=missing_value()))) x+y

# Building a function with an arbitrary list of arguments:
argnames <- letters[1:4]
qq(function(.=...(put(missing_value(length(argnames)), names, argnames))) {
  list(...(lapply(argnames, as.name)))
})
#function(a, b, c, d) list(a, b, c, d)

# The poor .() function is overloaded. Usually can escape it by adding
# parens:
dfname <- "baseball"
qq(ddply(.(as.name(dfname)), (.)(id, team), identity))
#ddply(baseball, .(id.team), identity)

crowding/vadr documentation built on May 14, 2019, 11:33 a.m.