Vector unpack assignment syntax for R

The ‘unpack’ package allows unpacking multiple values from a vector or list in a single assignment expression. Unlike other packages, this package does not require using a new operator or other, “convoluted” means.

Usage

For example, we can unpack a specific, known number of variables:

library(unpack)

c[a, b, c] = c('foo', 'bar', 'baz')

message(a) # foo
message(b) # bar
message(c) # baz

This looks like a subset assignment to a multi-dimensional array, c. But no such array exists. The sole purpose of this syntax is to enable compound assignment. The c[…] = values syntax intentionally mimics the inverse syntax to create a vector from multiple values, i.e. values = c(…).

And even though the ‘unpack’ package exports the name c to enable this, you can still use the c function from base R. Both work side by side.

We can also specify that we want to unpack a fixed number of items, and keep the rest inside a vector:

c[first, second, rest[]] = month.name

message(first)          # January
message(second)         # February
message(toString(rest)) # March, April, May, …

Great!

And maybe we only cared about February, and want to ignore the rest:

c[., feb, .[]] = month.name
message(feb) # February

Installation

if (! requireNamespace('pak')) {
    install.packages("pak", repos = "https://r-lib.github.io/p/pak/dev/")
}

pak::pkg_install('klmr/unpack@main')