Environments as Objects

Part of the Object Oriented Programming section of Coddy's R journey. Lesson 18 of 57.

An environment is a set of named values, like a list, created with new.env(). The difference is that an environment is never copied: every variable that refers to it sees the same object, so a change through one name is visible through all of them:

e <- new.env()
e$count <- 1
alias <- e
alias$count <- 99
print(e$count)
l <- list(count = 1)
copy <- l
copy$count <- 99
print(l$count)

Output:

[1] 99
[1] 1

A function that receives an environment changes the caller's object. It does not have to return anything, and the caller does not assign a result:

deposit <- function(acc, x) acc$balance <- acc$balance + x
acc <- new.env()
acc$balance <- 10
deposit(acc, 5)
deposit(acc, 5)
print(acc$balance)

Output:

[1] 20

assign(), get() and exists() work with a name held in a variable, and ls() lists the names in an environment in sorted order:

e <- new.env()
for (fruit in c("pear", "apple")) assign(fruit, nchar(fruit), envir = e)
print(ls(e))
print(get("pear", envir = e))
print(exists("kiwi", envir = e))

Output:

[1] "apple" "pear" 
[1] 4
[1] FALSE

An environment with a class attribute is an S3 object with reference semantics: its methods work as usual, and functions that change it do so in place:

new_wallet <- function(coins = 0) {
  w <- new.env()
  w$coins <- coins
  class(w) <- "wallet"
  w
}
print.wallet <- function(x, ...) {
  cat("wallet with", x$coins, "coins\n")
  invisible(x)
}
spend <- function(w, n) w$coins <- w$coins - n
w <- new_wallet(10)
spend(w, 3)
w

Output:

wallet with 7 coins

Challenge

Easy

An inventory is an environment of class inventory with one variable, items, a named numeric vector of stock. Complete Inventory.R:

new_inventory() creates the environment with an empty items (c() is fine) and the class.
add_stock(inv, item, n) adds n to the item's stock, starting from 0 for a new item.
take_stock(inv, item, n) removes n and returns TRUE when there is enough, otherwise changes nothing and returns FALSE.
stock_of(inv, item) returns the stock, 0 for an unknown item.

The functions change the inventory in place and do not return it. The supplied code reads commands add apple 5 and take apple 2, prints the result of each take, then every item in alphabetical order.

Your code goes in Inventory.R. main.R holds the supplied input/output code and cannot be edited.

Try it yourself

source("Inventory.R")

# Supplied input/output code: keep it as it is
input <- suppressWarnings(readLines(file("stdin")))
inv <- new_inventory()
for (line in input) {
  p <- strsplit(line, " ")[[1]]
  n <- as.numeric(p[3])
  if (p[1] == "add") add_stock(inv, p[2], n)
  if (p[1] == "take") cat("take ", p[2], ": ", if (take_stock(inv, p[2], n)) "ok" else "not enough", "\n", sep = "")
}
for (item in sort(names(inv$items))) cat(item, ": ", stock_of(inv, item), "\n", sep = "")