In Scala 3.3, a map is the core dictionary type. We link values from keys. Usually we have Strings and Ints, but any type can be used.
Scala maps are immutable—we cannot modify them once created. We can perform lookups, and loop over maps with foreach
.
Here we build a String
map—it has String
keys and Int
values. We initialize it with just two pairs. Map is immutable so a new map would be needed to add elements.
object Program { def main(args: Array[String]): Unit = { // Use simple map initialization syntax. val weights = Map("cat" -> 10, "elephant" -> 200000) // Look up animal weights. val weight = weights("elephant") println(weight) } }200000
We can initialize a Map with a different syntax based on a list of pairs. Here we use a String
to String
map. We look up two animal Strings.
object Program { def main(args: Array[String]): Unit = { // Create map of animals to colors. // ... Has string keys and string values. val colors = Map(("bird", "blue"), ("fox", "red")) // Get value for this key. val result1 = colors("bird") println(result1) val result2 = colors("fox") println(result2) } }blue red
A Map is immutable so we cannot add an element to an existing map. But we can add a key-value pair to a new, copied map. Here we add a zebra in a new map creation statement.
object Program { def main(args: Array[String]): Unit = { // Create an immutable map. val zoo = Map("frog" -> 1, "lion" -> 1) // Add a pair to the map. // ... This creates a new map. val all = zoo + ("zebra" -> 1) println(all) } }Map(frog -> 1, lion -> 1, zebra -> 1)
Keys
, valuesHere we create a map with String
keys and Int
values. We then access the keys and values properties to get collections of those elements.
foreach
function on the keys collection (which is an iterable
) and use println
on each key.for
-loop to call println
on each value in the Map.object Program { def main(args: Array[String]): Unit = { // Create map of String keys and Int values. val ids = Map(("abc", 10), ("def", 20)) // Use foreach function to print each key. ids.keys.foreach(println(_)) // Use for-loop to iterate over all values. for (value <- ids.values) { println(value) } } }abc def 10 20
getOrElse
With get, an option is returned. If the value exists in the Map, isDefined
will return true on that option. We can then call get()
on the option.
Get()
returns an option, not the value type of the Map. We must always test the option.GetOrElse()
lets us provide a default value that is returned if the key does not exist in the Map. Here we use 0 as a default.object Program { def main(args: Array[String]): Unit = { // Create a String, Int map. val sizes = Map(("Medium", 2), ("Large", 4)) // This key does not exist, so the option is not defined. val result1 = sizes.get("Small") if (!result1.isDefined) { println("Not defined") } // This key exists. // ... Get and print the option's internal value. val result2 = sizes.get("Large") if (result2.isDefined) { println(result2) val number = result2.get println(number) } // Use a default value if the key does not exist. val result3 = sizes.getOrElse("Small", 0) if (result3 == 0) { println("Zero, else value") } } }Not defined Some(4) 4 Zero, else value
Equals
With this method we test for structural equality of two maps. The order of the entries in the maps is not important. But the keys and values must be the same.
map1
and map2
are equal because their key-value pairs are the same. But map3
has a different value so it is not equal.object Program { def main(args: Array[String]): Unit = { // Create three maps. // ... The first two are equal but have different orders. // The third one has different entries. val map1 = Map((10, true), (20, false)) val map2 = Map((20, false), (10, true)) val map3 = Map((20, true), (10, true)) // These two maps are structurally equal. if (map1.equals(map2)) { println("Maps equal") } // Not equal. if (!map1.equals(map3)) { println("Maps not equal") } } }Maps equal Maps not equal
WithDefaultValue
A map can have a default value or function. We call withDefaultValue
to specify a default value—in this program we use a default integer of -1.
WithDefaultValue()
returns a default value based on a function. We can pass a lambda expression to it.object Program { def main(args: Array[String]): Unit = { val animalMap = Map(("cat", 10), ("bird", 5)) // The default value is now negative 1. val animalMapDefault = animalMap.withDefaultValue(-1); // Use map with default. val result1 = animalMapDefault("cat") println(result1) val result2 = animalMapDefault("bear") println(result2) } }10 -1
A Map is a good optimization. Here we have three elements "abc" and we store them in a Map and a List
. We search for one element in the collections.
string
argument. We repeat this operation in a for
-loop.List
. We use a for
-loop to search. The operation is repeated many times.List
. The fast lookup time helps this benchmark.object Program { def main(args: Array[String]): Unit = { // Map and list used. val lookup = Map(("a", 1), ("b", 1), ("c", 1)) val list = List("a", "b", "c") var total = 0 val t1 = System.currentTimeMillis() // Version 1: lookup element in a map. for (i <- 0 until 10000000) { val result = lookup("c") total += result } val t2 = System.currentTimeMillis() // Version 2: search for element in a list. for (i <- 0 until 10000000) { var result = 0 for (v <- list) { if (v == "c") { result = 1 } } total += result } val t3 = System.currentTimeMillis() // Results. println(total) println(t2 - t1) println(t3 - t2) } }20000000 77 ms, Map 140 ms, List
We use the Map type to associate keys with values. This provides fast, hashed lookups. Maps solve many performance problems, and can even simplify code by removing duplicates.