Multiparadigm Programming in Scala

1. Multiparadigm Programming in Scala H. Conrad Cunningham James C. Church Computer and Information Science University of Mississippi

2. What is Multiparadigm Programming? Definition: A multiparadigm programming language provides “a framework in which programmers can work in a variety of styles, freely intermixing constructs from different paradigms.” [Tim Budd] Programming paradigms: • imperative versus declarative (e.g., functional, logic) • other dimensions – object-oriented, component-oriented, concurrency-oriented, language-oriented

3. Why Learn Multiparadigm Programming? Tim Budd: “Research results from the psychology of programming indicate that expertise in programming is far more strongly related to the number of different programming styles understood by an individual than it is the number of years’ experience in programming.” The “goal of multiparadigm computing is to provide ... a number of different problem-solving styles” so that a programmer can “select a solution technique that best matches the characteristics of the problem to be solved”.

4. Why Teach Multiparadigm Programming? • Contemporary imperative and object-oriented languages increasingly have functional programming features, e.g., • higher order functions (closures) • list comprehensions • New explicitly multiparadigm (object-oriented/functional) languages are appearing, e.g., • Scala on the Java platform (and .Net in future) • F# on the .Net platform

5. Scala Programming language developed by Martin Odersky’s team at EPFL in Switzerland • Executes on the Java platform • Integrates with Java • Has growing usage (e.g., twitter.com) Multiparadigm language • Object-oriented (with generics and mixins) • Functional (similar to Haskell and SML) • Extensible (method calls as operators, currying, closures, by-name parameters) • Actor-based concurrency-oriented programming • Language-oriented programming • Statically typed with Hindley-Milner type inference

6. Scala References • Websitehttp://www.scala-lang.org • Martin Odersky. Scala Tutorial for Java Programmers. • Martin Odersky. Scala By Example. • Martin Odersky, Lex Spoon, and Bill Venners. Programming in Scala: A Comprehensive Step-By-Step Guide, Artima, Inc., 2009. • Books from Apress and Pragmatic Bookshelf in May, O’Reilly in August, Cambridge late 2009

7. Defining Hello World object HelloWorld { // Mississippi version def main(args: Array[String]){ println("Hey world!") } } • Singleton object named HelloWorld (also replaces static methods and variables) • Method main defined (procedure) • Parameter args of type Array[String] • Array is generic class with type parameter

8. Interpreting Hello World > scala This is a Scala shell. Type in expressions to have them evaluated. Type :help for more information. scala> object HelloWorld { | def main(args: Array[String]) { | println("Hey world!") | } |} defined module HelloWorld scala> HelloWorld.main(null) Hey world! unnamed0: Unit = () scala>:q

9. Compiling & Executing Hello World > scalac HelloWorld.scala > scala HelloWorld Hey world!

10. Numbers are Objects • Consider expression 1 + 2 * 3 / x • Operators are method calls (like Smalltalk) • Operator symbols are identifiers • Expression above is same as (1).+(((2).*(3))./(x))

11. Functions are Objects object Timer { def oncePerSecond(callback:() => Unit){ while (true) { callback(); Thread sleep 1000 } // 1-arg method sleep used as operator } def welcome() { println("Welcome to CCSC:MS!") } def main(args: Array[String]) { oncePerSecond(welcome) } }

12. Timer Execution scala> :l Timer.scala Loading Timer.scala... defined module Timer scala> Timer.main(null) Welcome to CCSC:MS! Welcome to CCSC:MS! Welcome to CCSC:MS! …

13. Anonymous Functions object Timer { def oncePerSecond(callback:() => Unit){ while (true) { callback(); Thread sleep 1000 } } def main(args: Array[String]) { oncePerSecond( () => println("Welcome to CCSC:MS!") ) } }

14. Classes class Complex(real: Double, imag: Double){ def re = real def im = imag } • Class primary constructor combined with class body • Parameters of class private constants within class • Parameterless methods re and im • Return types of re and im inferred from expression (cannot be inferred for recursive functions) • Thus more concise syntax

15. Method Overriding // Complex.scala class Complex(real: Double, imag: Double) { def re = real def im = imag override def toString = re + (if (im < 0.0) "" else "+") + im + "i“ } • Classes extend class AnyRef by default • Methods must explicitly override parent method • if expressions

16. Using Classes and Objects scala> :load Complex.scala Loading Complex.scala... defined class Complex scala> val x = new Complex(1,-3) x: Complex = 1.0-3.0i scala> x.toString res0: java.lang.String = 1.0-3.0i

17. Case Classes abstract class Tree // Expression Trees case class Sum(l: Tree, r: Tree) extends Tree case class Var(n: String) extends Tree case class Const(v: int) extends Tree • Algebraic data types as in functional languages • Keyword new not needed to create instances (objects) • Getters defined automatically for constructor parameters • equals method defined on structure of instances • Pattern matching can be used to decompose

18. Pattern Matching object Expressions { type Environ = String => Int def eval(t: Tree, env: Environ): Int = t match { case Sum(l,r) => eval(l,env) + eval(r,env) case Var(n) => env(n) case Const(v) => v } def derive(t: Tree, v: String): Tree = t match { case Sum(l,r) => Sum(derive(l,v), derive(r,v)) case Var(n) if (v == n) => Const(1) case _ => Const(0) }

19. Test Expression Trees def main(args: Array[String]) { val exp: Tree = Sum(Sum(Var("x"),Var("x")), Sum(Const(7),Var("y"))) val env: Environ = { case "x" => 5 case "y" => 7 } println("Expression: " + exp) println("Evaluation with x=5, y=7: " + eval(exp,env)) println("Derivative relative to x:\n " + derive(exp, "x")) println("Derivative relative to y:\n " + derive(exp, "y")) } }

20. Execute Expression Trees scala> :load Expressions.scala Loading Expressions.scala... … scala> Expressions.main(null) Expression: Sum(Sum(Var(x),Var(x)),Sum(Const(7),Var(y))) Evaluation with x=5, y=7: 24 Derivative relative to x: Sum(Sum(Const(1),Const(1)),Sum(Const(0),Const(0))) Derivative relative to y: Sum(Sum(Const(0),Const(0)),Sum(Const(0),Const(1)))

21. Defs, Vals, and Vars Three types of identifier definitions def defines functions with parameters; RHS expression evaluated each time called val defines unchanging values; RHS expression evaluated immediately to initialize var defines storage location whose values can be changed by assignment statements; RHS expression evaluated immediately to initialize

22. Traits trait Ord { // Order comparison operators def < (that: Any): Boolean // abstract def <=(that: Any): Boolean = (this < that) || (this == that) def > (that: Any): Boolean = !(this <= that) def >=(that: Any): Boolean = !(this < that) } • Like Java interfaces except can have concrete methods • Can be “mixed-in” to class • Note < abstract; others defined with < and equals

23. Date Class with Mixin Trait Ord class Date(y: Int, m: Int, d: Int) extends Ord { def year = y def month = m def day = d override def toString(): String = year + "-" + month + "-" + day // need definition of < and equals } • Can only extend only one class or trait • May mix-in additional classes using keyword with

24. Date Class Equals Method override def equals(that: Any): Boolean = that.isInstanceOf[Date] && { val o = that.asInstanceOf[Date] o.day == day && o.month == month && o.year == year } • isInstanceOf[T] checks whether object is an instance of the given type T • asInstanceOf[T] casts static type to T if compatible with dynamic type of object • Value of last statement of function is returned

25. Date Class < Method def <(that: Any): Boolean = { if (!that.isInstanceOf[Date]) error("Cannot compare " + that + " and a Date") val o = that.asInstanceOf[Date] (year < o.year) || (year == o.year && (month < o.month || (month == o.month && day < o.day))) }

26. DateTest object DateTest { def main(args: Array[String]) { val x = new Date(1,1,2000) val y = new Date(12,31,2001) println("x = " + x) println("y = " + y) println("x < y: " + (x<y)) println("x > y: " + (x>y)) } }

27. DateTest Output > scala DateTest x = 1-1-2000 y = 12-31-2001 x < y: true x > y: false

28. Scala Functions • Are first-class values – i.e., functions are objects • Can be higher-order – take functions as arguments or return them as result • Can be anonymous • May be curried – take arguments one at a time, allowing partial application • Are often passed in a closure – with references to free variables they maninpulate • Provide ability to build powerful libraries of higher-order functions

29. Curried Functions scala> def add(x: Int, y: Int) = x + y add: (Int,Int)Int scala> add(1,3) res0: Int = 4 scala> def addc(x: Int)(y: Int) = x + y addc: (Int)(Int)Int scala> addc(1)(3) res1: Int = 4

30. Partial Application scala> def addc(x: Int)(y: Int) = x + y addc: (Int)(Int)Int scala> val z = addc(1) _ z: (Int) => Int = <function> scala> z(3) res2: Int = 4

31. Closures scala> val inc = 10 inc: Int = 10 scala> def incre(x: Int) = x + inc incre: (Int)Int scala> def app(y: Int, g: (Int=>Int)) = g(y) app: (Int,(Int) => Int)Int scala> app(13,incre) res0: Int = 23

32. List Processing // Not actual Scala API code abstract class List[+A]{ … def map[B](f: (A)=> B): List[B] = this match { case Nil => this case x::xs => f(x)::xs.map(f) } … } case object Nil extends List[Nothing] case final class ::[B] (private hd: B, val tl: List[B]) extends List[B]

33. Using List Map scala> val xs = List(3,4,5) xs: List[Int] = List(3, 4, 5) scala> val triples = xs.map(x => 3*x) triples: List[Int] = List(9, 12, 15)

34. More List Processing // Not actual Scala API code abstract class List[+A]{ … def filter(p: (A) => Boolean): List[A] = this match { case Nil => this case x::xs => if (p(x)) x::xs.filter(p) else xs.filter(p) } … }

35. Using List Filter scala> val xs = List(3,4,5,6) xs: List[Int] = List(3, 4, 5, 6) scala> val evens = xs.filter(x => x%2==0) evens: List[Int] = List(4, 6)

36. Other Higher Order List Methods • flatMap • foldLeft, foldRight • reduceLeft, reduceRight • takeWhile, dropWhile • span, break • foreach

37. For Comprehensions scala> for(i <- 1 to 30; | j <- List(2,3,5,7); | if i % j == 0) yield (i,j) res0: Seq.Projection[(Int, Int)] = RangeG((2,2), (3,3), (4,2), (5,5), (6,2), (6,3), (7,7), (8,2), (9,3), (10,2), (10,5), (12,2), (12,3), (14,2), (14,7), (15,3), (15,5), (16,2), (18,2), (18,3), (20,2), (20,5), (21,3), (21,7), (22,2), (24,2), (24,3), (25,5), (26,2), (27,3), (28,2), (28,7), (30,2), (30,3), (30,5))

38. Actors in Scala

39. Motivation • Concurrency is hard! • Real World is parallel and distributed. • Erlang's notion of a process: • Concurrent processes should pass messages to other processes rather than share memory. • Erlang's processes are part of the language. • Scala's actors are part of the library.

40. Actors • Actors act independent of other actors. • Actors have mailboxes. • Actors communicate by sending messages to other actors. • Actors will check their mailbox and react to their messages.

41. Message in a Bottle • Any object can be sent to an Actor case object myMessageObject ... myActor ! myMessageObject

42. Please Mr. Postman • How urgent is it? • react: I need it now! • receiveWithin: I need it soon! • receive: I'll wait. • All three methods will perform pattern matching on the objects received.

43. Overacting import scala.actors._ object SillyActor extends Actor { def act() { // Defines how our actor acts for (i <- 1 to 5) { println(“I'm acting!”)‏ Thread.sleep(1000)‏ } } } ... SillyActor.start() // Begins acting

44. Vegetable Launcher case object Tomato case object Lettuce object VegetableLauncher extends Actor { def act() { for (i <- 1 to 5) { VegetableCatcher ! Tomato // Send it! Thread.sleep(1000)‏ VegetableCatcher ! Lettuce // Send it! Thread.sleep(1000)‏ } } }

45. Vegetable Catcher object VegetableCatcher extends Actor { def act() { loop { react { // Non-blocking call // Pattern Matching case Lettuce => println(“I caught a lettuce!”)‏ case Tomato => println(“I caught a tomato!”)‏ } } } }

46. Lights, Camera, ... VegtableLauncher.start()‏ VegtableCatcher.start()‏ SillyActor.start()‏ I'm acting! I caught a tomato! I'm acting! I caught a lettuce! I'm acting! I caught a tomato! I'm acting! I caught a lettuce! I'm acting! I caught a tomato! I caught a lettuce! I caught a tomato! I caught a lettuce! I caught a tomato! I caught a lettuce!

47. Dining Philosophers • Five philosophers compete for limited resources. • Deadlocks are possible. • Solution implemented with a waiter. • 11 actors total: • 5 philosophers • 5 chopsticks (A mutex actor)‏ • 1 waiter (A singleton object actor)‏

48. The waiter only allows four to sit. After that, names are on the wait list. • Philosophers must still wait for chopsticks after sitting.

49. Six Messages import scala.actors._ import scala.actors.Actor._ import java.util.Random // Philosopher and Waiter communication case object NeedToEat case object HaveASeat case object DoneEating // Philosopher and Chopstick communication case object NeedChopstick case object HeresAChopstick case object GiveBackChopstick

50. Chopstick “Mutex” class Chopstick extends Actor { def act() { loop { react { case NeedChopstick => sender ! HeresAChopstick receive { case GiveBackChopstick => 1 } } } } }