Scala By Example DRAFT October 28, 2009 Martin Odersky PROGRAMMINGMETHODSLABORATORY EPFL SWITZERLAND Contents 1 Introduction1 2 A First Example3 3 Programming with Actors and Messages7 4 Expressions and Simple Functions 11 4.1 Expressions And Simple Functions...................... 11 4.2 Parameters.................................... 12 4.3 Conditional Expressions............................ 15 4.4 Example: Square Roots by Newton’s Method................ 15 4.5 Nested Functions................................ 16 4.6 Tail Recursion.................................. 18 5 First-Class Functions 21 5.1 Anonymous Functions............................. 22 5.2 Currying..................................... 23 5.3 Example: Finding Fixed Points of Functions................ 25 5.4 Summary..................................... 27 5.5 Language Elements Seen So Far....................... 28 6 Classes and Objects 31 7 Case Classes and Pattern Matching 43 7.1 Case Classes and Case Objects........................ 46 7.2 Pattern Matching................................ 47 8 Generic Types and Methods 51 8.1 Type Parameter Bounds............................ 53 8.2 Variance Annotations.............................. 56 iv CONTENTS 8.3 Lower Bounds.................................. 58 8.4 Least Types.................................... 58 8.5 Tuples....................................... 60 8.6 Functions.................................... 61 9 Lists 63 9.1 Using Lists.................................... 63 9.2 Deﬁnition of class List I: First Order Methods............... 65 9.3 Example: Merge sort.............................. 68 9.4 Deﬁnition of class List II: Higher-Order Methods............. 70 9.5 Summary..................................... 77 10 For-Comprehensions 79 10.1 The N-Queens Problem............................ 80 10.2 Querying with For-Comprehensions..................... 81 10.3 Translation of For-Comprehensions..................... 82 10.4 For-Loops.................................... 84 10.5 Generalizing For................................. 84 11 Mutable State 87 11.1 Stateful Objects................................. 87 11.2 Imperative Control Structures......................... 91 11.3 Extended Example: Discrete Event Simulation............... 92 11.4 Summary..................................... 97 12 Computing with Streams 99 13 Iterators 103 13.1 Iterator Methods................................ 103 13.2 Constructing Iterators............................. 106 13.3 Using Iterators.................................. 107 14 Lazy Values 109 15 Implicit Parameters and Conversions 113 CONTENTS v 16 Hindley/Milner Type Inference 117 17 Abstractions for Concurrency 125 17.1 Signals and Monitors.............................. 125 17.2 SyncVars..................................... 127 17.3 Futures...................................... 127 17.4 Parallel Computations............................. 128 17.5 Semaphores................................... 129 17.6 Readers/Writers................................. 129 17.7 Asynchronous Channels............................ 130 17.8 Synchronous Channels............................. 131 17.9 Workers...................................... 132 17.10Mailboxes.................................... 134 17.11Actors....................................... 137 Chapter 1 Introduction Scala smoothly integrates object-oriented and functional programming. It is de- signed to express common programming patterns in a consise, elegant, and type- safe way. Scala introduces several innovative language constructs. For instance: • Abstract types and mixin composition unify concepts from object and module systems. • Pattern matching over class hierarchies uniﬁes functional and object- oriented data access. It greatly simpliﬁes the processing of XML trees. • A ﬂexible syntax and type system enables the construction of advanced li- braries and new domain speciﬁc languages. At the same time, Scala is compatible with Java. Java libraries and frameworks can be used without glue code or additional declarations. This document introduces Scala in an informal way, through a sequence of exam- ples. Chapters2 and3 highlight some of the features that make Scala interesting. The fol- lowing chapters introduce the language constructs of Scala in a more thorough way, starting with simple expressions and functions, and working up through objects and classes, lists and streams, mutable state, pattern matching to more complete exam- ples that show interesting programming techniques. The present informal exposi- tion is meant to be complemented by the Scala Language Reference Manual which speciﬁes Scala in a more detailed and precise way. Acknowledgment. We owe a great debt to Abelson’s and Sussman’s wonderful book “Structure and Interpretation of Computer Programs”[ASS96]. Many of their examples and exercises are also present here. Of course, the working language has in each case been changed from Scheme to Scala. Furthermore, the examples make use of Scala’s object-oriented constructs where appropriate. Chapter 2 A First Example As a ﬁrst example, here is an implementation of Quicksort in Scala. def sort(xs: Array[Int]) { def swap(i: Int, j: Int) { val t = xs(i); xs(i) = xs(j); xs(j) = t } def sort1(l: Int, r: Int) { val pivot = xs((l + r) / 2) var i = l; var j = r while (i <= j) { while (xs(i) < pivot) i += 1 while (xs(j) > pivot) j -= 1 if (i <= j) { swap(i, j) i += 1 j -= 1 } } if (l < j) sort1(l, j) if (j < r) sort1(i, r) } sort1(0, xs.length - 1) } The implementation looks quite similar to what one would write in Java or C. We use the same operators and similar control structures. There are also some minor syntactical differences. In particular: • Deﬁnitions start with a reserved word. Function deﬁnitions start with def, variable deﬁnitions start with var and deﬁnitions of values (i.e. read only vari- ables) start with val. 4 A First Example • The declared type of a symbol is given after the symbol and a colon. The de- clared type can often be omitted, because the compiler can infer it from the context. • Array types are written Array[T] rather than T[], and array selections are writ- ten a(i) rather than a[i]. • Functions can be nested inside other functions. Nested functions can access parameters and local variables of enclosing functions. For instance, the name of the array xs is visible in functions swap and sort1, and therefore need not be passed as a parameter to them. So far, Scala looks like a fairly conventional language with some syntactic peculiar- ities. In fact it is possible to write programs in a conventional imperative or object- oriented style. This is important because it is one of the things that makes it easy to combine Scala components with components written in mainstream languages such as Java, C# or Visual Basic. However, it is also possible to write programs in a style which looks completely dif- ferent. Here is Quicksort again, this time written in functional style. def sort(xs: Array[Int]): Array[Int] = { if (xs.length <= 1) xs else { val pivot = xs(xs.length / 2) Array.concat( sort(xs filter (pivot >)), xs filter (pivot ==), sort(xs filter (pivot <))) } } The functional program captures the essence of the quicksort algorithm in a concise way: • If the array is empty or consists of a single element, it is already sorted, so return it immediately. • If the array is not empty, pick an an element in the middle of it as a pivot. • Partition the array into two sub-arrays containing elements that are less than, respectively greater than the pivot element, and a third array which contains elements equal to pivot. • Sort the ﬁrst two sub-arrays by a recursive invocation of the sort function.1 • The result is obtained by appending the three sub-arrays together. 1This is not quite what the imperative algorithm does; the latter partitions the array into two sub-arrays containing elements less than or greater or equal to pivot. 5 Both the imperative and the functional implementation have the same asymptotic complexity – O(N log(N)) in the average case and O(N 2) in the worst case. But where the imperative implementation operates in place by modifying the argument array, the functional implementation returns a new sorted array and leaves the ar- gument array unchanged. The functional implementation thus requires more tran- sient memory than the imperative one. The functional implementation makes it look like Scala is a language that’s special- ized for functional operations on arrays. In fact, it is not; all of the operations used in the example are simple library methods of a sequence class Seq[T] which is part of the standard Scala library, and which itself is implemented in Scala. Because arrays are instances of Seq all sequence methods are available for them. In particular, there is the method filter which takes as argument a predicate func- tion. This predicate function must map array elements to boolean values. The result of filter is an array consisting of all the elements of the original array for which the given predicate function is true. The filter method of an object of type Array[T] thus has the signature def filter(p: T => Boolean): Array[T] Here, T => Boolean is the type of functions that take an element of type t and return a Boolean. Functions like filter that take another function as argument or return one as result are called higher-order functions. Scala does not distinguish between identiﬁers and operator names. An identiﬁer can be either a sequence of letters and digits which begins with a letter, or it can be a sequence of special characters, such as “+”, “*”, or “:”. Any identiﬁer can be used as an inﬁx operator in Scala. The binary operation E op E0 is always interpreted as the method call E.op(E0). This holds also for binary inﬁx operators which start with a letter. Hence, the expression xs filter (pivot >) is equivalent to the method call xs.filter(pivot >). In the quicksort program, filter is applied three times to an anonymous function argument. The ﬁrst argument, pivot >, represents a function that takes an argu- ment x and returns the value pivot > x. This is an example of a partially applied function. Another, equivalent way to write this function which makes the missing argument explicit is x => pivot > x. The function is anonymous, i.e. it is not de- ﬁned with a name. The type of the x parameter is omitted because a Scala compiler can infer it automatically from the context where the function is used. To summa- rize, xs.filter(pivot >) returns a list consisting of all elements of the list xs that are smaller than pivot. Looking again in detail at the ﬁrst, imperative implementation of Quicksort, we ﬁnd that many of the language constructs used in the second solution are also present, albeit in a disguised form. 6 A First Example For instance, “standard” binary operators such as +,-, or < are not treated in any special way. Like append, they are methods of their left operand. Consequently, the expression i + 1 is regarded as the invocation i.+(1) of the + method of the integer value x. Of course, a compiler is free (if it is moderately smart, even expected) to recognize the special case of calling the + method over integer arguments and to generate efﬁcient inline code for it. For efﬁciency and better error diagnostics the while loop is a primitive construct in Scala. But in principle, it could have just as well been a predeﬁned function. Here is a possible implementation of it: def While (p: => Boolean) (s: => Unit) { if (p) { s ; While(p)(s) } } The While function takes as ﬁrst parameter a test function, which takes no parame- ters and yields a boolean value. As second parameter it takes a command function which also takes no parameters and yields a result of type Unit. While invokes the command function as long as the test function yields true. Scala’s Unit type roughly corresponds to void in Java; it is used whenever a func- tion does not return an interesting result. In fact, because Scala is an expression- oriented language, every function returns some result. If no explicit return expres- sion is given, the value (), which is pronounced “unit”, is assumed. This value is of type Unit. Unit-returning functions are also called procedures. Here’s a more “expression-oriented” formulation of the swap function in the ﬁrst implementation of quicksort, which makes this explicit: def swap(i: Int, j: Int) { val t = xs(i); xs(i) = xs(j); xs(j) = t () } The result value of this function is simply its last expression – a return keyword is not necessary. Note that functions returning an explicit value always need an “=” before their body or deﬁning expression. Chapter 3 Programming with Actors and Mes- sages Here’s an example that shows an application area for which Scala is particularly well suited. Consider the task of implementing an electronic auction service. We use an Erlang-style actor process model to implement the participants of the auction. Actors are objects to which messages are sent. Every actor has a “mailbox” of its in- coming messages which is represented as a queue. It can work sequentially through the messages in its mailbox, or search for messages matching some pattern. For every traded item there is an auctioneer actor that publishes information about the traded item, that accepts offers from clients and that communicates with the seller and winning bidder to close the transaction. We present an overview of a simple implementation here. As a ﬁrst step, we deﬁne the messages that are exchanged during an auction. There are two abstract base classes AuctionMessage for messages from clients to the auc- tion service, and AuctionReply for replies from the service to the clients. For both base classes there exists a number of cases, which are deﬁned in Figure 3.1. For each base class, there are a number of case classes which deﬁne the format of particular messages in the class. These messages might well be ultimately mapped to small XML documents. We expect automatic tools to exist that convert between XML documents and internal data structures like the ones deﬁned above. Figure 3.2 presents a Scala implementation of a class Auction for auction actors that coordinate the bidding on one item. Objects of this class are created by indicating • a seller actor which needs to be notiﬁed when the auction is over, • a minimal bid, • the date when the auction is to be closed. The behavior of the actor is deﬁned by its act method. That method repeatedly 8 Programming with Actors and Messages import scala.actors.Actor abstract class AuctionMessage case class Offer(bid: Int, client: Actor) extends AuctionMessage case class Inquire(client: Actor) extends AuctionMessage abstract class AuctionReply case class Status(asked: Int, expire: Date) extends AuctionReply case object BestOffer extends AuctionReply case class BeatenOffer(maxBid: Int) extends AuctionReply case class AuctionConcluded(seller: Actor, client: Actor) extends AuctionReply case object AuctionFailed extends AuctionReply case object AuctionOver extends AuctionReply Listing 3.1: Message Classes for an Auction Service selects (using receiveWithin) a message and reacts to it, until the auction is closed, which is signaled by a TIMEOUT message. Before ﬁnally stopping, it stays active for another period determined by the timeToShutdown constant and replies to further offers that the auction is closed. Here are some further explanations of the constructs used in this program: • The receiveWithin method of class Actor takes as parameters a time span given in milliseconds and a function that processes messages in the mailbox. The function is given by a sequence of cases that each specify a pattern and an action to perform for messages matching the pattern. The receiveWithin method selects the ﬁrst message in the mailbox which matches one of these patterns and applies the corresponding action to it. • The last case of receiveWithin is guarded by a TIMEOUT pattern. If no other messages are received in the meantime, this pattern is triggered after the time span which is passed as argument to the enclosing receiveWithin method. TIMEOUT is a special message, which is triggered by the Actor implementation itself. • Reply messages are sent using syntax of the form destination ! SomeMessage.! is used here as a binary operator with an actor and a message as arguments. This is equivalent in Scala to the method call destination.!(SomeMessage), i.e. the invocation of the ! method of the destination actor with the given message as parameter. The preceding discussion gave a ﬂavor of distributed programming in Scala. It might seem that Scala has a rich set of language constructs that support actor pro- cesses, message sending and receiving, programming with timeouts, etc. In fact, the 9 class Auction(seller: Actor, minBid: Int, closing: Date) extends Actor { val timeToShutdown = 36000000 // msec val bidIncrement = 10 def act() { var maxBid = minBid - bidIncrement var maxBidder: Actor = null var running = true while (running) { receiveWithin ((closing.getTime() - new Date().getTime())) { case Offer(bid, client) => if (bid >= maxBid + bidIncrement) { if (maxBid >= minBid) maxBidder ! BeatenOffer(bid) maxBid = bid; maxBidder = client; client ! BestOffer } else { client ! BeatenOffer(maxBid) } case Inquire(client) => client ! Status(maxBid, closing) case TIMEOUT => if (maxBid >= minBid) { val reply = AuctionConcluded(seller, maxBidder) maxBidder ! reply; seller ! reply } else { seller ! AuctionFailed } receiveWithin(timeToShutdown) { case Offer(_, client) => client ! AuctionOver case TIMEOUT => running = false } } } } } Listing 3.2: Implementation of an Auction Service 10 Programming with Actors and Messages opposite is true. All the constructs discussed above are offered as methods in the li- brary class Actor. That class is itself implemented in Scala, based on the underlying thread model of the host language (e.g. Java, or .NET). The implementation of all features of class Actor used here is given in Section 17.11. The advantages of the library-based approach are relative simplicity of the core lan- guage and ﬂexibility for library designers. Because the core language need not spec- ify details of high-level process communication, it can be kept simpler and more general. Because the particular model of messages in a mailbox is a library module, it can be freely modiﬁed if a different model is needed in some applications. The approach requires however that the core language is expressive enough to provide the necessary language abstractions in a convenient way. Scala has been designed with this in mind; one of its major design goals was that it should be ﬂexible enough to act as a convenient host language for domain speciﬁc languages implemented by library modules. For instance, the actor communication constructs presented above can be regarded as one such domain speciﬁc language, which conceptually extends the Scala core. Chapter 4 Expressions and Simple Functions The previous examples gave an impression of what can be done with Scala. We now introduce its constructs one by one in a more systematic fashion. We start with the smallest level, expressions and functions. 4.1 Expressions And Simple Functions A Scala system comes with an interpreter which can be seen as a fancy calculator. A user interacts with the calculator by typing in expressions. The calculator returns the evaluation results and their types. For example: scala> 87 + 145 unnamed0: Int = 232 scala> 5 + 2 * 3 unnamed1: Int = 11 scala> "hello" + " world!" unnamed2: java.lang.String = hello world! It is also possible to name a sub-expression and use the name instead of the expres- sion afterwards: scala> def scale = 5 scale: Int scala> 7 * scale unnamed3: Int = 35 scala> def pi = 3.141592653589793 pi: Double 12 Expressions and Simple Functions scala> def radius = 10 radius: Int scala> 2 * pi * radius unnamed4: Double = 62.83185307179586 Deﬁnitions start with the reserved word def; they introduce a name which stands for the expression following the = sign. The interpreter will answer with the intro- duced name and its type. Executing a deﬁnition such as def x = e will not evaluate the expression e. In- stead e is evaluated whenever x is used. Alternatively, Scala offers a value deﬁni- tion val x = e, which does evaluate the right-hand-side e as part of the evaluation of the deﬁnition. If x is then used subsequently, it is immediately replaced by the pre-computed value of e, so that the expression need not be evaluated again. How are expressions evaluated? An expression consisting of operators and operands is evaluated by repeatedly applying the following simpliﬁcation steps. • pick the left-most operation • evaluate its operands • apply the operator to the operand values. A name deﬁned by def is evaluated by replacing the name by the (unevaluated) deﬁnition’s right hand side. A name deﬁned by val is evaluated by replacing the name by the value of the deﬁnitions’s right-hand side. The evaluation process stops once we have reached a value. A value is some data item such as a string, a number, an array, or a list. Example 4.1.1 Here is an evaluation of an arithmetic expression. (2 * pi) * radius → (2 * 3.141592653589793) * radius → 6.283185307179586 * radius → 6.283185307179586 * 10 → 62.83185307179586 The process of stepwise simpliﬁcation of expressions to values is called reduction. 4.2 Parameters Using def, one can also deﬁne functions with parameters. For example: 4.2 Parameters 13 scala> def square(x: Double) = x * x square: (Double)Double scala> square(2) unnamed0: Double = 4.0 scala> square(5 + 3) unnamed1: Double = 64.0 scala> square(square(4)) unnamed2: Double = 256.0 scala> def sumOfSquares(x: Double, y: Double) = square(x) + square(y) sumOfSquares: (Double,Double)Double scala> sumOfSquares(3, 2 + 2) unnamed3: Double = 25.0 Function parameters follow the function name and are always enclosed in paren- theses. Every parameter comes with a type, which is indicated following the param- eter name and a colon. At the present time, we only need basic numeric types such as the type scala.Double of double precision numbers. Scala deﬁnes type aliases for some standard types, so we can write numeric types as in Java. For instance double is a type alias of scala.Double and int is a type alias for scala.Int. Functions with parameters are evaluated analogously to operators in expressions. First, the arguments of the function are evaluated (in left-to-right order). Then, the function application is replaced by the function’s right hand side, and at the same time all formal parameters of the function are replaced by their corresponding ac- tual arguments. Example 4.2.1 sumOfSquares(3, 2+2) → sumOfSquares(3, 4) → square(3) + square(4) → 3 * 3 + square(4) → 9 + square(4) → 9 + 4 * 4 → 9 + 16 → 25 The example shows that the interpreter reduces function arguments to values be- fore rewriting the function application. One could instead have chosen to apply the function to unreduced arguments. This would have yielded the following reduction sequence: 14 Expressions and Simple Functions sumOfSquares(3, 2+2) → square(3) + square(2+2) → 3 * 3 + square(2+2) → 9 + square(2+2) → 9 + (2+2) *(2+2) → 9 + 4 *(2+2) → 9 + 4 * 4 → 9 + 16 → 25 The second evaluation order is known as call-by-name, whereas the ﬁrst one is known as call-by-value. For expressions that use only pure functions and that there- fore can be reduced with the substitution model, both schemes yield the same ﬁnal values. Call-by-value has the advantage that it avoids repeated evaluation of arguments. Call-by-name has the advantage that it avoids evaluation of arguments when the parameter is not used at all by the function. Call-by-value is usually more efﬁcient than call-by-name, but a call-by-value evaluation might loop where a call-by-name evaluation would terminate. Consider: scala> def loop: Int = loop loop: Int scala> def first(x: Int, y: Int) = x first: (Int,Int)Int Then first(1, loop) reduces with call-by-name to 1, whereas the same term re- duces with call-by-value repeatedly to itself, hence evaluation does not terminate. first(1, loop) → first(1, loop) → first(1, loop) → ... Scala uses call-by-value by default, but it switches to call-by-name evaluation if the parameter type is preceded by =>. Example 4.2.2 scala> def constOne(x: Int, y: => Int) = 1 constOne: (Int,=> Int)Int scala> constOne(1, loop) unnamed0: Int = 1 scala> constOne(loop, 2) // gives an infinite loop. 4.3 Conditional Expressions 15 ^C // stops execution with Ctrl-C 4.3 Conditional Expressions Scala’s if-else lets one choose between two alternatives. Its syntax is like Java’s if-else. But where Java’s if-else can be used only as an alternative of state- ments, Scala allows the same syntax to choose between two expressions. That’s why Scala’s if-else serves also as a substitute for Java’s conditional expression ... ? ... : .... Example 4.3.1 scala> def abs(x: Double) = if (x >= 0) x else -x abs: (Double)Double Scala’s boolean expressions are similar to Java’s; they are formed from the constants true and false, comparison operators, boolean negation ! and the boolean opera- tors && and ||. 4.4 Example: Square Roots by Newton’s Method We now illustrate the language elements introduced so far in the construction of a more interesting program. The task is to write a function def sqrt(x: Double): Double = ... which computes the square root of x. A common way to compute square roots is by Newton’s method of successive ap- proximations. One starts with an initial guess y (say: y = 1). One then repeatedly improves the current guess y by taking the average of y and x/y. As an example, the next three columns indicate the guess y, the quotient x/y, and their average for the ﬁrst approximations of p2. 1 2/1 = 2 1.5 1.5 2/1.5 = 1.3333 1.4167 1.4167 2/1.4167 = 1.4118 1.4142 1.4142 ... ... y x/y (y + x/y)/2 One can implement this algorithm in Scala by a set of small functions, which each represent one of the elements of the algorithm. We ﬁrst deﬁne a function for iterating from a guess to the result: 16 Expressions and Simple Functions def sqrtIter(guess: Double, x: Double): Double = if (isGoodEnough(guess, x)) guess else sqrtIter(improve(guess, x), x) Note that sqrtIter calls itself recursively. Loops in imperative programs can always be modeled by recursion in functional programs. Note also that the deﬁnition of sqrtIter contains a return type, which follows the parameter section. Such return types are mandatory for recursive functions. For a non-recursive function, the return type is optional; if it is missing the type checker will compute it from the type of the function’s right-hand side. However, even for non-recursive functions it is often a good idea to include a return type for better documentation. As a second step, we deﬁne the two functions called by sqrtIter: a function to improve the guess and a termination test isGoodEnough. Here is their deﬁnition. def improve(guess: Double, x: Double) = (guess + x / guess) / 2 def isGoodEnough(guess: Double, x: Double) = abs(square(guess) - x) < 0.001 Finally, the sqrt function itself is deﬁned by an application of sqrtIter. def sqrt(x: Double) = sqrtIter(1.0, x) Exercise 4.4.1 The isGoodEnough test is not very precise for small numbers and might lead to non-termination for very large ones (why?). Design a different ver- sion of isGoodEnough which does not have these problems. Exercise 4.4.2 Trace the execution of the sqrt(4) expression. 4.5 Nested Functions The functional programming style encourages the construction of many small helper functions. In the last example, the implementation of sqrt made use of the helper functions sqrtIter, improve and isGoodEnough. The names of these func- tions are relevant only for the implementation of sqrt. We normally do not want users of sqrt to access these functions directly. We can enforce this (and avoid name-space pollution) by including the helper func- tions within the calling function itself: def sqrt(x: Double) = { def sqrtIter(guess: Double, x: Double): Double = 4.5 Nested Functions 17 if (isGoodEnough(guess, x)) guess else sqrtIter(improve(guess, x), x) def improve(guess: Double, x: Double) = (guess + x / guess) / 2 def isGoodEnough(guess: Double, x: Double) = abs(square(guess) - x) < 0.001 sqrtIter(1.0, x) } In this program, the braces { ... } enclose a block. Blocks in Scala are themselves expressions. Every block ends in a result expression which deﬁnes its value. The result expression may be preceded by auxiliary deﬁnitions, which are visible only in the block itself. Every deﬁnition in a block must be followed by a semicolon, which separates this deﬁnition from subsequent deﬁnitions or the result expression. However, a semi- colon is inserted implicitly at the end of each line, unless one of the following con- ditions is true. 1. Either the line in question ends in a word such as a period or an inﬁx-operator which would not be legal as the end of an expression. 2. Or the next line begins with a word that cannot start a expression. 3. Or we are inside parentheses (...) or brackets , because these cannot contain multiple statements anyway. Therefore, the following are all legal: def f(x: Int) = x + 1; f(1) + f(2) def g1(x: Int) = x + 1 g(1) + g(2) def g2(x: Int) = {x + 1}; /*‘;’ mandatory */ g2(1) + g2(2) def h1(x) = x + y h1(1) * h1(2) def h2(x: Int) = ( x // parentheses mandatory, otherwise a semicolon + y // would be inserted after the ‘x’. ) h2(1) / h2(2) 18 Expressions and Simple Functions Scala uses the usual block-structured scoping rules. A name deﬁned in some outer block is visible also in some inner block, provided it is not redeﬁned there. This rule permits us to simplify our sqrt example. We need not pass x around as an additional parameter of the nested functions, since it is always visible in them as a parameter of the outer function sqrt. Here is the simpliﬁed code: def sqrt(x: Double) = { def sqrtIter(guess: Double): Double = if (isGoodEnough(guess)) guess else sqrtIter(improve(guess)) def improve(guess: Double) = (guess + x / guess) / 2 def isGoodEnough(guess: Double) = abs(square(guess) - x) < 0.001 sqrtIter(1.0) } 4.6 Tail Recursion Consider the following function to compute the greatest common divisor of two given numbers. def gcd(a: Int, b: Int): Int = if (b == 0) a else gcd(b, a % b) Using our substitution model of function evaluation, gcd(14, 21) evaluates as fol- lows: gcd(14, 21) → if (21 == 0) 14 else gcd(21, 14 % 21) → if (false) 14 else gcd(21, 14 % 21) → gcd(21, 14 % 21) → gcd(21, 14) → if (14 == 0) 21 else gcd(14, 21 % 14) → → gcd(14, 21 % 14) → gcd(14, 7) → if (7 == 0) 14 else gcd(7, 14 % 7) → → gcd(7, 14 % 7) → gcd(7, 0) → if (0 == 0) 7 else gcd(0, 7 % 0) → → 7 Contrast this with the evaluation of another recursive function, factorial: def factorial(n: Int): Int = if (n == 0) 1 else n * factorial(n - 1) 4.6 Tail Recursion 19 The application factorial(5) rewrites as follows: factorial(5) → if (5 == 0) 1 else 5 * factorial(5 - 1) → 5 * factorial(5 - 1) → 5 * factorial(4) → ... → 5 *(4 * factorial(3)) → ... → 5 *(4 *(3 * factorial(2))) → ... → 5 *(4 *(3 *(2 * factorial(1)))) → ... → 5 *(4 *(3 *(2 *(1 * factorial(0)))) → ... → 5 *(4 *(3 *(2 *(1 * 1)))) → ... → 120 There is an important difference between the two rewrite sequences: The terms in the rewrite sequence of gcd have again and again the same form. As evaluation pro- ceeds, their size is bounded by a constant. By contrast, in the evaluation of factorial we get longer and longer chains of operands which are then multiplied in the last part of the evaluation sequence. Even though actual implementations of Scala do not work by rewriting terms, they nevertheless should have the same space behavior as in the rewrite sequences. In the implementation of gcd, one notes that the recursive call to gcd is the last action performed in the evaluation of its body. One also says that gcd is “tail-recursive”. The ﬁnal call in a tail-recursive function can be implemented by a jump back to the beginning of that function. The arguments of that call can overwrite the parameters of the current instantiation of gcd, so that no new stack space is needed. Hence, tail recursive functions are iterative processes, which can be executed in constant space. By contrast, the recursive call in factorial is followed by a multiplication. Hence, a new stack frame is allocated for the recursive instance of factorial, and is deallo- cated after that instance has ﬁnished. The given formulation of the factorial func- tion is not tail-recursive; it needs space proportional to its input parameter for its execution. More generally, if the last action of a function is a call to another (possibly the same) function, only a single stack frame is needed for both functions. Such calls are called “tail calls”. In principle, tail calls can always re-use the stack frame of the calling function. However, some run-time environments (such as the Java VM) lack the primitives to make stack frame re-use for tail calls efﬁcient. A production quality Scala implementation is therefore only required to re-use the stack frame of a di- rectly tail-recursive function whose last action is a call to itself. Other tail calls might be optimized also, but one should not rely on this across implementations. Exercise 4.6.1 Design a tail-recursive version of factorial. Chapter 5 First-Class Functions A function in Scala is a “ﬁrst-class value”. Like any other value, it may be passed as a parameter or returned as a result. Functions which take other functions as pa- rameters or return them as results are called higher-order functions. This chapter introduces higher-order functions and shows how they provide a ﬂexible mecha- nism for program composition. As a motivating example, consider the following three related tasks: 1. Write a function to sum all integers between two given numbers a and b: def sumInts(a: Int, b: Int): Int = if (a > b) 0 else a + sumInts(a + 1, b) 2. Write a function to sum the squares of all integers between two given numbers a and b: def square(x: Int): Int = x * x def sumSquares(a: Int, b: Int): Int = if (a > b) 0 else square(a) + sumSquares(a + 1, b) 3. Write a function to sum the powers 2n of all integers n between two given numbers a and b: def powerOfTwo(x: Int): Int = if (x == 0) 1 else 2 * powerOfTwo(x - 1) def sumPowersOfTwo(a: Int, b: Int): Int = if (a > b) 0 else powerOfTwo(a) + sumPowersOfTwo(a + 1, b) These functions are all instances of Pb a f (n) for different values of f . We can factor out the common pattern by deﬁning a function sum: def sum(f: Int => Int, a: Int, b: Int): Int = if (a > b) 0 else f(a) + sum(f, a + 1, b) 22 First-Class Functions The type Int => Int is the type of functions that take arguments of type Int and return results of type Int. So sum is a function which takes another function as a parameter. In other words, sum is a higher-order function. Using sum, we can formulate the three summing functions as follows. def sumInts(a: Int, b: Int): Int = sum(id, a, b) def sumSquares(a: Int, b: Int): Int = sum(square, a, b) def sumPowersOfTwo(a: Int, b: Int): Int = sum(powerOfTwo, a, b) where def id(x: Int): Int = x def square(x: Int): Int = x * x def powerOfTwo(x: Int): Int = if (x == 0) 1 else 2 * powerOfTwo(x - 1) 5.1 Anonymous Functions Parameterization by functions tends to create many small functions. In the previous example, we deﬁned id, square and power as separate functions, so that they could be passed as arguments to sum. Instead of using named function deﬁnitions for these small argument functions, we can formulate them in a shorter way as anonymous functions. An anonymous func- tion is an expression that evaluates to a function; the function is deﬁned without giving it a name. As an example consider the anonymous square function: (x: Int) => x * x The part before the arrow ‘=>’ are the parameters of the function, whereas the part following the ‘=>’ is its body. For instance, here is an anonymous function which multiples its two arguments. (x: Int, y: Int) => x * y Using anonymous functions, we can reformulate the ﬁrst two summation functions without named auxiliary functions: def sumInts(a: Int, b: Int): Int = sum((x: Int) => x, a, b) def sumSquares(a: Int, b: Int): Int = sum((x: Int) => x * x, a, b) Often, the Scala compiler can deduce the parameter type(s) from the context of the anonymous function in which case they can be omitted. For instance, in the case of sumInts or sumSquares, one knows from the type of sum that the ﬁrst parameter must be a function of type Int => Int. Hence, the parameter type Int is redundant and may be omitted. If there is a single parameter without a type, we may also omit 5.2 Currying 23 the parentheses around it: def sumInts(a: Int, b: Int): Int = sum(x => x, a, b) def sumSquares(a: Int, b: Int): Int = sum(x => x * x, a, b) Generally, the Scala term (x1:T1, ..., xn:Tn) => E deﬁnes a function which maps its parameters x1, ..., xn to the result of the expression E(where E may refer to x1, ..., xn). Anonymous functions are not essential language elements of Scala, as they can always be expressed in terms of named functions. Indeed, the anonymous function (x1:T1, ..., xn:Tn) => E is equivalent to the block { def f (x1:T1, ..., xn:Tn) = E ; f _ } where f is fresh name which is used nowhere else in the program. We also say, anonymous functions are “syntactic sugar”. 5.2 Currying The latest formulation of the summing functions is already quite compact. But we can do even better. Note that a and b appear as parameters and arguments of every function but they do not seem to take part in interesting combinations. Is there a way to get rid of them? Let’s try to rewrite sum so that it does not take the bounds a and b as parameters: def sum(f: Int => Int): (Int, Int) => Int = { def sumF(a: Int, b: Int): Int = if (a > b) 0 else f(a) + sumF(a + 1, b) sumF } In this formulation, sum is a function which returns another function, namely the specialized summing function sumF. This latter function does all the work; it takes the bounds a and b as parameters, applies sum’s function parameter f to all integers between them, and sums up the results. Using this new formulation of sum, we can now deﬁne: def sumInts = sum(x => x) def sumSquares = sum(x => x * x) def sumPowersOfTwo = sum(powerOfTwo) Or, equivalently, with value deﬁnitions: 24 First-Class Functions val sumInts = sum(x => x) val sumSquares = sum(x => x * x) val sumPowersOfTwo = sum(powerOfTwo) sumInts, sumSquares, and sumPowersOfTwo can be applied like any other function. For instance, scala> sumSquares(1, 10) + sumPowersOfTwo(10, 20) unnamed0: Int = 267632001 How are function-returning functions applied? As an example, in the expression sum(x => x * x)(1, 10) , the function sum is applied to the squaring function (x => x * x). The resulting function is then applied to the second argument list, (1, 10). This notation is possible because function application associates to the left. That is, if args1 and args2 are argument lists, then f (args1)(args2) is equivalent to (f (args1))(args2) In our example, sum(x => x * x)(1, 10) is equivalent to the following expression: (sum(x => x * x))(1, 10). The style of function-returning functions is so useful that Scala has special syntax for it. For instance, the next deﬁnition of sum is equivalent to the previous one, but is shorter: def sum(f: Int => Int)(a: Int, b: Int): Int = if (a > b) 0 else f(a) + sum(f)(a + 1, b) Generally, a curried function deﬁnition def f (args1) ... (argsn) = E where n > 1 expands to def f (args1) ... (argsn−1) = { def g (argsn) = E ; g } where g is a fresh identiﬁer. Or, shorter, using an anonymous function: def f (args1) ... (argsn−1) = ( argsn ) => E . Performing this step n times yields that def f (args1) ... (argsn) = E is equivalent to 5.3 Example: Finding Fixed Points of Functions 25 def f = (args1) => ... => (argsn) => E . Or, equivalently, using a value deﬁnition: val f = (args1) => ... => (argsn) => E . This style of function deﬁnition and application is called currying after its promoter, Haskell B. Curry, a logician of the 20th century, even though the idea goes back fur- ther to Moses Schönﬁnkel and Gottlob Frege. The type of a function-returning function is expressed analogously to its param- eter list. Taking the last formulation of sum as an example, the type of sum is (Int => Int) => (Int, Int) => Int. This is possible because function types as- sociate to the right. I.e. T1 => T2 => T3 is equivalent to T1 => (T2 => T3) Exercise 5.2.1 1. The sum function uses a linear recursion. Can you write a tail- recursive one by ﬁlling in the ??’s? def sum(f: Int => Double)(a: Int, b: Int): Double = { def iter(a, result) = { if (??) ?? else iter(??, ??) } iter(??, ??) } Exercise 5.2.2 Write a function product that computes the product of the values of functions at points over a given range. Exercise 5.2.3 Write factorial in terms of product. Exercise 5.2.4 Can you write an even more general function which generalizes both sum and product? 5.3 Example: Finding Fixed Points of Functions A number x is called a ﬁxed point of a function f if f(x) = x . For some functions f we can locate the ﬁxed point by beginning with an initial guess and then applying f repeatedly, until the value does not change anymore (or the change is within a small tolerance). This is possible if the sequence 26 First-Class Functions x, f(x), f(f(x)), f(f(f(x))), ... converges to ﬁxed point of f . This idea is captured in the following “ﬁxed-point ﬁnding function”: val tolerance = 0.0001 def isCloseEnough(x: Double, y: Double) = abs((x - y) / x) < tolerance def fixedPoint(f: Double => Double)(firstGuess: Double) = { def iterate(guess: Double): Double = { val next = f(guess) if (isCloseEnough(guess, next)) next else iterate(next) } iterate(firstGuess) } We now apply this idea in a reformulation of the square root function. Let’s start with a speciﬁcation of sqrt: sqrt(x) = the y such that y * y = x = the y such that y = x / y Hence, sqrt(x) is a ﬁxed point of the function y => x / y. This suggests that sqrt(x) can be computed by ﬁxed point iteration: def sqrt(x: double) = fixedPoint(y => x / y)(1.0) But if we try this, we ﬁnd that the computation does not converge. Let’s instrument the ﬁxed point function with a print statement which keeps track of the current guess value: def fixedPoint(f: Double => Double)(firstGuess: Double) = { def iterate(guess: Double): Double = { val next = f(guess) println(next) if (isCloseEnough(guess, next)) next else iterate(next) } iterate(firstGuess) } Then, sqrt(2) yields: 2.0 1.0 2.0 1.0 2.0 5.4 Summary 27 ... One way to control such oscillations is to prevent the guess from changing too much. This can be achieved by averaging successive values of the original sequence: scala> def sqrt(x: Double) = fixedPoint(y => (y + x/y) / 2)(1.0) sqrt: (Double)Double scala> sqrt(2.0) 1.5 1.4166666666666665 1.4142156862745097 1.4142135623746899 1.4142135623746899 In fact, expanding the fixedPoint function yields exactly our previous deﬁnition of ﬁxed point from Section 4.4. The previous examples showed that the expressive power of a language is consid- erably enhanced if functions can be passed as arguments. The next example shows that functions which return functions can also be very useful. Consider again ﬁxed point iterations. We started with the observation that p(x) is a ﬁxed point of the function y => x / y. Then we made the iteration converge by averaging successive values. This technique of average damping is so general that it can be wrapped in another function. def averageDamp(f: Double => Double)(x: Double) = (x + f(x)) / 2 Using averageDamp, we can reformulate the square root function as follows. def sqrt(x: Double) = fixedPoint(averageDamp(y => x/y))(1.0) This expresses the elements of the algorithm as clearly as possible. Exercise 5.3.1 Write a function for cube roots using fixedPoint and averageDamp. 5.4 Summary We have seen in the previous chapter that functions are essential abstractions, be- cause they permit us to introduce general methods of computing as explicit, named elements in our programming language. The present chapter has shown that these abstractions can be combined by higher-order functions to create further abstrac- tions. As programmers, we should look out for opportunities to abstract and to reuse. The highest possible level of abstraction is not always the best, but it is im- portant to know abstraction techniques, so that one can use abstractions where ap- propriate. 28 First-Class Functions 5.5 Language Elements Seen So Far Chapters4 and5 have covered Scala’s language elements to express expressions and types comprising of primitive data and functions. The context-free syntax of these language elements is given below in extended Backus-Naur form, where ‘|’ denotes alternatives, [...] denotes option (0 or 1 occurrence), and {...} denotes repetition (0 or more occurrences). Characters Scala programs are sequences of (Unicode) characters. We distinguish the following character sets: • whitespace, such as ‘’, tabulator, or newline characters, • letters ‘a’ to ‘z’, ‘A’ to ‘Z’, • digits ‘0’ to ‘9’, • the delimiter characters .,;(){}[]\"’ • operator characters, such as ‘#’‘+’, ‘:’. Essentially, these are printable charac- ters which are in none of the character sets above. Lexemes: ident = letter {letter | digit} | operator { operator } | ident ’_’ ident literal = “as in Java” Literals are as in Java. They deﬁne numbers, characters, strings, or boolean values. Examples of literals as 0, 1.0e10,’x’,"he said "hi!"", or true. Identiﬁers can be of two forms. They either start with a letter, which is followed by a (possibly empty) sequence of letters or symbols, or they start with an operator char- acter, which is followed by a (possibly empty) sequence of operator characters. Both forms of identiﬁers may contain underscore characters ‘_’. Furthermore, an under- score character may be followed by either sort of identiﬁer. Hence, the following are all legal identiﬁers: x Room10a + -- foldl_: +_vector It follows from this rule that subsequent operator-identiﬁers need to be separated by whitespace. For instance, the input x+-y is parsed as the three token sequence x, 5.5 Language Elements Seen So Far 29 +-, y. If we want to express the sum of x with the negated value of y, we need to add at least one space, e.g. x+ -y. The $ character is reserved for compiler-generated identiﬁers; it should not be used in source programs. The following are reserved words, they may not be used as identiﬁers: abstract case catch class def do else extends false final finally for if implicit import match new null object override package private protected requires return sealed super this throw trait try true type val var while with yield _ : = => <- <: <% >: # @ Types: Type = SimpleType | FunctionType FunctionType = SimpleType ’=>’ Type | ’(’ [Types] ’)’ ’=>’ Type SimpleType = Byte | Short | Char | Int | Long | Float | Double | Boolean | Unit | String Types = Type {‘,’ Type} Types can be: • number types Byte, Short, Char, Int, Long, Float and Double (these are as in Java), • the type Boolean with values true and false, • the type Unit with the only value (), • the type String, • function types such as (Int, Int) => Int or String => Int => String. Expressions: Expr = InfixExpr | FunctionExpr | if ’(’ Expr ’)’ Expr else Expr InfixExpr = PrefixExpr | InfixExpr Operator InfixExpr Operator = ident PrefixExpr = [’+’ | ’-’ | ’!’ | ’~’ ] SimpleExpr SimpleExpr = ident | literal | SimpleExpr ’.’ ident | Block FunctionExpr = (Bindings | Id) ’=>’ Expr Bindings = ‘(’ Binding {‘,’ Binding} ‘)’ Binding = ident [’:’ Type] Block = ’{’ {Def ’;’} Expr ’}’ 30 First-Class Functions Expressions can be: • identiﬁers such as x, isGoodEnough,*, or +-, • literals, such as 0, 1.0, or "abc", • ﬁeld and method selections, such as System.out.println, • function applications, such as sqrt(x), • operator applications, such as -x or y + x, • conditionals, such as if (x < 0) -x else x, • blocks, such as { val x = abs(y) ; x * 2 }, • anonymous functions, such as x => x + 1 or (x: Int, y: Int) => x + y. Deﬁnitions: Def = FunDef | ValDef FunDef = ’def’ ident {’(’ [Parameters] ’)’} [’:’ Type] ’=’ Expr ValDef = ’val’ ident [’:’ Type] ’=’ Expr Parameters = Parameter {’,’ Parameter} Parameter = ident ’:’ [’=>’] Type Deﬁnitions can be: • function deﬁnitions such as def square(x: Int): Int = x * x, • value deﬁnitions such as val y = square(2). Chapter 6 Classes and Objects Scala does not have a built-in type of rational numbers, but it is easy to deﬁne one, using a class. Here’s a possible implementation. class Rational(n: Int, d: Int) { private def gcd(x: Int, y: Int): Int = { if (x == 0) y else if (x < 0) gcd(-x, y) else if (y < 0) -gcd(x, -y) else gcd(y % x, x) } private val g = gcd(n, d) val numer: Int = n/g val denom: Int = d/g def +(that: Rational) = new Rational(numer * that.denom + that.numer * denom, denom * that.denom) def -(that: Rational) = new Rational(numer * that.denom - that.numer * denom, denom * that.denom) def *(that: Rational) = new Rational(numer * that.numer, denom * that.denom) def /(that: Rational) = new Rational(numer * that.denom, denom * that.numer) } This deﬁnes Rational as a class which takes two constructor arguments n and d, containing the number’s numerator and denominator parts. The class provides ﬁelds which return these parts as well as methods for arithmetic over rational num- bers. Each arithmetic method takes as parameter the right operand of the opera- tion. The left operand of the operation is always the rational number of which the 32 Classes and Objects method is a member. Private members. The implementation of rational numbers deﬁnes a private method gcd which computes the greatest common denominator of two integers, as well as a private ﬁeld g which contains the gcd of the constructor arguments. These members are inaccessible outside class Rational. They are used in the implementa- tion of the class to eliminate common factors in the constructor arguments in order to ensure that numerator and denominator are always in normalized form. Creating and Accessing Objects. As an example of how rational numbers can be used, here’s a program that prints the sum of all numbers 1/i where i ranges from 1 to 10. var i = 1 var x = new Rational(0, 1) while (i <= 10) { x += new Rational(1, i) i += 1 } println("" + x.numer + "/" + x.denom) The + takes as left operand a string and as right operand a value of arbitrary type. It returns the result of converting its right operand to a string and appending it to its left operand. Inheritance and Overriding. Every class in Scala has a superclass which it ex- tends. If a class does not mention a superclass in its deﬁnition, the root type scala.AnyRef is implicitly assumed (for Java implementations, this type is an alias for java.lang.Object. For instance, class Rational could equivalently be deﬁned as class Rational(n: Int, d: Int) extends AnyRef { ...// as before } A class inherits all members from its superclass. It may also redeﬁne (or: override) some inherited members. For instance, class java.lang.Object deﬁnes a method toString which returns a representation of the object as a string: class Object { ... def toString: String = ... } 33 The implementation of toString in Object forms a string consisting of the object’s class name and a number. It makes sense to redeﬁne this method for objects that are rational numbers: class Rational(n: Int, d: Int) extends AnyRef { ...// as before override def toString = "" + numer + "/" + denom } Note that, unlike in Java, redeﬁning deﬁnitions need to be preceded by an override modiﬁer. If class A extends class B, then objects of type A may be used wherever objects of type B are expected. We say in this case that type A conforms to type B. For instance, Rational conforms to AnyRef, so it is legal to assign a Rational value to a variable of type AnyRef: var x: AnyRef = new Rational(1, 2) Parameterless Methods. Unlike in Java, methods in Scala do not necessarily take a parameter list. An example is the square method below. This method is invoked by simply mentioning its name. class Rational(n: Int, d: Int) extends AnyRef { ...// as before def square = new Rational(numer*numer, denom*denom) } val r = new Rational(3, 4) println(r.square) // prints‘‘9/16’’* That is, parameterless methods are accessed just as value ﬁelds such as numer are. The difference between values and parameterless methods lies in their deﬁnition. The right-hand side of a value is evaluated when the object is created, and the value does not change afterwards. A right-hand side of a parameterless method, on the other hand, is evaluated each time the method is called. The uniform access of ﬁelds and parameterless methods gives increased ﬂexibility for the implementer of a class. Often, a ﬁeld in one version of a class becomes a computed value in the next version. Uniform access ensures that clients do not have to be rewritten because of that change. Abstract Classes. Consider the task of writing a class for sets of integer numbers with two operations, incl and contains.(s incl x) should return a new set which contains the element x together with all the elements of set s.(s contains x) should return true if the set s contains the element x, and should return false oth- erwise. The interface of such sets is given by: 34 Classes and Objects abstract class IntSet { def incl(x: Int): IntSet def contains(x: Int): Boolean } IntSet is labeled as an abstract class. This has two consequences. First, abstract classes may have deferred members which are declared but which do not have an implementation. In our case, both incl and contains are such members. Second, because an abstract class might have unimplemented members, no objects of that class may be created using new. By contrast, an abstract class may be used as a base class of some other class, which implements the deferred members. Traits. Instead of abstract class one also often uses the keyword trait in Scala. Traits are abstract classes that are meant to be added to some other class. This might be because a trait adds some methods or ﬁelds to an unknown parent class. For instance, a trait Bordered might be used to add a border to a various graphical components. Another usage scenario is where the trait collects signatures of some functionality provided by different classes, much in the way a Java interface would work. Since IntSet falls in this category, one can alternatively deﬁne it as a trait: trait IntSet { def incl(x: Int): IntSet def contains(x: Int): Boolean } Implementing Abstract Classes. Let’s say, we plan to implement sets as binary trees. There are two possible forms of trees. A tree for the empty set, and a tree consisting of an integer and two subtrees. Here are their implementations. class EmptySet extends IntSet { def contains(x: Int): Boolean = false def incl(x: Int): IntSet = new NonEmptySet(x, new EmptySet, new EmptySet) } class NonEmptySet(elem: Int, left: IntSet, right: IntSet) extends IntSet { def contains(x: Int): Boolean = if (x < elem) left contains x else if (x > elem) right contains x else true def incl(x: Int): IntSet = if (x < elem) new NonEmptySet(elem, left incl x, right) else if (x > elem) new NonEmptySet(elem, left, right incl x) else this 35 } Both EmptySet and NonEmptySet extend class IntSet. This implies that types EmptySet and NonEmptySet conform to type IntSet – a value of type EmptySet or NonEmptySet may be used wherever a value of type IntSet is required. Exercise 6.0.1 Write methods union and intersection to form the union and in- tersection between two sets. Exercise 6.0.2 Add a method def excl(x: Int) to return the given set without the element x. To accomplish this, it is useful to also implement a test method def isEmpty: Boolean for sets. Dynamic Binding. Object-oriented languages (Scala included) use dynamic dis- patch for method invocations. That is, the code invoked for a method call depends on the run-time type of the object which contains the method. For example, con- sider the expression s contains 7 where s is a value of declared type s: IntSet. Which code for contains is executed depends on the type of value of s at run-time. If it is an EmptySet value, it is the implementation of contains in class EmptySet that is executed, and analogously for NonEmptySet values. This behavior is a direct consequence of our substitution model of evaluation. For instance, (new EmptySet).contains(7) -> (by replacing contains by its body in class EmptySet) false Or, new NonEmptySet(7, new EmptySet, new EmptySet).contains(1) -> (by replacing contains by its body in class NonEmptySet) if (1 < 7) new EmptySet contains 1 else if (1 > 7) new EmptySet contains 1 else true -> (by rewriting the conditional) 36 Classes and Objects new EmptySet contains 1 -> (by replacing contains by its body in class EmptySet) false . Dynamic method dispatch is analogous to higher-order function calls. In both cases, the identity of code to be executed is known only at run-time. This similarity is not just superﬁcial. Indeed, Scala represents every function value as an object (see Section 8.6). Objects. In the previous implementation of integer sets, empty sets were ex- pressed with new EmptySet; so a new object was created every time an empty set value was required. We could have avoided unnecessary object creations by deﬁn- ing a value empty once and then using this value instead of every occurrence of new EmptySet. For example: val EmptySetVal = new EmptySet One problem with this approach is that a value deﬁnition such as the one above is not a legal top-level deﬁnition in Scala; it has to be part of another class or object. Also, the deﬁnition of class EmptySet now seems a bit of an overkill – why deﬁne a class of objects, if we are only interested in a single object of this class? A more direct approach is to use an object deﬁnition. Here is a more streamlined alternative deﬁnition of the empty set: object EmptySet extends IntSet { def contains(x: Int): Boolean = false def incl(x: Int): IntSet = new NonEmptySet(x, EmptySet, EmptySet) } The syntax of an object deﬁnition follows the syntax of a class deﬁnition; it has an optional extends clause as well as an optional body. As is the case for classes, the extends clause deﬁnes inherited members of the object whereas the body de- ﬁnes overriding or new members. However, an object deﬁnition deﬁnes a single object only it is not possible to create other objects with the same structure using new. Therefore, object deﬁnitions also lack constructor parameters, which might be present in class deﬁnitions. Object deﬁnitions can appear anywhere in a Scala program; including at top-level. Since there is no ﬁxed execution order of top-level entities in Scala, one might ask exactly when the object deﬁned by an object deﬁnition is created and initialized. The answer is that the object is created the ﬁrst time one of its members is accessed. This strategy is called lazy evaluation. 37 Standard Classes. Scala is a pure object-oriented language. This means that every value in Scala can be regarded as an object. In fact, even primitive types such as int or boolean are not treated specially. They are deﬁned as type aliases of Scala classes in module Predef: type boolean = scala.Boolean type int = scala.Int type long = scala.Long ... For efﬁciency, the compiler usually represents values of type scala.Int by 32 bit integers, values of type scala.Boolean by Java’s booleans, etc. But it converts these specialized representations to objects when required, for instance when a primitive Int value is passed to a function with a parameter of type AnyRef. Hence, the special representation of primitive values is just an optimization, it does not change the meaning of a program. Here is a speciﬁcation of class Boolean. package scala abstract class Boolean { def && (x: => Boolean): Boolean def || (x: => Boolean): Boolean def ! : Boolean def == (x: Boolean) : Boolean def != (x: Boolean) : Boolean def < (x: Boolean) : Boolean def > (x: Boolean) : Boolean def <= (x: Boolean) : Boolean def >= (x: Boolean) : Boolean } Booleans can be deﬁned using only classes and objects, without reference to a built- in type of booleans or numbers. A possible implementation of class Boolean is given below. This is not the actual implementation in the standard Scala library. For efﬁ- ciency reasons the standard implementation uses built-in booleans. package scala abstract class Boolean { def ifThenElse(thenpart: => Boolean, elsepart: => Boolean) def && (x: => Boolean): Boolean = ifThenElse(x, false) def || (x: => Boolean): Boolean = ifThenElse(true, x) def ! : Boolean = ifThenElse(false, true) def == (x: Boolean) : Boolean = ifThenElse(x, x.!) 38 Classes and Objects def != (x: Boolean) : Boolean = ifThenElse(x.!, x) def < (x: Boolean) : Boolean = ifThenElse(false, x) def > (x: Boolean) : Boolean = ifThenElse(x.!, false) def <= (x: Boolean) : Boolean = ifThenElse(x, true) def >= (x: Boolean) : Boolean = ifThenElse(true, x.!) } case object True extends Boolean { def ifThenElse(t: => Boolean, e: => Boolean) = t } case object False extends Boolean { def ifThenElse(t: => Boolean, e: => Boolean) = e } Here is a partial speciﬁcation of class Int. package scala abstract class Int extends AnyVal { def toLong: Long def toFloat: Float def toDouble: Double def + (that: Double): Double def + (that: Float): Float def + (that: Long): Long def + (that: Int): Int // analogous for -, *, /, % def << (cnt: Int): Int // analogous for >>, >>> def & (that: Long): Long def & (that: Int): Int // analogous for |, ^ def == (that: Double): Boolean def == (that: Float): Boolean def == (that: Long): Boolean // analogous for !=, <, >, <=, >= } Class Int can in principle also be implemented using just objects and classes, with- out reference to a built in type of integers. To see how, we consider a slightly simpler problem, namely how to implement a type Nat of natural (i.e. non-negative) num- bers. Here is the deﬁnition of an abstract class Nat: abstract class Nat { def isZero: Boolean def predecessor: Nat def successor: Nat def + (that: Nat): Nat def - (that: Nat): Nat 39 } To implement the operations of class Nat, we deﬁne a sub-object Zero and a sub- class Succ (for successor). Each number N is represented as N applications of the Succ constructor to Zero: new Succ( ... new Succ | {z }N times (Zero) ... ) The implementation of the Zero object is straightforward: object Zero extends Nat { def isZero: Boolean = true def predecessor: Nat = error("negative number") def successor: Nat = new Succ(Zero) def + (that: Nat): Nat = that def - (that: Nat): Nat = if (that.isZero) Zero else error("negative number") } The implementation of the predecessor and subtraction functions on Zero throws an Error exception, which aborts the program with the given error message. Here is the implementation of the successor class: class Succ(x: Nat) extends Nat { def isZero: Boolean = false def predecessor: Nat = x def successor: Nat = new Succ(this) def + (that: Nat): Nat = x + that.successor def - (that: Nat): Nat = if (that.isZero) this else x - that.predecessor } Note the implementation of method successor. To create the successor of a num- ber, we need to pass the object itself as an argument to the Succ constructor. The object itself is referenced by the reserved name this. The implementations of + and - each contain a recursive call with the constructor argument as receiver. The recursion will terminate once the receiver is the Zero object (which is guaranteed to happen eventually because of the way numbers are formed). Exercise 6.0.3 Write an implementation Integer of integer numbers The imple- mentation should support all operations of class Nat while adding two methods def isPositive: Boolean def negate: Integer 40 Classes and Objects The ﬁrst method should return true if the number is positive. The second method should negate the number. Do not use any of Scala’s standard numeric classes in your implementation. (Hint: There are two possible ways to implement Integer. One can either make use the existing implementation of Nat, representing an inte- ger as a natural number and a sign. Or one can generalize the given implementation of Nat to Integer, using the three subclasses Zero for 0, Succ for positive numbers and Pred for negative numbers.) Language Elements Introduced In This Chapter Types: Type = ... | ident Types can now be arbitrary identiﬁers which represent classes. Expressions: Expr = ... | Expr ’.’ ident | ’new’ Expr | ’this’ An expression can now be an object creation, or a selection E.m of a member m from an object-valued expression E, or it can be the reserved name this. Deﬁnitions and Declarations: Def = FunDef | ValDef | ClassDef | TraitDef | ObjectDef ClassDef = [’abstract’] ’class’ ident [’(’ [Parameters] ’)’] [’extends’ Expr] [‘{’ {TemplateDef} ‘}’] TraitDef = ’trait’ ident [’extends’ Expr] [’{’ {TemplateDef} ’}’] ObjectDef = ’object’ ident [’extends’ Expr] [’{’ {ObjectDef} ’}’] TemplateDef = [Modifier] (Def | Dcl) ObjectDef = [Modifier] Def Modifier = ’private’ | ’override’ Dcl = FunDcl | ValDcl FunDcl = ’def’ ident {’(’ [Parameters] ’)’} ’:’ Type ValDcl = ’val’ ident ’:’ Type A deﬁnition can now be a class, trait or object deﬁnition such as class C(params) extends B { defs } trait T extends B { defs } object O extends B { defs } The deﬁnitions defs in a class, trait or object may be preceded by modiﬁers private or override. Abstract classes and traits may also contain declarations. These introduce deferred functions or values with their types, but do not give an implementation. Deferred members have to be implemented in subclasses before objects of an abstract class 41 or trait can be created. Chapter 7 Case Classes and Pattern Match- ing Say, we want to write an interpreter for arithmetic expressions. To keep things sim- ple initially, we restrict ourselves to just numbers and + operations. Such expres- sions can be represented as a class hierarchy, with an abstract base class Expr as the root, and two subclasses Number and Sum. Then, an expression 1 + (3 + 7) would be represented as new Sum(new Number(1), new Sum(new Number(3), new Number(7))) Now, an evaluator of an expression like this needs to know of what form it is (either Sum or Number) and also needs to access the components of the expression. The following implementation provides all necessary methods. abstract class Expr { def isNumber: Boolean def isSum: Boolean def numValue: Int def leftOp: Expr def rightOp: Expr } class Number(n: Int) extends Expr { def isNumber: Boolean = true def isSum: Boolean = false def numValue: Int = n def leftOp: Expr = error("Number.leftOp") def rightOp: Expr = error("Number.rightOp") } class Sum(e1: Expr, e2: Expr) extends Expr { def isNumber: Boolean = false def isSum: Boolean = true 44 Case Classes and Pattern Matching def numValue: Int = error("Sum.numValue") def leftOp: Expr = e1 def rightOp: Expr = e2 } With these classiﬁcation and access methods, writing an evaluator function is sim- ple: def eval(e: Expr): Int = { if (e.isNumber) e.numValue else if (e.isSum) eval(e.leftOp) + eval(e.rightOp) else error("unrecognized expression kind") } However, deﬁning all these methods in classes Sum and Number is rather tedious. Furthermore, the problem becomes worse when we want to add new forms of ex- pressions. For instance, consider adding a new expression form Prod for products. Not only do we have to implement a new class Prod, with all previous classiﬁcation and access methods; we also have to introduce a new abstract method isProduct in class Expr and implement that method in subclasses Number, Sum, and Prod. Having to modify existing code when a system grows is always problematic, since it intro- duces versioning and maintenance problems. The promise of object-oriented programming is that such modiﬁcations should be unnecessary, because they can be avoided by re-using existing, unmodiﬁed code through inheritance. Indeed, a more object-oriented decomposition of our prob- lem solves the problem. The idea is to make the “high-level” operation eval a method of each expression class, instead of implementing it as a function outside the expression class hierarchy, as we have done before. Because eval is now a mem- ber of all expression nodes, all classiﬁcation and access methods become superﬂu- ous, and the implementation is simpliﬁed considerably: abstract class Expr { def eval: Int } class Number(n: Int) extends Expr { def eval: Int = n } class Sum(e1: Expr, e2: Expr) extends Expr { def eval: Int = e1.eval + e2.eval } Furthermore, adding a new Prod class does not entail any changes to existing code: class Prod(e1: Expr, e2: Expr) extends Expr { def eval: Int = e1.eval * e2.eval } 45 The conclusion we can draw from this example is that object-oriented decomposi- tion is the technique of choice for constructing systems that should be extensible with new types of data. But there is also another possible way we might want to ex- tend the expression example. We might want to add new operations on expressions. For instance, we might want to add an operation that pretty-prints an expression tree to standard output. If we have deﬁned all classiﬁcation and access methods, such an operation can eas- ily be written as an external function. Here is an example: def print(e: Expr) { if (e.isNumber) Console.print(e.numValue) else if (e.isSum) { Console.print("(") print(e.leftOp) Console.print("+") print(e.rightOp) Console.print(")") } else error("unrecognized expression kind") } However, if we had opted for an object-oriented decomposition of expressions, we would need to add a new print procedure to each class: abstract class Expr { def eval: Int def print } class Number(n: Int) extends Expr { def eval: Int = n def print { Console.print(n) } } class Sum(e1: Expr, e2: Expr) extends Expr { def eval: Int = e1.eval + e2.eval def print { Console.print("(") print(e1) Console.print("+") print(e2) Console.print(")") } } Hence, classical object-oriented decomposition requires modiﬁcation of all existing classes when a system is extended with new operations. 46 Case Classes and Pattern Matching As yet another way we might want to extend the interpreter, consider expression simpliﬁcation. For instance, we might want to write a function which rewrites ex- pressions of the form a * b + a * c to a *(b + c). This operation requires in- spection of more than a single node of the expression tree at the same time. Hence, it cannot be implemented by a method in each expression kind, unless that method can also inspect other nodes. So we are forced to have classiﬁcation and access methods in this case. This seems to bring us back to square one, with all the prob- lems of verbosity and extensibility. Taking a closer look, one observes that the only purpose of the classiﬁcation and access functions is to reverse the data construction process. They let us determine, ﬁrst, which sub-class of an abstract base class was used and, second, what were the constructor arguments. Since this situation is quite common, Scala has a way to automate it with case classes. 7.1 Case Classes and Case Objects Case classes and case objects are deﬁned like a normal classes or objects, except that the deﬁnition is preﬁxed with the modiﬁer case. For instance, the deﬁnitions abstract class Expr case class Number(n: Int) extends Expr case class Sum(e1: Expr, e2: Expr) extends Expr introduce Number and Sum as case classes. The case modiﬁer in front of a class or object deﬁnition has the following effects. 1. Case classes implicitly come with a constructor function, with the same name as the class. In our example, the two functions def Number(n: Int) = new Number(n) def Sum(e1: Expr, e2: Expr) = new Sum(e1, e2) would be added. Hence, one can now construct expression trees a bit more concisely, as in Sum(Sum(Number(1), Number(2)), Number(3)) 2. Case classes and case objects implicitly come with implementations of meth- ods toString, equals and hashCode, which override the methods with the same name in class AnyRef. The implementation of these methods takes in each case the structure of a member of a case class into account. The toString method represents an expression tree the way it was constructed. So, Sum(Sum(Number(1), Number(2)), Number(3)) 7.2 Pattern Matching 47 would be converted to exactly that string, whereas the default implementa- tion in class AnyRef would return a string consisting of the outermost con- structor name Sum and a number. The equals methods treats two case mem- bers of a case class as equal if they have been constructed with the same con- structor and with arguments which are themselves pairwise equal. This also affects the implementation of == and !=, which are implemented in terms of equals in Scala. So, Sum(Number(1), Number(2)) == Sum(Number(1), Number(2)) will yield true. If Sum or Number were not case classes, the same expression would be false, since the standard implementation of equals in class AnyRef always treats objects created by different constructor calls as being differ- ent. The hashCode method follows the same principle as other two meth- ods. It computes a hash code from the case class constructor name and the hash codes of the constructor arguments, instead of from the object’s address, which is what the as the default implementation of hashCode does. 3. Case classes implicitly come with nullary accessor methods which retrieve the constructor arguments. In our example, Number would obtain an acces- sor method def n: Int which returns the constructor parameter n, whereas Sum would obtain two accessor methods def e1: Expr, e2: Expr Hence, if for a value s of type Sum, say, one can now write s.e1, to access the left operand. However, for a value e of type Expr, the term e.e1 would be illegal since e1 is deﬁned in Sum; it is not a member of the base class Expr. So, how do we determine the constructor and access constructor arguments for values whose static type is the base class Expr? This is solved by the fourth and ﬁnal particularity of case classes. 4. Case classes allow the constructions of patterns which refer to the case class constructor. 7.2 Pattern Matching Pattern matching is a generalization of C or Java’s switch statement to class hier- archies. Instead of a switch statement, there is a standard method match, which is deﬁned in Scala’s root class Any, and therefore is available for all objects. The match method takes as argument a number of cases. For instance, here is an implementa- tion of eval using pattern matching. 48 Case Classes and Pattern Matching def eval(e: Expr): Int = e match { case Number(n) => n case Sum(l, r) => eval(l) + eval(r) } In this example, there are two cases. Each case associates a pattern with an expres- sion. Patterns are matched against the selector values e. The ﬁrst pattern in our example, Number(n), matches all values of the form Number(v), where v is an arbi- trary value. In that case, the pattern variable n is bound to the value v. Similarly, the pattern Sum(l, r) matches all selector values of form Sum(v1, v2) and binds the pattern variables l and r to v1 and v2, respectively. In general, patterns are built from • Case class constructors, e.g. Number, Sum, whose arguments are again patterns, • pattern variables, e.g. n, e1, e2, • the “wildcard” pattern _, • literals, e.g. 1, true, "abc", • constant identiﬁers, e.g. MAXINT, EmptySet. Pattern variables always start with a lower-case letter, so that they can be distin- guished from constant identiﬁers, which start with an upper case letter. Each vari- able name may occur only once in a pattern. For instance, Sum(x, x) would be illegal as a pattern, since the pattern variable x occurs twice in it. Meaning of Pattern Matching. A pattern matching expression e match { case p1 => e1 ... case pn => en } matches the patterns p1,..., pn in the order they are written against the selector value e. • A constructor pattern C(p1,..., pn) matches all values that are of type C(or a subtype thereof) and that have been constructed with C-arguments matching patterns p1,..., pn. • A variable pattern x matches any value and binds the variable name to that value. • The wildcard pattern ‘_’ matches any value but does not bind a name to that value. • A constant pattern C matches a value which is equal (in terms of ==) to C. 7.2 Pattern Matching 49 The pattern matching expression rewrites to the right-hand-side of the ﬁrst case whose pattern matches the selector value. References to pattern variables are re- placed by corresponding constructor arguments. If none of the patterns matches, the pattern matching expression is aborted with a MatchError exception. Example 7.2.1 Our substitution model of program evaluation extends quite natu- rally to pattern matching, For instance, here is how eval applied to a simple expres- sion is re-written: eval(Sum(Number(1), Number(2))) -> (by rewriting the application) Sum(Number(1), Number(2)) match { case Number(n) => n case Sum(e1, e2) => eval(e1) + eval(e2) } -> (by rewriting the pattern match) eval(Number(1)) + eval(Number(2)) -> (by rewriting the ﬁrst application) Number(1) match { case Number(n) => n case Sum(e1, e2) => eval(e1) + eval(e2) } + eval(Number(2)) -> (by rewriting the pattern match) 1 + eval(Number(2)) ->∗ 1 + 2 -> 3 Pattern Matching and Methods. In the previous example, we have used pattern matching in a function which was deﬁned outside the class hierarchy over which it matches. Of course, it is also possible to deﬁne a pattern matching function in that class hierarchy itself. For instance, we could have deﬁned eval is a method of the base class Expr, and still have used pattern matching in its implementation: abstract class Expr { def eval: Int = this match { case Number(n) => n case Sum(e1, e2) => e1.eval + e2.eval 50 Case Classes and Pattern Matching } } Exercise 7.2.2 Consider the following deﬁnitions representing trees of integers. These deﬁnitions can be seen as an alternative representation of IntSet: abstract class IntTree case object EmptyTree extends IntTree case class Node(elem: Int, left: IntTree, right: IntTree) extends IntTree Complete the following implementations of function contains and insert for IntTree’s. def contains(t: IntTree, v: Int): Boolean = t match { ... ... } def insert(t: IntTree, v: Int): IntTree = t match { ... ... } Pattern Matching Anonymous Functions. So far, case-expressions always ap- peared in conjunction with a match operation. But it is also possible to use case- expressions by themselves. A block of case-expressions such as { case P1 => E1 ... case Pn => En } is seen by itself as a function which matches its arguments against the patterns P1,...,Pn, and produces the result of one of E1,...,En. (If no pattern matches, the function would throw a MatchError exception instead). In other words, the expres- sion above is seen as a shorthand for the anonymous function (x => x match { case P1 => E1 ... case Pn => En }) where x is a fresh variable which is not used otherwise in the expression. Chapter 8 Generic Types and Methods Classes in Scala can have type parameters. We demonstrate the use of type parame- ters with functional stacks as an example. Say, we want to write a data type of stacks of integers, with methods push, top, pop, and isEmpty. This is achieved by the fol- lowing class hierarchy: abstract class IntStack { def push(x: Int): IntStack = new IntNonEmptyStack(x, this) def isEmpty: Boolean def top: Int def pop: IntStack } class IntEmptyStack extends IntStack { def isEmpty = true def top = error("EmptyStack.top") def pop = error("EmptyStack.pop") } class IntNonEmptyStack(elem: Int, rest: IntStack) extends IntStack { def isEmpty = false def top = elem def pop = rest } Of course, it would also make sense to deﬁne an abstraction for a stack of Strings. To do that, one could take the existing abstraction for IntStack, rename it to StringStack and at the same time rename all occurrences of type Int to String. A better way, which does not entail code duplication, is to parameterize the stack deﬁnitions with the element type. Parameterization lets us generalize from a spe- ciﬁc instance of a problem to a more general one. So far, we have used parameteri- zation only for values, but it is available also for types. To arrive at a generic version of Stack, we equip it with a type parameter. 52 Generic Types and Methods abstract class Stack[A] { def push(x: A): Stack[A] = new NonEmptyStack[A](x, this) def isEmpty: Boolean def top: A def pop: Stack[A] } class EmptyStack[A] extends Stack[A] { def isEmpty = true def top = error("EmptyStack.top") def pop = error("EmptyStack.pop") } class NonEmptyStack[A](elem: A, rest: Stack[A]) extends Stack[A] { def isEmpty = false def top = elem def pop = rest } In the deﬁnitions above, ‘A’ is a type parameter of class Stack and its subclasses. Type parameters are arbitrary names; they are enclosed in brackets instead of parentheses, so that they can be easily distinguished from value parameters. Here is an example how the generic classes are used: val x = new EmptyStack[Int] val y = x.push(1).push(2) println(y.pop.top) The ﬁrst line creates a new empty stack of Int’s. Note the actual type argument [Int] which replaces the formal type parameter A. It is also possible to parameterize methods with types. As an example, here is a generic method which determines whether one stack is a preﬁx of another. def isPrefix[A](p: Stack[A], s: Stack[A]): Boolean = { p.isEmpty || p.top == s.top && isPrefix[A](p.pop, s.pop) } The method parameters are called polymorphic. Generic methods are also called polymorphic. The term comes from the Greek, where it means “having many forms”. To apply a polymorphic method such as isPrefix, we pass type parameters as well as value parameters to it. For instance, val s1 = new EmptyStack[String].push("abc") val s2 = new EmptyStack[String].push("abx").push(s1.top) println(isPrefix[String](s1, s2)) 8.1 Type Parameter Bounds 53 Local Type Inference. Passing type parameters such as [Int] or [String] all the time can become tedious in applications where generic functions are used a lot. Quite often, the information in a type parameter is redundant, because the correct parameter type can also be determined by inspecting the function’s value parame- ters or expected result type. Taking the expression isPrefix[String](s1, s2) as an example, we know that its value parameters are both of type Stack[String], so we can deduce that the type parameter must be String. Scala has a fairly powerful type inferencer which allows one to omit type parameters to polymorphic functions and constructors in situations like these. In the example above, one could have writ- ten isPrefix(s1, s2) and the missing type argument [String] would have been inserted by the type inferencer. 8.1 Type Parameter Bounds Now that we know how to make classes generic it is natural to generalize some of the earlier classes we have written. For instance class IntSet could be generalized to sets with arbitrary element types. Let’s try. The abstract class for generic sets is easily written. abstract class Set[A] { def incl(x: A): Set[A] def contains(x: A): Boolean } However, if we still want to implement sets as binary search trees, we encounter a problem. The contains and incl methods both compare elements using methods < and >. For IntSet this was OK, since type Int has these two methods. But for an arbitrary type parameter a, we cannot guarantee this. Therefore, the previous implementation of, say, contains would generate a compiler error. def contains(x: Int): Boolean = if (x < elem) left contains x ^ < not a member of type A. One way to solve the problem is to restrict the legal types that can be substituted for type A to only those types that contain methods < and > of the correct types. There is a trait Ordered[A] in the standard class library Scala which represents values which are comparable (via < and >) to values of type A. This trait is deﬁned as follows: /** A class for totally ordered data. */ trait Ordered[A] { /** Result of comparing ‘this’ with operand ‘that’. * returns ‘x’ where * x < 0 iff this < that 54 Generic Types and Methods * x == 0 iff this == that * x > 0 iff this > that */ def compare(that: A): Int def < (that: A): Boolean = (this compare that) < 0 def > (that: A): Boolean = (this compare that) > 0 def <= (that: A): Boolean = (this compare that) <= 0 def >= (that: A): Boolean = (this compare that) >= 0 def compareTo(that: A): Int = compare(that) } We can enforce the comparability of a type by demanding that the type is a subtype of Ordered. This is done by giving an upper bound to the type parameter of Set: trait Set[A <: Ordered[A]] { def incl(x: A): Set[A] def contains(x: A): Boolean } The parameter declaration A <: Ordered[A] introduces A as a type parameter which must be a subtype of Ordered[A], i.e. its values must be comparable to values of the same type. With this restriction, we can now implement the rest of the generic set abstraction as we did in the case of IntSets before. class EmptySet[A <: Ordered[A]] extends Set[A] { def contains(x: A): Boolean = false def incl(x: A): Set[A] = new NonEmptySet(x, new EmptySet[A], new EmptySet[A]) } class NonEmptySet[A <: Ordered[A]] (elem: A, left: Set[A], right: Set[A]) extends Set[A] { def contains(x: A): Boolean = if (x < elem) left contains x else if (x > elem) right contains x else true def incl(x: A): Set[A] = if (x < elem) new NonEmptySet(elem, left incl x, right) else if (x > elem) new NonEmptySet(elem, left, right incl x) else this } Note that we have left out the type argument in the object creations new NonEmptySet(...). In the same way as for polymorphic methods, missing type arguments in constructor calls are inferred from value arguments and/or the ex- 8.1 Type Parameter Bounds 55 pected result type. Here is an example that uses the generic set abstraction. Let’s ﬁrst create a subclass of Ordered, like this: case class Num(value: Double) extends Ordered[Num] { def compare(that: Num): Int = if (this.value < that.value) -1 else if (this.value > that.value) 1 else 0 } Then: val s = new EmptySet[Num].incl(Num(1.0)).incl(Num(2.0)) s.contains(Num(1.5)) This is OK, as type Num implements the trait Ordered[Num]. However, the following example is in error. val s = new EmptySet[java.io.File] ^ java.io.File does not conform to type parameter bound Ordered[java.io.File]. One probem with type parameter bounds is that they require forethought: if we had not declared Num a subclass of Ordered, we would not have been able to use Num elements in sets. By the same token, types inherited from Java, such as Int, Double, or String are not subclasses of Ordered, so values of these types cannot be used as set elements. A more ﬂexible design, which admits elements of these types, uses view bounds in- stead of the plain type bounds we have seen so far. The only change this entails in the example above is in the type parameters: trait Set[A <% Ordered[A]] ... class EmptySet[A <% Ordered[A]] ... class NonEmptySet[A <% Ordered[A]] ... View bounds <% are weaker than plain bounds <:: A view bounded type parameter clause [A <% T] only speciﬁes that the bounded type A must be convertible to the bound type T, using an implicit conversion. The Scala library predeﬁnes implicit conversions for a number of types, including the primitive types and String. Therefore, the redesign set abstraction can be in- stantiated with these types as well. More explanations on implicit conversions and view bounds are given in Section15. 56 Generic Types and Methods 8.2 Variance Annotations The combination of type parameters and subtyping poses some interesting ques- tions. For instance, should Stack[String] be a subtype of Stack[AnyRef]? Intu- itively, this seems OK, since a stack of Strings is a special case of a stack of AnyRefs. More generally, if T is a subtype of type S then Stack[T] should be a subtype of Stack[S]. This property is called co-variant subtyping. In Scala, generic types have by default non-variant subtyping. That is, with Stack deﬁned as above, stacks with different element types would never be in a subtype relation. However, we can enforce co-variant subtyping of stacks by changing the ﬁrst line of the deﬁnition of class Stack as follows. class Stack[+A] { Preﬁxing a formal type parameter with a + indicates that subtyping is covariant in that parameter. Besides +, there is also a preﬁx - which indicates contra-variant subtyping. If Stack was deﬁned class Stack[-A] ..., then T a subtype of type S would imply that Stack[S] is a subtype of Stack[T] (which in the case of stacks would be rather surprising!). In a purely functional world, all types could be co-variant. However, the situation changes once we introduce mutable data. Consider the case of arrays in Java or .NET. Such arrays are represented in Scala by a generic class Array. Here is a partial deﬁnition of this class. class Array[A] { def apply(index: Int): A def update(index: Int, elem: A) } The class above deﬁnes the way Scala arrays are seen from Scala user programs. The Scala compiler will map this abstraction to the underlying arrays of the host system in most cases where this possible. In Java, arrays are indeed covariant; that is, for reference types T and S, if T is a sub- type of S, then also Array[T] is a subtype of Array[S]. This might seem natural but leads to safety problems that require special runtime checks. Here is an example: val x = new Array[String](1) val y: Array[Any] = x y(0) = new Rational(1, 2) // this is syntactic sugar for // y.update(0, new Rational(1, 2)) In the ﬁrst line, a new array of strings is created. In the second line, this array is bound to a variable y, of type Array[Any]. Assuming arrays are covariant, this is OK, since Array[String] is a subtype of Array[Any]. Finally, in the last line a rational number is stored in the array. This is also OK, since type Rational is a subtype of 8.2 Variance Annotations 57 the element type Any of the array y. We thus end up storing a rational number in an array of strings, which clearly violates type soundness. Java solves this problem by introducing a run-time check in the third line which tests whether the stored element is compatible with the element type with which the array was created. We have seen in the example that this element type is not necessarily the static element type of the array being updated. If the test fails, an ArrayStoreException is raised. Scala solves this problem instead statically, by disallowing the second line at compile-time, because arrays in Scala have non-variant subtyping. This raises the question how a Scala compiler veriﬁes that variance annotations are correct. If we had simply declared arrays co-variant, how would the potential problem have been detected? Scala uses a conservative approximation to verify soundness of variance annota- tions. A covariant type parameter of a class may only appear in co-variant posi- tions inside the class. Among the co-variant positions are the types of values in the class, the result types of methods in the class, and type arguments to other covariant types. Not co-variant are types of formal method parameters. Hence, the following class deﬁnition would have been rejected class Array[+A] { def apply(index: Int): A def update(index: Int, elem: A) ^ covariant type parameter A appears in contravariant position. } So far, so good. Intuitively, the compiler was correct in rejecting the update proce- dure in a co-variant class because update potentially changes state, and therefore undermines the soundness of co-variant subtyping. However, there are also methods which do not mutate state, but where a type pa- rameter still appears contra-variantly. An example is push in type Stack. Again the Scala compiler will reject the deﬁnition of this method for co-variant stacks. class Stack[+A] { def push(x: A): Stack[A] = ^ covariant type parameter A appears in contravariant position. This is a pity, because, unlike arrays, stacks are purely functional data structures and therefore should enable co-variant subtyping. However, there is a a way to solve the problem by using a polymorphic method with a lower type parameter bound. 58 Generic Types and Methods 8.3 Lower Bounds We have seen upper bounds for type parameters. In a type parameter declaration such as T <: U, the type parameter T is restricted to range only over subtypes of type U. Symmetrical to this are lower bounds in Scala. In a type parameter declaration T >: S, the type parameter T is restricted to range only over supertypes of type S. (One can also combine lower and upper bounds, as in T >: S <: U.) Using lower bounds, we can generalize the push method in Stack as follows. class Stack[+A] { def push[B >: A](x: B): Stack[B] = new NonEmptyStack(x, this) Technically, this solves our variance problem since now the type parameter A ap- pears no longer as a parameter type of method push. Instead, it appears as lower bound for another type parameter of a method, which is classiﬁed as a co-variant position. Hence, the Scala compiler accepts the new deﬁnition of push. In fact, we have not only solved the technical variance problem but also have gen- eralized the deﬁnition of push. Before, we were required to push only elements with types that conform to the declared element type of the stack. Now, we can push also elements of a supertype of this type, but the type of the returned stack will change accordingly. For instance, we can now push an AnyRef onto a stack of Strings, but the resulting stack will be a stack of AnyRefs instead of a stack of Strings! In summary, one should not hesitate to add variance annotations to your data struc- tures, as this yields rich natural subtyping relationships. The compiler will detect potential soundness problems. Even if the compiler’s approximation is too conser- vative, as in the case of method push of class Stack, this will often suggest a useful generalization of the contested method. 8.4 Least Types Scala does not allow one to parameterize objects with types. That’s why we orig- inally deﬁned a generic class EmptyStack[A], even though a single value denoting empty stacks of arbitrary type would do. For co-variant stacks, however, one can use the following idiom: object EmptyStack extends Stack[Nothing] { ... } The bottom type Nothing contains no value, so the type Stack[Nothing] expresses the fact that an EmptyStack contains no elements. Furthermore, Nothing is a sub- type of all other types. Hence, for co-variant stacks, Stack[Nothing] is a subtype of Stack[T], for any other type T. This makes it possible to use a single empty stack object in user code. For instance: 8.4 Least Types 59 val s = EmptyStack.push("abc").push(new AnyRef()) Let’s analyze the type assignment for this expression in detail. The EmptyStack ob- ject is of type Stack[Nothing], which has a method push[B >: Nothing](elem: B): Stack[B] . Local type inference will determine that the type parameter B should be instanti- ated to String in the application EmptyStack.push("abc"). The result type of that application is hence Stack[String], which in turn has a method push[B >: String](elem: B): Stack[B] . The ﬁnal part of the value deﬁnition above is the application of this method to new AnyRef(). Local type inference will determine that the type parameter b should this time be instantiated to AnyRef, with result type Stack[AnyRef]. Hence, the type assigned to value s is Stack[AnyRef]. Besides Nothing, which is a subtype of every other type, there is also the type Null, which is a subtype of scala.AnyRef, and every class derived from it. The null lit- eral in Scala is the only value of that type. This makes null compatible with every reference type, but not with a value type such as Int. We conclude this section with the complete improved deﬁnition of stacks. Stacks have now co-variant subtyping, the push method has been generalized, and the empty stack is represented by a single object. abstract class Stack[+A] { def push[B >: A](x: B): Stack[B] = new NonEmptyStack(x, this) def isEmpty: Boolean def top: A def pop: Stack[A] } object EmptyStack extends Stack[Nothing] { def isEmpty = true def top = error("EmptyStack.top") def pop = error("EmptyStack.pop") } class NonEmptyStack[+A](elem: A, rest: Stack[A]) extends Stack[A] { def isEmpty = false def top = elem def pop = rest } Many classes in the Scala library are generic. We now present two commonly used families of generic classes, tuples and functions. The discussion of another com- mon class, lists, is deferred to the next chapter. 60 Generic Types and Methods 8.5 Tuples Sometimes, a function needs to return more than one result. For instance, take the function divmod which returns the integer quotient and rest of two given integer arguments. Of course, one can deﬁne a class to hold the two results of divmod, as in: case class TwoInts(first: Int, second: Int) def divmod(x: Int, y: Int): TwoInts = new TwoInts(x / y, x % y) However, having to deﬁne a new class for every possible pair of result types is very tedious. In Scala one can use instead a generic class Tuple2, which is deﬁned as follows: package scala case class Tuple2[A, B](_1: A, _2: B) With Tuple2, the divmod method can be written as follows. def divmod(x: Int, y: Int) = new Tuple2[Int, Int](x / y, x % y) As usual, type parameters to constructors can be omitted if they are deducible from value arguments. There exist also tuple classes for every other number of elements (the current Scala implementation limits this to tuples of some reasonable number of elements). How are elements of tuples accessed? Since tuples are case classes, there are two possibilities. One can either access a tuple’s ﬁelds using the names of the construc- tor parameters _i, as in the following example: val xy = divmod(x, y) println("quotient: " + xy._1 + ", rest: " + xy._2) Or one uses pattern matching on tuples, as in the following example: divmod(x, y) match { case Tuple2(n, d) => println("quotient: " + n + ", rest: " + d) } Note that type parameters are never used in patterns; it would have been illegal to write case Tuple2[Int, Int](n, d). Tuples are so convenient that Scala deﬁnes special syntax for them. To form a tuple with n elements x1,..., xn one can write (x1,..., xn). This is equivalent to Tuplen(x1,..., xn). The (...) syntax works equivalently for types and for patterns. With that tuple syntax, the divmod example is written as follows: def divmod(x: Int, y: Int): (Int, Int) = (x / y, x % y) 8.6 Functions 61 divmod(x, y) match { case (n, d) => println("quotient: " + n + ", rest: " + d) } 8.6 Functions Scala is a functional language in that functions are ﬁrst-class values. Scala is also an object-oriented language in that every value is an object. It follows that functions are objects in Scala. For instance, a function from type String to type Int is repre- sented as an instance of the trait Function1[String, Int]. The Function1 trait is deﬁned as follows. package scala trait Function1[-A, +B] { def apply(x: A): B } Besides Function1, there are also deﬁnitions of for functions of all other arities (the current implementation implements this only up to a reasonable limit). That is, there is one deﬁnition for each possible number of function parameters. Scala’s function type syntax (T1,...,Tn) => S is simply an abbreviation for the parame- terized type Functionn[T1,...,Tn,S]. Scala uses the same syntax f (x) for function application, no matter whether f is a method or a function object. This is made possible by the following convention: A function application f (x) where f is an object (as opposed to a method) is taken to be a shorthand for f .apply(x). Hence, the apply method of a function type is inserted automatically where this is necessary. That’s also why we deﬁned array subscripting in Section 8.2 by an apply method. For any array a, the subscript operation a(i) is taken to be a shorthand for a.apply(i). Functions are an example where a contra-variant type parameter declaration is use- ful. For example, consider the following code: val f: (AnyRef => Int) = x => x.hashCode() val g: (String => Int) = f g("abc") It’s sound to bind the value g of type String => Int to f, which is of type AnyRef => Int. Indeed, all one can do with function of type String => Int is pass it a string in order to obtain an integer. Clearly, the same works for function f: If we pass it a string (or any other object), we obtain an integer. This demonstrates that function subtyping is contra-variant in its argument type whereas it is covariant in its result type. In short, S ⇒ T is a subtype of S0 ⇒ T 0, provided S0 is a subtype of S 62 Generic Types and Methods and T is a subtype of T 0. Example 8.6.1 Consider the Scala code val plus1: (Int => Int) = (x: Int) => x + 1 plus1(2) This is expanded into the following object code. val plus1: Function1[Int, Int] = new Function1[Int, Int] { def apply(x: Int): Int = x + 1 } plus1.apply(2) Here, the object creation new Function1[Int, Int]{ ... } represents an instance of an anonymous class. It combines the creation of a new Function1 object with an implementation of the apply method (which is abstract in Function1). Equivalently, but more verbosely, one could have used a local class: val plus1: Function1[Int, Int] = { class Local extends Function1[Int, Int] { def apply(x: Int): Int = x + 1 } new Local: Function1[Int, Int] } plus1.apply(2) Chapter 9 Lists Lists are an important data structure in many Scala programs. A list containing the elements x1,..., xn is written List(x1, ..., xn). Examples are: val fruit = List("apples", "oranges", "pears") val nums = List(1, 2, 3, 4) val diag3 = List(List(1, 0, 0), List(0, 1, 0), List(0, 0, 1)) val empty = List() Lists are similar to arrays in languages such as C or Java, but there are also three important differences. First, lists are immutable. That is, elements of a list cannot be changed by assignment. Second, lists have a recursive structure, whereas arrays are ﬂat. Third, lists support a much richer set of operations than arrays usually do. 9.1 Using Lists The List type. Like arrays, lists are homogeneous. That is, the elements of a list all have the same type. The type of a list with elements of type T is written List[T] (compare to T[] in Java). val fruit: List[String] = List("apples", "oranges", "pears") val nums : List[Int] = List(1, 2, 3, 4) val diag3: List[List[Int]] = List(List(1, 0, 0), List(0, 1, 0), List(0, 0, 1)) val empty: List[Int] = List() List constructors. All lists are built from two more fundamental constructors, Nil and :: (pronounced “cons”). Nil represents an empty list. The inﬁx operator :: expresses list extension. That is, x :: xs represents a list whose ﬁrst element is x, which is followed by (the elements of) list xs. Hence, the list values above could also 64 Lists have been deﬁned as follows (in fact their previous deﬁnition is simply syntactic sugar for the deﬁnitions below). val fruit = "apples" :: ("oranges" :: ("pears" :: Nil)) val nums = 1 :: (2 :: (3 :: (4 :: Nil))) val diag3 = (1 :: (0 :: (0 :: Nil))) :: (0 :: (1 :: (0 :: Nil))) :: (0 :: (0 :: (1 :: Nil))) :: Nil val empty = Nil The ‘::’ operation associates to the right: A :: B :: C is interpreted as A :: (B :: C). Therefore, we can drop the parentheses in the deﬁnitions above. For instance, we can write shorter val nums = 1 :: 2 :: 3 :: 4 :: Nil Basic operations on lists. All operations on lists can be expressed in terms of the following three: head returns the ﬁrst element of a list, tail returns the list consisting of all elements except the ﬁrst element, isEmpty returns true iff the list is empty These operations are deﬁned as methods of list objects. So we invoke them by se- lecting from the list that’s operated on. Examples: empty.isEmpty = true fruit.isEmpty = false fruit.head = "apples" fruit.tail.head = "oranges" diag3.head = List(1, 0, 0) The head and tail methods are deﬁned only for non-empty lists. When selected from an empty list, they throw an exception. As an example of how lists can be processed, consider sorting the elements of a list of numbers into ascending order. One simple way to do so is insertion sort, which works as follows: To sort a non-empty list with ﬁrst element x and rest xs, sort the remainder xs and insert the element x at the right position in the result. Sorting an empty list will yield the empty list. Expressed as Scala code: def isort(xs: List[Int]): List[Int] = if (xs.isEmpty) Nil else insert(xs.head, isort(xs.tail)) Exercise 9.1.1 Provide an implementation of the missing function insert. 9.2 Deﬁnition of class List I: First Order Methods 65 List patterns. In fact, :: is deﬁned as a case class in Scala’s standard library. Hence, it is possible to decompose lists by pattern matching, using patterns com- posed from the Nil and :: constructors. For instance, isort can be written alterna- tively as follows. def isort(xs: List[Int]): List[Int] = xs match { case List() => List() case x :: xs1 => insert(x, isort(xs1)) } where def insert(x: Int, xs: List[Int]): List[Int] = xs match { case List() => List(x) case y :: ys => if (x <= y) x :: xs else y :: insert(x, ys) } 9.2 Deﬁnition of class List I: First Order Methods Lists are not built in in Scala; they are deﬁned by an abstract class List, which comes with two subclasses for :: and Nil. In the following we present a tour through class List. package scala abstract class List[+A] { List is an abstract class, so one cannot deﬁne elements by calling the empty List constructor (e.g. by new List). The class has a type parameter a. It is co-variant in this parameter, which means that List[S] <: List[T] for all types S and T such that S <: T. The class is situated in the package scala. This is a package containing the most important standard classes of Scala. List deﬁnes a number of methods, which are explained in the following. Decomposing lists. First, there are the three basic methods isEmpty, head, tail. Their implementation in terms of pattern matching is straightforward: def isEmpty: Boolean = this match { case Nil => true case x :: xs => false } def head: A = this match { case Nil => error("Nil.head") case x :: xs => x } def tail: List[A] = this match { 66 Lists case Nil => error("Nil.tail") case x :: xs => xs } The next function computes the length of a list. def length: Int = this match { case Nil => 0 case x :: xs => 1 + xs.length } Exercise 9.2.1 Design a tail-recursive version of length. The next two functions are the complements of head and tail. def last: A def init: List[A] xs.last returns the last element of list xs, whereas xs.init returns all elements of xs except the last. Both functions have to traverse the entire list, and are thus less efﬁcient than their head and tail analogues. Here is the implementation of last. def last: A = this match { case Nil => error("Nil.last") case x :: Nil => x case x :: xs => xs.last } The implementation of init is analogous. The next three functions return a preﬁx of the list, or a sufﬁx, or both. def take(n: Int): List[A] = if (n == 0 || isEmpty) Nil else head :: tail.take(n-1) def drop(n: Int): List[A] = if (n == 0 || isEmpty) this else tail.drop(n-1) def split(n: Int): (List[A], List[A]) = (take(n), drop(n)) (xs take n) returns the ﬁrst n elements of list xs, or the whole list, if its length is smaller than n.(xs drop n) returns all elements of xs except the n ﬁrst ones. Fi- nally, (xs split n) returns a pair consisting of the lists resulting from xs take n and xs drop n. The next function returns an element at a given index in a list. It is thus analogous to array subscripting. Indices start at 0. 9.2 Deﬁnition of class List I: First Order Methods 67 def apply(n: Int): A = drop(n).head The apply method has a special meaning in Scala. An object with an apply method can be applied to arguments as if it was a function. For instance, to pick the 3’rd element of a list xs, one can write either xs.apply(3) or xs(3) – the latter expression expands into the ﬁrst. With take and drop, we can extract sublists consisting of consecutive elements of the original list. To extract the sublist xsm,..., xsn−1 of a list xs, use: xs.drop(m).take(n - m) Zipping lists. The next function combines two lists into a list of pairs. Given two lists xs = List(x1, ..., xn), and ys = List(y1, ..., yn), xs zip ys constructs the list List((x1, y1), ..., (xn, yn)). If the two lists have different lengths, the longer one of the two is truncated. Here is the deﬁnition of zip – note that it is a polymorphic method. def zip[B](that: List[B]): List[(a,b)] = if (this.isEmpty || that.isEmpty) Nil else (this.head, that.head) :: (this.tail zip that.tail) Consing lists.. Like any inﬁx operator, :: is also implemented as a method of an object. In this case, the object is the list that is extended. This is possible, because operators ending with a ‘:’ character are treated specially in Scala. All such opera- tors are treated as methods of their right operand. E.g., x :: y = y.::(x) whereas x + y = x.+(y) Note, however, that operands of a binary operation are in each case evaluated from left to right. So, if D and E are expressions with possible side-effects, D :: E is translated to {val x = D; E.::(x)} in order to maintain the left-to-right order of operand evaluation. Another difference between operators ending in a ‘:’ and other operators concerns their associativity. Operators ending in ‘:’ are right-associative, whereas other op- erators are left-associative. E.g., x :: y :: z = x :: (y :: z) whereas x+y+z=(x+y)+z The deﬁnition of :: as a method in class List is as follows: 68 Lists def ::[B >: A](x: B): List[B] = new scala.::(x, this) Note that :: is deﬁned for all elements x of type B and lists of type List[A] such that the type B of x is a supertype of the list’s element type A. The result is in this case a list of B’s. This is expressed by the type parameter B with lower bound A in the signature of ::. Concatenating lists. An operation similar to :: is list concatenation, written ‘:::’. The result of (xs ::: ys) is a list consisting of all elements of xs, followed by all elements of ys. Because it ends in a colon, ::: is right-associative and is considered as a method of its right-hand operand. Therefore, xs ::: ys ::: zs = xs ::: (ys ::: zs) = zs.:::(ys).:::(xs) Here is the implementation of the ::: method: def :::[B >: A](prefix: List[B]): List[B] = prefix match { case Nil => this case p :: ps => this.:::(ps).::(p) } Reversing lists. Another useful operation is list reversal. There is a method reverse in List to that effect. Let’s try to give its implementation: def reverse[A](xs: List[A]): List[A] = xs match { case Nil => Nil case x :: xs => reverse(xs) ::: List(x) } This implementation has the advantage of being simple, but it is not very efﬁcient. Indeed, one concatenation is executed for every element in the list. List concatena- tion takes time proportional to the length of its ﬁrst operand. Therefore, the com- plexity of reverse(xs) is n +(n −1)+...+1 = n(n +1)/2 where n is the length of xs. Can reverse be implemented more efﬁciently? We will see later that there exists another implementation which has only linear complexity. 9.3 Example: Merge sort The insertion sort presented earlier in this chapter is simple to formulate, but also not very efﬁcient. It’s average complexity is proportional to the square of the length 9.3 Example: Merge sort 69 of the input list. We now design a program to sort the elements of a list which is more efﬁcient than insertion sort. A good algorithm for this is merge sort, which works as follows. First, if the list has zero or one elements, it is already sorted, so one returns the list unchanged. Longer lists are split into two sub-lists, each containing about half the elements of the original list. Each sub-list is sorted by a recursive call to the sort function, and the resulting two sorted lists are then combined in a merge operation. For a general implementation of merge sort, we still have to specify the type of list elements to be sorted, as well as the function to be used for the comparison of el- ements. We obtain a function of maximal generality by passing these two items as parameters. This leads to the following implementation. def msort[A](less: (A, A) => Boolean)(xs: List[A]): List[A] = { def merge(xs1: List[A], xs2: List[A]): List[A] = if (xs1.isEmpty) xs2 else if (xs2.isEmpty) xs1 else if (less(xs1.head, xs2.head)) xs1.head :: merge(xs1.tail, xs2) else xs2.head :: merge(xs1, xs2.tail) val n = xs.length/2 if (n == 0) xs else merge(msort(less)(xs take n), msort(less)(xs drop n)) } The complexity of msort is O(N log(N)), where N is the length of the input list. To see why, note that splitting a list in two and merging two sorted lists each take time proportional to the length of the argument list(s). Each recursive call of msort halves the number of elements in its input, so there are O(log(N)) consecutive recursive calls until the base case of lists of length 1 is reached. However, for longer lists each call spawns off two further calls. Adding everything up we obtain that at each of the O(log(N)) call levels, every element of the original lists takes part in one split operation and in one merge operation. Hence, every call level has a total cost pro- portional to O(N). Since there are O(log(N)) call levels, we obtain an overall cost of O(N log(N)). That cost does not depend on the initial distribution of elements in the list, so the worst case cost is the same as the average case cost. This makes merge sort an attractive algorithm for sorting lists. Here is an example how msort is used. msort((x: Int, y: Int) => x < y)(List(5, 7, 1, 3)) The deﬁnition of msort is curried, to make it easy to specialize it with particular comparison functions. For instance, val intSort = msort((x: Int, y: Int) => x < y) val reverseSort = msort((x: Int, y: Int) => x > y) 70 Lists 9.4 Deﬁnition of class List II: Higher-Order Methods The examples encountered so far show that functions over lists often have similar structures. We can identify several patterns of computation over lists, like: • transforming every element of a list in some way. • extracting from a list all elements satisfying a criterion. • combine the elements of a list using some operator. Functional programming languages enable programmers to write general functions which implement patterns like this by means of higher order functions. We now discuss a set of commonly used higher-order functions, which are implemented as methods in class List. Mapping over lists. A common operation is to transform each element of a list and then return the lists of results. For instance, to scale each element of a list by a given factor. def scaleList(xs: List[Double], factor: Double): List[Double] = xs match { case Nil => xs case x :: xs1 => x * factor :: scaleList(xs1, factor) } This pattern can be generalized to the map method of class List: 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) } Using map, scaleList can be more concisely written as follows. def scaleList(xs: List[Double], factor: Double) = xs map (x => x * factor) As another example, consider the problem of returning a given column of a matrix which is represented as a list of rows, where each row is again a list. This is done by the following function column. def column[A](xs: List[List[A]], index: Int): List[A] = xs map (row => row(index)) Closely related to map is the foreach method, which applies a given function to all elements of a list, but does not construct a list of results. The function is thus applied only for its side effect. foreach is deﬁned as follows. 9.4 Deﬁnition of class List II: Higher-Order Methods 71 def foreach(f: A => Unit) { this match { case Nil => () case x :: xs => f(x); xs.foreach(f) } } This function can be used for printing all elements of a list, for instance: xs foreach (x => println(x)) Exercise 9.4.1 Consider a function which squares all elements of a list and re- turns a list with the results. Complete the following two equivalent deﬁnitions of squareList. def squareList(xs: List[Int]): List[Int] = xs match { case List() => ?? case y :: ys => ?? } def squareList(xs: List[Int]): List[Int] = xs map ?? Filtering Lists. Another common operation selects from a list all elements fulﬁll- ing a given criterion. For instance, to return a list of all positive elements in some given lists of integers: def posElems(xs: List[Int]): List[Int] = xs match { case Nil => xs case x :: xs1 => if (x > 0) x :: posElems(xs1) else posElems(xs1) } This pattern is generalized to the filter method of class List: 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) } Using filter, posElems can be more concisely written as follows. def posElems(xs: List[Int]): List[Int] = xs filter (x => x > 0) An operation related to ﬁltering is testing whether all elements of a list satisfy a cer- tain condition. Dually, one might also be interested in the question whether there 72 Lists exists an element in a list that satisﬁes a certain condition. These operations are embodied in the higher-order functions forall and exists of class List. def forall(p: A => Boolean): Boolean = isEmpty || (p(head) && (tail forall p)) def exists(p: A => Boolean): Boolean = !isEmpty && (p(head) || (tail exists p)) To illustrate the use of forall, consider the question of whether a number if prime. Remember that a number n is prime of it can be divided without remainder only by one and itself. The most direct translation of this deﬁnition would test that n divided by all numbers from 2 up to and excluding itself gives a non-zero remainder. This list of numbers can be generated using a function List.range which is deﬁned in object List as follows. package scala object List { ... def range(from: Int, end: Int): List[Int] = if (from >= end) Nil else from :: range(from + 1, end) For example, List.range(2, n) generates the list of all integers from 2 up to and excluding n. The function isPrime can now simply be deﬁned as follows. def isPrime(n: Int) = List.range(2, n) forall (x => n % x != 0) We see that the mathematical deﬁnition of prime-ness has been translated directly into Scala code. Exercise: Deﬁne forall and exists in terms of filter. Folding and Reducing Lists. Another common operation is to combine the ele- ments of a list with some operator. For instance: sum(List(x1, ..., xn)) = 0 + x1 + ... + xn product(List(x1, ..., xn)) = 1 * x1 *...* xn Of course, we can implement both functions with a recursive scheme: def sum(xs: List[Int]): Int = xs match { case Nil => 0 case y :: ys => y + sum(ys) } def product(xs: List[Int]): Int = xs match { case Nil => 1 case y :: ys => y * product(ys) } 9.4 Deﬁnition of class List II: Higher-Order Methods 73 But we can also use the generalization of this program scheme embodied in the reduceLeft method of class List. This method inserts a given binary operator be- tween adjacent elements of a given list. E.g. List(x1, ..., xn).reduceLeft(op) = (...(x1 op x2) op ... ) op xn Using reduceLeft, we can make the common pattern in sum and product apparent: def sum(xs: List[Int]) = (0 :: xs) reduceLeft {(x, y) => x + y} def product(xs: List[Int]) = (1 :: xs) reduceLeft {(x, y) => x * y} Here is the implementation of reduceLeft. def reduceLeft(op: (A, A) => A): A = this match { case Nil => error("Nil.reduceLeft") case x :: xs => (xs foldLeft x)(op) } def foldLeft[B](z: B)(op: (B, A) => B): B = this match { case Nil => z case x :: xs => (xs foldLeft op(z, x))(op) } } We see that the reduceLeft method is deﬁned in terms of another generally use- ful method, foldLeft. The latter takes as additional parameter an accumulator z, which is returned when foldLeft is applied on an empty list. That is, (List(x1, ..., xn) foldLeft z)(op) = (...(z op x1) op ... ) op xn The sum and product methods can be deﬁned alternatively using foldLeft: def sum(xs: List[Int]) = (xs foldLeft 0) {(x, y) => x + y} def product(xs: List[Int]) = (xs foldLeft 1) {(x, y) => x * y} FoldRight and ReduceRight. Applications of foldLeft and reduceLeft expand to left-leaning trees. . They have duals foldRight and reduceRight, which produce right-leaning trees. List(x1, ..., xn).reduceRight(op) = x1 op ( ... (xn−1 op xn)...) (List(x1, ..., xn) foldRight acc)(op) = x1 op ( ... (xn op acc)...) These are deﬁned as follows. def reduceRight(op: (A, A) => A): A = this match { case Nil => error("Nil.reduceRight") case x :: Nil => x case x :: xs => op(x, xs.reduceRight(op)) 74 Lists } def foldRight[B](z: B)(op: (A, B) => B): B = this match { case Nil => z case x :: xs => op(x, (xs foldRight z)(op)) } Class List deﬁnes also two symbolic abbreviations for foldLeft and foldRight: def /:[B](z: B)(f: (B, A) => B): B = foldLeft(z)(f) def :\[B](z: B)(f: (A, B) => B): B = foldRight(z)(f) The method names picture the left/right leaning trees of the fold operations by for- ward or backward slashes. The : points in each case to the list argument whereas the end of the slash points to the accumulator (or: zero) argument z. That is, (z /: List(x1, ..., xn))(op) = (...(z op x1) op ... ) op xn (List(x1, ..., xn) :\ z)(op) = x1 op ( ... (xn op acc)...) For associative and commutative operators, /: and :\ are equivalent (even though there may be a difference in efﬁciency). Exercise 9.4.2 Consider the problem of writing a function flatten, which takes a list of element lists as arguments. The result of flatten should be the concatenation of all element lists into a single list. Here is the an implementation of this method in terms of :\. def flatten[A](xs: List[List[A]]): List[A] = (xs :\ (Nil: List[A])) {(x, xs) => x ::: xs} Consider replacing the body of flatten by ((Nil: List[A]) /: xs) ((xs, x) => xs ::: x) What would be the difference in asymptotic complexity between the two versions of flatten? In fact flatten is predeﬁned together with a set of other userful function in an ob- ject called List in the standatd Scala library. It can be accessed from user program by calling List.flatten. Note that flatten is not a method of class List – it would not make sense there, since it applies only to lists of lists, not to all lists in general. List Reversal Again. We have seen in Section 9.2 an implementation of method reverse whose run-time was quadratic in the length of the list to be reversed. We now develop a new implementation of reverse, which has linear cost. The idea is to use a foldLeft operation based on the following program scheme. class List[+A] { ... 9.4 Deﬁnition of class List II: Higher-Order Methods 75 def reverse: List[A] = (z? /: this)(op?) It only remains to ﬁll in the z? and op? parts. Let’stry to deduce them from examples. Nil = Nil.reverse // by specification = (z /: Nil)(op) // by the template for reverse = (Nil foldLeft z)(op) // by the definition of /: = z // by definition of foldLeft Hence, z? must be Nil. To deduce the second operand, let’s study reversal of a list of length one. List(x) = List(x).reverse // by specification = (Nil /: List(x))(op) // by the template for reverse, with z = Nil = (List(x) foldLeft Nil)(op) // by the definition of /: = op(Nil, x) // by definition of foldLeft Hence, op(Nil, x) equals List(x), which is the same as x :: Nil. This suggests to take as op the :: operator with its operands exchanged. Hence, we arrive at the following implementation for reverse, which has linear complexity. def reverse: List[A] = ((Nil: List[A]) /: this) {(xs, x) => x :: xs} (Remark: The type annotation of Nil is necessary to make the type inferencer work.) Exercise 9.4.3 Fill in the missing expressions to complete the following deﬁnitions of some basic list-manipulation operations as fold operations. def mapFun[A, B](xs: List[A], f: A => B): List[B] = (xs :\ List[B]()){ ?? } def lengthFun[A](xs: List[A]): int = (0 /: xs){ ?? } Nested Mappings. We can employ higher-order list processing functions to ex- press many computations that are normally expressed as nested loops in imperative languages. As an example, consider the following problem: Given a positive integer n, ﬁnd all pairs of positive integers i and j, where 1 ≤ j < i < n such that i + j is prime. For 76 Lists instance, if n = 7, the pairs are i 2 3 4 4 5 6 6 j 1 2 1 3 2 1 5 i + j 3 5 5 7 7 7 11 A natural way to solve this problem consists of two steps. In a ﬁrst step, one gener- ates the sequence of all pairs (i, j) of integers such that 1 ≤ j < i < n. In a second step one then ﬁlters from this sequence all pairs (i, j) such that i + j is prime. Looking at the ﬁrst step in more detail, a natural way to generate the sequence of pairs consists of three sub-steps. First, generate all integers between 1 and n for i. Second, for each integer i between 1 and n, generate the list of pairs (i,1) up to (i,i −1). This can be achieved by a combination of range and map: List.range(1, i) map (x => (i, x)) Finally, combine all sublists using foldRight with :::. Putting everything together gives the following expression: List.range(1, n) .map(i => List.range(1, i).map(x => (i, x))) .foldRight(List[(Int, Int)]()) {(xs, ys) => xs ::: ys} .filter(pair => isPrime(pair._1 + pair._2)) Flattening Maps. The combination of mapping and then concatenating sublists resulting from the map is so common that we there is a special method for it in class List: abstract class List[+A] { ... def flatMap[B](f: A => List[B]): List[B] = this match { case Nil => Nil case x :: xs => f(x) ::: (xs flatMap f) } } With flatMap, the pairs-whose-sum-is-prime expression could have been written more concisely as follows. List.range(1, n) .flatMap(i => List.range(1, i).map(x => (i, x))) .filter(pair => isPrime(pair._1 + pair._2)) 9.5 Summary 77 9.5 Summary This chapter has introduced lists as a fundamental data structure in programming. Since lists are immutable, they are a common data type in functional programming languages. They have a role comparable to arrays in imperative languages. How- ever, the access patterns between arrays and lists are quite different. Where array accessing is always done by indexing, this is much less common for lists. We have seen that scala.List deﬁnes a method called apply for indexing however this oper- ation is much more costly than in the case of arrays (linear as opposed to constant time). Instead of indexing, lists are usually traversed recursively, where recursion steps are usually based on a pattern match over the traversed list. There is also a rich set of higher-order combinators which allow one to instantiate a set of prede- ﬁned patterns of computations over lists. Chapter 10 For-Comprehensions The last chapter demonstrated that higher-order functions such as map, flatMap, filter provide powerful constructions for dealing with lists. But sometimes the level of abstraction required by these functions makes a program hard to under- stand. To help understandability, Scala has a special notation which simpliﬁes common patterns of applications of higher-order functions. This notation builds a bridge between set-comprehensions in mathematics and for-loops in imperative lan- guages such as C or Java. It also closely resembles the query notation of relational databases. As a ﬁrst example, say we are given a list persons of persons with name and age ﬁelds. To print the names of all persons in the sequence which are aged over 20, one can write: for (p <- persons if p.age > 20) yield p.name This is equivalent to the following expression , which uses higher-order functions filter and map: persons filter (p => p.age > 20) map (p => p.name) The for-comprehension looks a bit like a for-loop in imperative languages, except that it constructs a list of the results of all iterations. Generally, a for-comprehension is of the form for ( s ) yield e Here, s is a sequence of generators, deﬁnitions and ﬁlters.A generator is of the form val x <- e, where e is a list-valued expression. It binds x to successive values in the list. A deﬁnition is of the form val x = e. It introduces x as a name for the value of e in the rest of the comprehension. A ﬁlter is an expression f of type Boolean. It omits 80 For-Comprehensions from consideration all bindings for which f is false. The sequence s starts in each case with a generator. If there are several generators in a sequence, later generators vary more rapidly than earlier ones. The sequence s may also be enclosed in braces instead of parentheses, in which case the semicolons between generators, deﬁnitions and ﬁlters can be omitted. Here are two examples that show how for-comprehensions are used. First, let’s redo an example of the previous chapter: Given a positive integer n, ﬁnd all pairs of positive integers i and j, where 1 ≤ j < i < n such that i + j is prime. With a for- comprehension this problem is solved as follows: for { i <- List.range(1, n) j <- List.range(1, i) if isPrime(i+j) } yield {i, j} This is arguably much clearer than the solution using map, flatMap and filter that we have developed previously. As a second example, consider computing the scalar product of two vectors xs and ys. Using a for-comprehension, this can be written as follows. sum(for ((x, y) <- xs zip ys) yield x * y) 10.1 The N-Queens Problem For-comprehensions are especially useful for solving combinatorial puzzles. An ex- ample of such a puzzle is the 8-queens problem: Given a standard chess-board, place 8 queens such that no queen is in check from any other (a queen can check another piece if they are on the same column, row, or diagonal). We will now de- velop a solution to this problem, generalizing it to chess-boards of arbitrary size. Hence, the problem is to place n queens on a chess-board of size n ×n. To solve this problem, note that we need to place a queen in each row. So we could place queens in successive rows, each time checking that a newly placed queen is not in check from any other queens that have already been placed. In the course of this search, it might arrive that a queen to be placed in row k would be in check in all ﬁelds of that row from queens in row 1 to k − 1. In that case, we need to abort that part of the search in order to continue with a different conﬁguration of queens in columns 1 to k −1. This suggests a recursive algorithm. Assume that we have already generated all so- lutions of placing k −1 queens on a board of size n ×n. We can represent each such solution by a list of length k − 1 of column numbers (which can range from 1 to n). We treat these partial solution lists as stacks, where the column number of the queen in row k − 1 comes ﬁrst in the list, followed by the column number of the queen in row k −2, etc. The bottom of the stack is the column number of the queen 10.2 Querying with For-Comprehensions 81 placed in the ﬁrst row of the board. All solutions together are then represented as a list of lists, with one element for each solution. Now, to place the k’the queen, we generate all possible extensions of each previous solution by one more queen. This yields another list of solution lists, this time of length k. We continue the process until we have reached solutions of the size of the chess-board n. This algorithmic idea is embodied in function placeQueens below: def queens(n: Int): List[List[Int]] = { def placeQueens(k: Int): List[List[Int]] = if (k == 0) List(List()) else for { queens <- placeQueens(k - 1) column <- List.range(1, n + 1) if isSafe(column, queens, 1) } yield column :: queens placeQueens(n) } Exercise 10.1.1 Write the function def isSafe(col: Int, queens: List[Int], delta: Int): Boolean which tests whether a queen in the given column col is safe with respect to the queens already placed. Here, delta is the difference between the row of the queen to be placed and the row of the ﬁrst queen in the list. 10.2 Querying with For-Comprehensions The for-notation is essentially equivalent to common operations of database query languages. For instance, say we are given a database books, represented as a list of books, where Book is deﬁned as follows. case class Book(title: String, authors: List[String]) Here is a small example database: val books: List[Book] = List( Book("Structure and Interpretation of Computer Programs", List("Abelson, Harold", "Sussman, Gerald J.")), Book("Principles of Compiler Design", List("Aho, Alfred", "Ullman, Jeffrey")), Book("Programming in Modula-2", List("Wirth, Niklaus")), Book("Introduction to Functional Programming"), List("Bird, Richard")), Book("The Java Language Specification", List("Gosling, James", "Joy, Bill", "Steele, Guy", "Bracha, Gilad"))) 82 For-Comprehensions Then, to ﬁnd the titles of all books whose author’s last name is “Ullman”: for (b <- books; a <- b.authors if a startsWith "Ullman") yield b.title (Here, startsWith is a method in java.lang.String). Or, to ﬁnd the titles of all books that have the string “Program” in their title: for (b <- books if (b.title indexOf "Program") >= 0) yield b.title Or, to ﬁnd the names of all authors that have written at least two books in the database. for (b1 <- books; b2 <- books if b1 != b2; a1 <- b1.authors; a2 <- b2.authors if a1 == a2) yield a1 The last solution is not yet perfect, because authors will appear several times in the list of results. We still need to remove duplicate authors from result lists. This can be achieved with the following function. def removeDuplicates[A](xs: List[A]): List[A] = if (xs.isEmpty) xs else xs.head :: removeDuplicates(xs.tail filter (x => x != xs.head)) Note that the last expression in method removeDuplicates can be equivalently ex- pressed using a for-comprehension. xs.head :: removeDuplicates(for (x <- xs.tail if x != xs.head) yield x) 10.3 Translation of For-Comprehensions Every for-comprehension can be expressed in terms of the three higher-order func- tions map, flatMap and filter. Here is the translation scheme, which is also used by the Scala compiler. • A simple for-comprehension for (x <- e) yield e’ is translated to e.map(x => e’) • A for-comprehension 10.3 Translation of For-Comprehensions 83 for (x <- e if f; s) yield e’ where f is a ﬁlter and s is a (possibly empty) sequence of generators or ﬁlters is translated to for (x <- e.filter(x => f); s) yield e’ and then translation continues with the latter expression. • A for-comprehension for (x <- e; y <- e’; s) yield e’’ where s is a (possibly empty) sequence of generators or ﬁlters is translated to e.flatMap(x => for (y <- e’; s) yield e’’) and then translation continues with the latter expression. For instance, taking our "pairs of integers whose sum is prime" example: for { i <- range(1, n) j <- range(1, i) if isPrime(i+j) } yield {i, j} Here is what we get when we translate this expression: range(1, n) .flatMap(i => range(1, i) .filter(j => isPrime(i+j)) .map(j => (i, j))) Conversely, it would also be possible to express functions map, flatMap and filter using for-comprehensions. Here are the three functions again, this time imple- mented using for-comprehensions. object Demo { def map[A, B](xs: List[A], f: A => B): List[B] = for (x <- xs) yield f(x) def flatMap[A, B](xs: List[A], f: A => List[B]): List[B] = for (x <- xs; y <- f(x)) yield y def filter[A](xs: List[A], p: A => Boolean): List[A] = for (x <- xs if p(x)) yield x } 84 For-Comprehensions Not surprisingly, the translation of the for-comprehension in the body of Demo.map will produce a call to map in class List. Similarly, Demo.flatMap and Demo.filter translate to flatMap and filter in class List. Exercise 10.3.1 Deﬁne the following function in terms of for. def flatten[A](xss: List[List[A]]): List[A] = (xss :\ (Nil: List[A])) ((xs, ys) => xs ::: ys) Exercise 10.3.2 Translate for (b <- books; a <- b.authors if a startsWith "Bird") yield b.title for (b <- books if (b.title indexOf "Program") >= 0) yield b.title to higher-order functions. 10.4 For-Loops For-comprehensions resemble for-loops in imperative languages, except that they produce a list of results. Sometimes, a list of results is not needed but we would still like the ﬂexibility of generators and ﬁlters in iterations over lists. This is made possible by a variant of the for-comprehension syntax, which expresses for-loops: for ( s ) e This construct is the same as the standard for-comprehension syntax except that the keyword yield is missing. The for-loop is executed by executing the expression e for each element generated from the sequence of generators and ﬁlters s. As an example, the following expression prints out all elements of a matrix repre- sented as a list of lists: for (xs <- xss) { for (x <- xs) print(x + "\t") println() } The translation of for-loops to higher-order methods of class List is similar to the translation of for-comprehensions, but is simpler. Where for-comprehensions translate to map and flatMap, for-loops translate in each case to foreach. 10.5 Generalizing For We have seen that the translation of for-comprehensions only relies on the presence of methods map, flatMap, and filter. Therefore it is possible to apply the same 10.5 Generalizing For 85 notation to generators that produce objects other than lists; these objects only have to support the three key functions map, flatMap, and filter. The standard Scala library has several other abstractions that support these three methods and with them support for-comprehensions. We will encounter some of them in the following chapters. As a programmer you can also use this principle to enable for-comprehensions for types you deﬁne – these types just need to support methods map, flatMap, and filter. There are many examples where this is useful: Examples are database interfaces, XML trees, or optional values. One caveat: It is not assured automatically that the result translating a for- comprehension is well-typed. To ensure this, the types of map, flatMap and filter have to be essentially similar to the types of these methods in class List. To make this precise, assume you have a parameterized class C[A] for which you want to enable for-comprehensions. Then C should deﬁne map, flatMap and filter with the following types: def map[B](f: A => B): C[B] def flatMap[B](f: A => C[B]): C[B] def filter(p: A => Boolean): C[A] It would be attractive to enforce these types statically in the Scala compiler, for instance by requiring that any type supporting for-comprehensions implements a standard trait with these methods 1. The problem is that such a standard trait would have to abstract over the identity of the class C, for instance by taking C as a type pa- rameter. Note that this parameter would be a type constructor, which gets applied to several different types in the signatures of methods map and flatMap. Unfortu- nately, the Scala type system is too weak to express this construct, since it can han- dle only type parameters which are fully applied types. 1In the programming language Haskell, which has similar constructs, this abstraction is called a “monad with zero” Chapter 11 Mutable State Most programs we have presented so far did not have side-effects 1. Therefore, the notion of time did not matter. For a program that terminates, any sequence of ac- tions would have led to the same result! This is also reﬂected by the substitution model of computation, where a rewrite step can be applied anywhere in a term, and all rewritings that terminate lead to the same solution. In fact, this conﬂuence property is a deep result in λ-calculus, the theory underlying functional program- ming. In this chapter, we introduce functions with side effects and study their behavior. We will see that as a consequence we have to fundamentally modify up the substi- tution model of computation which we employed so far. 11.1 Stateful Objects We normally view the world as a set of objects, some of which have state that changes over time. Normally, state is associated with a set of variables that can be changed in the course of a computation. There is also a more abstract notion of state, which does not refer to particular constructs of a programming language: An object has state (or: is stateful) if its behavior is inﬂuenced by its history. For instance, a bank account object has state, because the question “can I withdraw 100 CHF?” might have different answers during the lifetime of the account. In Scala, all mutable state is ultimately built from variables. A variable deﬁnition is written like a value deﬁnition, but starts with var instead of val. For instance, the following two deﬁnitions introduce and initialize two variables x and count. var x: String = "abc" 1We ignore here the fact that some of our program printed to standard output, which technically is a side effect. 88 Mutable State var count = 111 Like a value deﬁnition, a variable deﬁnition associates a name with a value. But in the case of a variable deﬁnition, this association may be changed later by an assign- ment. Such assignments are written as in C or Java. Examples: x = "hello" count = count + 1 In Scala, every deﬁned variable has to be initialized at the point of its deﬁnition. For instance, the statement var x: Int; is not regarded as a variable deﬁnition, because the initializer is missing2. If one does not know, or does not care about, the appropriate initializer, one can use a wildcard instead. I.e. val x: T = _ will initialize x to some default value (null for reference types, false for booleans, and the appropriate version of 0 for numeric value types). Real-world objects with state are represented in Scala by objects that have variables as members. For instance, here is a class that represents bank accounts. class BankAccount { private var balance = 0 def deposit(amount: Int) { if (amount > 0) balance += amount } def withdraw(amount: Int): Int = if (0 < amount && amount <= balance) { balance -= amount balance } else error("insufficient funds") } The class deﬁnes a variable balance which contains the current balance of an ac- count. Methods deposit and withdraw change the value of this variable through assignments. Note that balance is private in class BankAccount – hence it can not be accessed directly outside the class. To create bank-accounts, we use the usual object creation notation: val myAccount = new BankAccount 2If a statement like this appears in a class, it is instead regarded as a variable declaration, which introduces abstract access methods for the variable, but does not associate these methods with a piece of state. 11.1 Stateful Objects 89 Example 11.1.1 Here is a scalaint session that deals with bank accounts. scala> :l bankaccount.scala Loading bankaccount.scala... defined class BankAccount scala> val account = new BankAccount account: BankAccount = BankAccount$class@1797795 scala> account deposit 50 unnamed0: Unit = () scala> account withdraw 20 unnamed1: Int = 30 scala> account withdraw 20 unnamed2: Int = 10 scala> account withdraw 15 java.lang.Error: insufficient funds at scala.Predef$error(Predef.scala:74) at BankAccount$class.withdraw(:14) at (:5) scala> The example shows that applying the same operation (withdraw 20) twice to an account yields different results. So, clearly, accounts are stateful objects. Sameness and Change. Assignments pose new problems in deciding when two expressions are “the same”. If assignments are excluded, and one writes val x = E; val y = E where E is some arbitrary expression, then x and y can reasonably be assumed to be the same. I.e. one could have equivalently written val x = E; val y = x (This property is usually called referential transparency). But once we admit assign- ments, the two deﬁnition sequences are different. Consider: val x = new BankAccount; val y = new BankAccount To answer the question whether x and y are the same, we need to be more pre- cise what “sameness” means. This meaning is captured in the notion of operational equivalence, which, somewhat informally, is stated as follows. Suppose we have two deﬁnitions of x and y. To test whether x and y deﬁne the same value, proceed as follows. • Execute the deﬁnitions followed by an arbitrary sequence S of operations that involve x and y. Observe the results (if any). 90 Mutable State • Then, execute the deﬁnitions with another sequence S’ which results from S by renaming all occurrences of y in S to x. • If the results of running S’ are different, then surely x and y are different. • On the other hand, if all possible pairs of sequences {S, S’} yield the same results, then x and y are the same. In other words, operational equivalence regards two deﬁnitions x and y as deﬁning the same value, if no possible experiment can distinguish between x and y. An ex- periment in this context are two version of an arbitrary program which use either x or y. Given this deﬁnition, let’s test whether val x = new BankAccount; val y = new BankAccount deﬁnes values x and y which are the same. Here are the deﬁnitions again, followed by a test sequence: > val x = new BankAccount > val y = new BankAccount > x deposit 30 30 > y withdraw 20 java.lang.RuntimeException: insufficient funds Now, rename all occurrences of y in that sequence to x. We get: > val x = new BankAccount > val y = new BankAccount > x deposit 30 30 > x withdraw 20 10 Since the ﬁnal results are different, we have established that x and y are not the same. On the other hand, if we deﬁne val x = new BankAccount; val y = x then no sequence of operations can distinguish between x and y, so x and y are the same in this case. Assignment and the Substitution Model. These examples show that our previous substitution model of computation cannot be used anymore. After all, under this model we could always replace a value name by its deﬁning expression. For instance in 11.2 Imperative Control Structures 91 val x = new BankAccount; val y = x the x in the deﬁnition of y could be replaced by new BankAccount. But we have seen that this change leads to a different program. So the substitution model must be invalid, once we add assignments. 11.2 Imperative Control Structures Scala has the while and do-while loop constructs known from the C and Java lan- guages. There is also a single branch if which leaves out the else-part as well as a return statement which aborts a function prematurely. This makes it possible to program in a conventional imperative style. For instance, the following function, which computes the n’th power of a given parameter x, is implemented using while and single-branch if. def power(x: Double, n: Int): Double = { var r = 1.0 var i = n var j = 0 while (j < 32) { r = r * r if (i < 0) r *= x i = i << 1 j += 1 } r } These imperative control constructs are in the language for convenience. They could have been left out, as the same constructs can be implemented using just functions. As an example, let’s develop a functional implementation of the while loop. whileLoop should be a function that takes two parameters: a condition, of type Boolean, and a command, of type Unit. Both condition and command need to be passed by-name, so that they are evaluated repeatedly for each loop iteration. This leads to the following deﬁnition of whileLoop. def whileLoop(condition: => Boolean)(command: => Unit) { if (condition) { command; whileLoop(condition)(command) } else () } Note that whileLoop is tail recursive, so it operates in constant stack space. 92 Mutable State Exercise 11.2.1 Write a function repeatLoop, which should be applied as follows: repeatLoop { command } ( condition ) Is there also a way to obtain a loop syntax like the following? repeatLoop { command } until ( condition ) Some other control constructs known from C and Java are missing in Scala: There are no break and continue jumps for loops. There are also no for-loops in the Java sense – these have been replaced by the more general for-loop construct discussed in Section 10.4. 11.3 Extended Example: Discrete Event Simulation We now discuss an example that demonstrates how assignments and higher-order functions can be combined in interesting ways. We will build a simulator for digital circuits. The example is taken from Abelson and Sussman’s book [ASS96]. We augment their basic (Scheme-) code by an object-oriented structure which allows code-reuse through inheritance. The example also shows how discrete event simulation pro- grams in general are structured and built. We start with a little language to describe digital circuits. A digital circuit is built from wires and function boxes. Wires carry signals which are transformed by func- tion boxes. We will represent signals by the booleans true and false. Basic function boxes (or: gates) are: • An inverter, which negates its signal • An and-gate, which sets its output to the conjunction of its input. • An or-gate, which sets its output to the disjunction of its input. Other function boxes can be built by combining basic ones. Gates have delays, so an output of a gate will change only some time after its inputs change. A Language for Digital Circuits. We describe the elements of a digital circuit by the following set of Scala classes and functions. First, there is a class Wire for wires. We can construct wires as follows. val a = new Wire val b = new Wire val c = new Wire 11.3 Extended Example: Discrete Event Simulation 93 Second, there are procedures def inverter(input: Wire, output: Wire) def andGate(a1: Wire, a2: Wire, output: Wire) def orGate(o1: Wire, o2: Wire, output: Wire) which “make” the basic gates we need (as side-effects). More complicated function boxes can now be built from these. For instance, to construct a half-adder, we can deﬁne: def halfAdder(a: Wire, b: Wire, s: Wire, c: Wire) { val d = new Wire val e = new Wire orGate(a, b, d) andGate(a, b, c) inverter(c, e) andGate(d, e, s) } This abstraction can itself be used, for instance in deﬁning a full adder: def fullAdder(a: Wire, b: Wire, cin: Wire, sum: Wire, cout: Wire) { val s = new Wire val c1 = new Wire val c2 = new Wire halfAdder(a, cin, s, c1) halfAdder(b, s, sum, c2) orGate(c1, c2, cout) } Class Wire and functions inverter, andGate, and orGate represent thus a little lan- guage in which users can deﬁne digital circuits. We now give implementations of this class and these functions, which allow one to simulate circuits. These imple- mentations are based on a simple and general API for discrete event simulation. The Simulation API. Discrete event simulation performs user-deﬁned actions at speciﬁed times. An action is represented as a function which takes no parameters and returns a Unit result: type Action = () => Unit The time is simulated; it is not the actual “wall-clock” time. A concrete simulation will be done inside an object which inherits from the abstract Simulation class. This class has the following signature: abstract class Simulation { 94 Mutable State def currentTime: Int def afterDelay(delay: Int, action: => Action) def run() } Here, currentTime returns the current simulated time as an integer number, afterDelay schedules an action to be performed at a speciﬁed delay after currentTime, and run runs the simulation until there are no further actions to be performed. The Wire Class. A wire needs to support three basic actions. getSignal: Boolean returns the current signal on the wire. setSignal(sig: Boolean) sets the wire’s signal to sig. addAction(p: Action) attaches the speciﬁed procedure p to the actions of the wire. All attached action procedures will be executed every time the signal of a wire changes. Here is an implementation of the Wire class: class Wire { private var sigVal = false private var actions: List[Action] = List() def getSignal = sigVal def setSignal(s: Boolean) = if (s != sigVal) { sigVal = s actions.foreach(action => action()) } def addAction(a: Action) { actions = a :: actions; a() } } Two private variables make up the state of a wire. The variable sigVal represents the current signal, and the variable actions represents the action procedures currently attached to the wire. The Inverter Class. We implement an inverter by installing an action on its input wire, namely the action which puts the negated input signal onto the output sig- nal. The action needs to take effect at InverterDelay simulated time units after the input changes. This suggests the following implementation: def inverter(input: Wire, output: Wire) { 11.3 Extended Example: Discrete Event Simulation 95 def invertAction() { val inputSig = input.getSignal afterDelay(InverterDelay) { output setSignal !inputSig } } input addAction invertAction } The And-Gate Class. And-gates are implemented analogously to inverters. The action of an andGate is to output the conjunction of its input signals. This should happen at AndGateDelay simulated time units after any one of its two inputs changes. Hence, the following implementation: def andGate(a1: Wire, a2: Wire, output: Wire) { def andAction() { val a1Sig = a1.getSignal val a2Sig = a2.getSignal afterDelay(AndGateDelay) { output setSignal (a1Sig & a2Sig) } } a1 addAction andAction a2 addAction andAction } Exercise 11.3.1 Write the implementation of orGate. Exercise 11.3.2 Another way is to deﬁne an or-gate by a combination of inverters and and gates. Deﬁne a function orGate in terms of andGate and inverter. What is the delay time of this function? The Simulation Class. Now, we just need to implement class Simulation, and we are done. The idea is that we maintain inside a Simulation object an agenda of actions to perform. The agenda is represented as a list of pairs of actions and the times they need to be run. The agenda list is sorted, so that earlier actions come before later ones. abstract class Simulation { case class WorkItem(time: Int, action: Action) private type Agenda = List[WorkItem] private var agenda: Agenda = List() There is also a private variable curtime to keep track of the current simulated time. private var curtime = 0 96 Mutable State An application of the method afterDelay(delay, block) inserts the element WorkItem(currentTime + delay, () => block) into the agenda list at the appro- priate place. private def insert(ag: Agenda, item: WorkItem): Agenda = if (ag.isEmpty || item.time < ag.head.time) item :: ag else ag.head :: insert(ag.tail, item) def afterDelay(delay: Int)(block: => Unit) { val item = WorkItem(currentTime + delay, () => block) agenda = insert(agenda, item) } An application of the run method removes successive elements from the agenda and performs their actions. It continues until the agenda is empty: private def next() { agenda match { case WorkItem(time, action) :: rest => agenda = rest; curtime = time; action() case List() => } } def run() { afterDelay(0) { println("*** simulation started ***") } while (!agenda.isEmpty) next() } Running the Simulator. To run the simulator, we still need a way to inspect changes of signals on wires. To this purpose, we write a function probe. def probe(name: String, wire: Wire) { wire addAction { println(name + " " + currentTime + " new_value = " + wire.getSignal) } } Now, to see the simulator in action, let’s deﬁne four wires, and place probes on two of them: scala> val input1, input2, sum, carry = new Wire scala> probe("sum", sum) sum 0 new_value = false 11.4 Summary 97 scala> probe("carry", carry) carry 0 new_value = false Now let’s deﬁne a half-adder connecting the wires: scala> halfAdder(input1, input2, sum, carry) Finally, set one after another the signals on the two input wires to true and run the simulation. scala> input1 setSignal true; run *** simulation started *** sum 8 new_value = true scala> input2 setSignal true; run carry 11 new_value = true sum 15 new_value = false 11.4 Summary We have seen in this chapter the constructs that let us model state in Scala – these are variables, assignments, and imperative control structures. State and Assign- ment complicate our mental model of computation. In particular, referential trans- parency is lost. On the other hand, assignment gives us new ways to formulate pro- grams elegantly. As always, it depends on the situation whether purely functional programming or programming with assignments works best. Chapter 12 Computing with Streams The previous chapters have introduced variables, assignment and stateful objects. We have seen how real-world objects that change with time can be modeled by changing the state of variables in a computation. Time changes in the real world thus are modeled by time changes in program execution. Of course, such time changes are usually stretched out or compressed, but their relative order is the same. This seems quite natural, but there is a also price to pay: Our simple and pow- erful substitution model for functional computation is no longer applicable once we introduce variables and assignment. Is there another way? Can we model state change in the real world using only im- mutable functions? Taking mathematics as a guide, the answer is clearly yes: A time-changing quantity is simply modeled by a function f(t) with a time parame- ter t. The same can be done in computation. Instead of overwriting a variable with successive values, we represent all these values as successive elements in a list. So, a mutable variable var x: T gets replaced by an immutable value val x: List[T]. In a sense, we trade space for time – the different values of the variable now all exist concurrently as different elements of the list. One advantage of the list-based view is that we can “time-travel”, i.e. view several successive values of the variable at the same time. Another advantage is that we can make use of the powerful library of list processing functions, which often simpliﬁes computation. For instance, consider the imperative way to compute the sum of all prime numbers in an interval: def sumPrimes(start: Int, end: Int): Int = { var i = start var acc = 0 while (i < end) { if (isPrime(i)) acc += i i += 1 } acc } 100 Computing with Streams Note that the variable i “steps through” all values of the interval [start .. end-1]. A more functional way is to represent the list of values of variable i directly as range(start, end). Then the function can be rewritten as follows. def sumPrimes(start: Int, end: Int) = sum(range(start, end) filter isPrime) No contest which program is shorter and clearer! However, the functional program is also considerably less efﬁcient since it constructs a list of all numbers in the in- terval, and then another one for the prime numbers. Even worse from an efﬁciency point of view is the following example: To ﬁnd the second prime number between 1000 and 10000: range(1000, 10000) filter isPrime at 1 Here, the list of all numbers between 1000 and 10000 is constructed. But most of that list is never inspected! However, we can obtain efﬁcient execution for examples like these by a trick: Avoid computing the tail of a sequence unless that tail is actually neces- sary for the computation. We deﬁne a new class for such sequences, which is called Stream. Streams are created using the constant empty and the constructor cons, which are both deﬁned in module scala.Stream. For instance, the following expression con- structs a stream with elements 1 and 2: Stream.cons(1, Stream.cons(2, Stream.empty)) As another example, here is the analogue of List.range, but returning a stream instead of a list: def range(start: Int, end: Int): Stream[Int] = if (start >= end) Stream.empty else Stream.cons(start, range(start + 1, end)) (This function is also deﬁned as given above in module Stream). Even though Stream.range and List.range look similar, their execution behavior is completely different: Stream.range immediately returns with a Stream object whose ﬁrst element is start. All other elements are computed only when they are demanded by calling the tail method (which might be never at all). Streams are accessed just as lists. Similarly to lists, the basic access methods are isEmpty, head and tail. For instance, we can print all elements of a stream as fol- lows. 101 def print(xs: Stream[A]) { if (!xs.isEmpty) { Console.println(xs.head); print(xs.tail) } } Streams also support almost all other methods deﬁned on lists (see below for where their methods sets differ). For instance, we can ﬁnd the second prime number be- tween 1000 and 10000 by applying methods filter and apply on an interval stream: Stream.range(1000, 10000) filter isPrime at 1 The difference to the previous list-based implementation is that now we do not needlessly construct and test for primality any numbers beyond 3. Consing and appending streams. Two methods in class List which are not sup- ported by class Stream are :: and :::. The reason is that these methods are dis- patched on their right-hand side argument, which means that this argument needs to be evaluated before the method is called. For instance, in the case of x :: xs on lists, the tail xs needs to be evaluated before :: can be called and the new list can be constructed. This does not work for streams, where we require that the tail of a stream should not be evaluated until it is demanded by a tail operation. The argument why list-append ::: cannot be adapted to streams is analogous. Instead of x :: xs, one uses Stream.cons(x, xs) for constructing a stream with ﬁrst element x and (unevaluated) rest xs. Instead of xs ::: ys, one uses the opera- tion xs append ys. Chapter 13 Iterators Iterators are the imperative version of streams. Like streams, iterators describe po- tentially inﬁnite lists. However, there is no data-structure which contains the el- ements of an iterator. Instead, iterators allow one to step through the sequence, using two abstract methods next and hasNext. trait Iterator[+A] { def hasNext: Boolean def next: A Method next returns successive elements. Method hasNext indicates whether there are still more elements to be returned by next. Iterators also support some other methods, which are explained later. As an example, here is an application which prints the squares of all numbers from 1 to 100. val it: Iterator[Int] = Iterator.range(1, 100) while (it.hasNext) { val x = it.next println(x * x) } 13.1 Iterator Methods Iterators support a rich set of methods besides next and hasNext, which is described in the following. Many of these methods mimic a corresponding functionality in lists. 104 Iterators Append. Method append constructs an iterator which resumes with the given iter- ator it after the current iterator has ﬁnished. def append[B >: A](that: Iterator[B]): Iterator[B] = new Iterator[B] { def hasNext = Iterator.this.hasNext || that.hasNext def next = if (Iterator.this.hasNext) Iterator.this.next else that.next } The terms Iterator.this.next and Iterator.this.hasNext in the deﬁnition of append call the corresponding methods as they are deﬁned in the enclosing Iterator class. If the Iterator preﬁx to this would have been missing, hasNext and next would have called recursively the methods being deﬁned in the result of append, which is not what we want. Map, FlatMap, Foreach. Method map constructs an iterator which returns all ele- ments of the original iterator transformed by a given function f. def map[B](f: A => B): Iterator[B] = new Iterator[B] { def hasNext = Iterator.this.hasNext def next = f(Iterator.this.next) } Method flatMap is like method map, except that the transformation function f now returns an iterator. The result of flatMap is the iterator resulting from appending together all iterators returned from successive calls of f. def flatMap[B](f: A => Iterator[B]): Iterator[B] = new Iterator[B] { private var cur: Iterator[B] = Iterator.empty def hasNext: Boolean = if (cur.hasNext) true else if (Iterator.this.hasNext) { cur = f(Iterator.this.next); hasNext } else false def next: B = if (cur.hasNext) cur.next else if (Iterator.this.hasNext) { cur = f(Iterator.this.next); next } else error("next on empty iterator") } Closely related to map is the foreach method, which applies a given function to all elements of an iterator, but does not construct a list of results def foreach(f: A => Unit): Unit = while (hasNext) { f(next) } 13.1 Iterator Methods 105 Filter. Method filter constructs an iterator which returns all elements of the original iterator that satisfy a criterion p. def filter(p: A => Boolean) = new BufferedIterator[a] { private val source = Iterator.this.buffered private def skip { while (source.hasNext && !p(source.head)) { source.next } } def hasNext: Boolean = { skip; source.hasNext } def next: A = { skip; source.next } def head: A = { skip; source.head } } In fact, filter returns instances of a subclass of iterators which are “buffered”. A BufferedIterator object is an iterator which has in addition a method head. This method returns the element which would otherwise have been returned by head, but does not advance beyond that element. Hence, the element returned by head is returned again by the next call to head or next. Here is the deﬁnition of the BufferedIterator trait. trait BufferedIterator[+A] extends Iterator[A] { def head: A } Since map, flatMap, filter, and foreach exist for iterators, it follows that for- comprehensions and for-loops can also be used on iterators. For instance, the ap- plication which prints the squares of numbers between 1 and 100 could have equiv- alently been expressed as follows. for (i <- Iterator.range(1, 100)) println(i * i) Zip. Method zip takes another iterator and returns an iterator consisting of pairs of corresponding elements returned by the two iterators. def zip[B](that: Iterator[B]) = new Iterator[(a, b)] { def hasNext = Iterator.this.hasNext && that.hasNext def next = {Iterator.this.next, that.next} } } 106 Iterators 13.2 Constructing Iterators Concrete iterators need to provide implementations for the two abstract methods next and hasNext in class Iterator. The simplest iterator is Iterator.empty which always returns an empty sequence: object Iterator { object empty extends Iterator[Nothing] { def hasNext = false def next = error("next on empty iterator") } A more interesting iterator enumerates all elements of an array. This iterator is con- structed by the fromArray method, which is also deﬁned in the object Iterator def fromArray[A](xs: Array[A]) = new Iterator[A] { private var i = 0 def hasNext: Boolean = i < xs.length def next: A = if (i < xs.length) { val x = xs(i); i += 1; x } else error("next on empty iterator") } Another iterator enumerates an integer interval. The Iterator.range function re- turns an iterator which traverses a given interval of integer values. It is deﬁned as follows. object Iterator { def range(start: Int, end: Int) = new Iterator[Int] { private var current = start def hasNext = current < end def next = { val r = current if (current < end) current += 1 else error("end of iterator") r } } } All iterators seen so far terminate eventually. It is also possible to deﬁne iterators that go on forever. For instance, the following iterator returns successive integers from some start value1. 1Due to the ﬁnite representation of type int, numbers will wrap around at 231. 13.3 Using Iterators 107 def from(start: Int) = new Iterator[Int] { private var last = start - 1 def hasNext = true def next = { last += 1; last } } 13.3 Using Iterators Here are two more examples how iterators are used. First, to print all elements of an array xs: Array[Int], one can write: Iterator.fromArray(xs) foreach (x => println(x)) Or, using a for-comprehension: for (x <- Iterator.fromArray(xs)) println(x) As a second example, consider the problem of ﬁnding the indices of all the elements in an array of doubles greater than some limit. The indices should be returned as an iterator. This is achieved by the following expression. import Iterator._ fromArray(xs) .zip(from(0)) .filter(case (x, i) => x > limit) .map(case (x, i) => i) Or, using a for-comprehension: import Iterator._ for ((x, i) <- fromArray(xs) zip from(0); x > limit) yield i Chapter 14 Lazy Values Lazy values provide a way to delay initialization of a value until the ﬁrst time it is accessed. This may be useful when dealing with values that might not be needed during execution, and whose computational cost is signifﬁcant. As a ﬁrst example, let’s consider a database of employees, containing for each employee its manager and its team. case class Employee(id: Int, name: String, managerId: Int) { val manager: Employee = Db.get(managerId) val team: List[Employee] = Db.team(id) } The Employee class given above will eagerly initialize all its ﬁelds, loading the whole employee table in memory. This is certainly not optimal, and it can be easily im- proved my making the ﬁelds lazy. This way we delay the database access until it is really needed, if it is ever needed. case class Employee(id: Int, name: String, managerId: Int) { lazy val manager: Employee = Db.get(managerId) lazy val team: List[Employee] = Db.team(id) } To see what is really happening, we can use this mockup database which shows when records are fetched: object Db { val table = Map(1 -> (1, "Haruki Murakami", -1), 2 -> (2, "Milan Kundera", 1), 110 Lazy Values 3 -> (3, "Jeffrey Eugenides", 1), 4 -> (4, "Mario Vargas Llosa", 1), 5 -> (5, "Julian Barnes", 2)) def team(id: Int) = { for (rec <- table.values.toList; if rec._3 == id) yield recToEmployee(rec) } def get(id: Int) = recToEmployee(table(id)) private def recToEmployee(rec: (Int, String, Int)) = { println("[db] fetching " + rec._1) Employee(rec._1, rec._2, rec._3) } } The output when running a program that retrieves one employee conﬁrms that the database is only accessed when referring the lazy values. Another use of lazy values is to resolve the initialization order of applications com- posed of several modules. Before lazy values were introduced, the same effect was achieved by using object deﬁnitions. As a second example, we consider a compiler composed of several modules. We look ﬁrst at a simple symbol table that deﬁnes a class for symbols and two predeﬁned functions. class Symbols(val compiler: Compiler) { import compiler.types._ val Add = new Symbol("+", FunType(List(IntType, IntType), IntType)) val Sub = new Symbol("-", FunType(List(IntType, IntType), IntType)) class Symbol(name: String, tpe: Type) { override def toString = name + ": " + tpe } } The symbols module is parameterized with a Compiler instance, which provides ac- cess to other services, such as the types module. In our example there are only two predeﬁned functions, addition and subtraction, and their deﬁnitions depend on the types module. class Types(val compiler: Compiler) { import compiler.symtab._ abstract class Type case class FunType(args: List[Type], res: Type) extends Type 111 case class NamedType(sym: Symbol) extends Type case object IntType extends Type } In order to hook the two components together a compiler object is created and passed as an argument to the two components. class Compiler { val symtab = new Symbols(this) val types = new Types(this) } Unfortunately, the straight-forward approach fails at runtime because the symtab module needs the types module. In general, the dependency between modules can be complicated and getting the right initialization order is difﬁcult, or even im- possible when there are cycles. The easy ﬁx is to make such ﬁelds lazy and let the compiler ﬁgure out the right order. class Compiler { lazy val symtab = new Symbols(this) lazy val types = new Types(this) } Now the two modules are initialized on ﬁrst access, and the compiler may run as expected. Syntax The lazy modiﬁer is allowed only on concrete value deﬁnitions. All typing rules for value deﬁnitions apply for lazy values as well, with one restriction removed: recursive local values are allowed. Chapter 15 Implicit Parameters and Conver- sions Implicit parameters and conversions are powerful tools for custimizing existing li- braries and for creating high-level abstractions. As an example, let’s start with an abstract class of semi-groups that support an unspeciﬁed add operation. abstract class SemiGroup[A] { def add(x: A, y: A): A } Here’s a subclass Monoid of SemiGroup which adds a unit element. abstract class Monoid[A] extends SemiGroup[A] { def unit: A } Here are two implementations of monoids: object stringMonoid extends Monoid[String] { def add(x: String, y: String): String = x.concat(y) def unit: String = "" } object intMonoid extends Monoid[Int] { def add(x: Int, y: Int): Int = x + y def unit: Int = 0 } A sum method, which works over arbitrary monoids, can be written in plain Scala as follows. def sum[A](xs: List[A])(m: Monoid[A]): A = 114 Implicit Parameters and Conversions if (xs.isEmpty) m.unit else m.add(xs.head, sum(m)(xs.tail) This sum method can be called as follows: sum(List("a", "bc", "def"))(stringMonoid) sum(List(1, 2, 3))(intMonoid) All this works, but it is not very nice. The problem is that the monoid implementa- tions have to be passed into all code that uses them. We would sometimes wish that the system could ﬁgure out the correct arguments automatically, similar to what is done when type arguments are inferred. This is what implicit parameters provide. Implicit Parameters: The Basics In Scala 2 there is a new implicit keyword that can be used at the beginning of a parameter list. Syntax: ParamClauses ::= {‘(’ [Param {‘,’ Param}] ’)’} [‘(’ implicit Param {‘,’ Param} ‘)’] If the keyword is present, it makes all parameters in the list implicit. For instance, the following version of sum has m as an implicit parameter. def sum[A](xs: List[A])(implicit m: Monoid[A]): A = if (xs.isEmpty) m.unit else m.add(xs.head, sum(xs.tail)) As can be seen from the example, it is possible to combine normal and implicit parameters. However, there may only be one implicit parameter list for a method or constructor, and it must come last. implicit can also be used as a modiﬁer for deﬁnitions and declarations. Examples: implicit object stringMonoid extends Monoid[String] { def add(x: String, y: String): String = x.concat(y) def unit: String = "" } implicit object intMonoid extends Monoid[Int] { def add(x: Int, y: Int): Int = x + y def unit: Int = 0 } The principal idea behind implicit parameters is that arguments for them can be left out from a method call. If the arguments corresponding to an implicit parameter section are missing, they are inferred by the Scala compiler. 115 The actual arguments that are eligible to be passed to an implicit parameter are all identiﬁers X that can be accessed at the point of the method call without a preﬁx and that denote an implicit deﬁnition or parameter. If there are several eligible arguments which match the implicit parameter’s type, the Scala compiler will chose a most speciﬁc one, using the standard rules of static overloading resolution. For instance, assume the call sum(List(1, 2, 3)) in a context where stringMonoid and intMonoid are visible. We know that the formal type parameter a of sum needs to be instantiated to int. The only eligible value which matches the implicit formal parameter type Monoid[Int] is intMonoid so this object will be passed as implicit parameter. This discussion also shows that implicit parameters are inferred after any type ar- guments are inferred. Implicit Conversions Say you have an expression E of type T which is expected to type S.T does not con- form to S and is not convertible to S by some other predeﬁned conversion. Then the Scala compiler will try to apply as last resort an implicit conversion I(E). Here, I is an identiﬁer denoting an implicit deﬁnition or parameter that is accessible without a preﬁx at the point of the conversion, that can be applied to arguments of type T and whose result type conforms to the expected type S. Implicit conversions can also be applied in member selections. Given a selection E.x where x is not a member of the type E, the Scala compiler will try to insert an implicit conversion I(E).x, so that x is a member of I(E). Here is an example of an implicit conversion function that converts integers into instances of class scala.Ordered: implicit def int2ordered(x: Int): Ordered[Int] = new Ordered[Int] { def compare(y: Int): Int = if (x < y1) -1 else if (x > y1) 1 else 0 } View Bounds View bounds are convenient syntactic sugar for implicit parameters. Consider for instance a generic sort method: def sort[A <% Ordered[A]](xs: List[A]): List[A] = if (xs.isEmpty || xs.tail.isEmpty) xs 116 Implicit Parameters and Conversions else { val {ys, zs} = xs.splitAt(xs.length / 2) merge(ys, zs) } The view bounded type parameter [a <% Ordered[a]] expresses that sort is ap- plicable to lists of type a such that there exists an implicit conversion from a to Ordered[a]. The deﬁnition is treated as a shorthand for the following method sig- nature with an implicit parameter: def sort[A](xs: List[A])(implicit c: A => Ordered[A]): List[A] = ... (Here, the parameter name c is chosen arbitrarily in a way that does not collide with other names in the program.) As a more detailed example, consider the merge method that comes with the sort method above: def merge[A <% Ordered[A]](xs: List[A], ys: List[A]): List[A] = if (xs.isEmpty) ys else if (ys.isEmpty) xs else if (xs.head < ys.head) xs.head :: merge(xs.tail, ys) else if ys.head :: merge(xs, ys.tail) After expanding view bounds and inserting implicit conversions, this method im- plementation becomes: def merge[A](xs: List[A], ys: List[A]) (implicit c: A => Ordered[A]): List[A] = if (xs.isEmpty) ys else if (ys.isEmpty) xs else if (c(xs.head) < ys.head) xs.head :: merge(xs.tail, ys) else if ys.head :: merge(xs, ys.tail)(c) The last two lines of this method deﬁnition illustrate two different uses of the im- plicit parameter c. It is applied in a conversion in the condition of the second to last line, and it is passed as implicit argument in the recursive call to merge on the last line. Chapter 16 Hindley/Milner Type Inference This chapter demonstrates Scala’s data types and pattern matching by developing a type inference system in the Hindley/Milner style [Mil78]. The source language for the type inferencer is lambda calculus with a let construct called Mini-ML. Abstract syntax trees for the Mini-ML are represented by the following data type of Terms. abstract class Term {} case class Var(x: String) extends Term { override def toString = x } case class Lam(x: String, e: Term) extends Term { override def toString = "(\\" + x + "." + e + ")" } case class App(f: Term, e: Term) extends Term { override def toString = "(" + f + " " + e + ")" } case class Let(x: String, e: Term, f: Term) extends Term { override def toString = "let " + x + " = " + e + " in " + f } There are four tree constructors: Var for variables, Lam for function abstractions, App for function applications, and Let for let expressions. Each case class overrides the toString method of class Any, so that terms can be printed in legible form. We next deﬁne the types that are computed by the inference system. sealed abstract class Type {} case class Tyvar(a: String) extends Type { override def toString = a } case class Arrow(t1: Type, t2: Type) extends Type { override def toString = "(" + t1 + "->" + t2 + ")" } 118 Hindley/Milner Type Inference case class Tycon(k: String, ts: List[Type]) extends Type { override def toString = k + (if (ts.isEmpty) "" else ts.mkString("[", ",", "]")) } There are three type constructors: Tyvar for type variables, Arrow for function types and Tycon for type constructors such as Boolean or List. Type constructors have as component a list of their type parameters. This list is empty for type constants such as Boolean. Again, the type constructors implement the toString method in order to display types legibly. Note that Type is a sealed class. This means that no subclasses or data constructors that extend Type can be formed outside the sequence of deﬁnitions in which Type is deﬁned. This makes Type a closed algebraic data type with exactly three alternatives. By contrast, type Term is an open algebraic type for which further alternatives can be deﬁned. The main parts of the type inferencer are contained in object typeInfer. We start with a utility function which creates fresh type variables: object typeInfer { private var n: Int = 0 def newTyvar(): Type = { n += 1; Tyvar("a" + n) } We next deﬁne a class for substitutions. A substitution is an idempotent function from type variables to types. It maps a ﬁnite number of type variables to some types, and leaves all other type variables unchanged. The meaning of a substitution is extended point-wise to a mapping from types to types. abstract class Subst extends Function1[Type,Type] { def lookup(x: Tyvar): Type def apply(t: Type): Type = t match { case tv @ Tyvar(a) => val u = lookup(tv); if (t == u) t else apply(u) case Arrow(t1, t2) => Arrow(apply(t1), apply(t2)) case Tycon(k, ts) => Tycon(k, ts map apply) } def extend(x: Tyvar, t: Type) = new Subst { def lookup(y: Tyvar): Type = if (x == y) t else Subst.this.lookup(y) } } val emptySubst = new Subst { def lookup(t: Tyvar): Type = t } 119 We represent substitutions as functions, of type Type => Type. This is achieved by making class Subst inherit from the unary function type Function1[Type, Type]1. To be an instance of this type, a substitution s has to implement an apply method that takes a Type as argument and yields another Type as result. A function applica- tion s(t) is then interpreted as s.apply(t). The lookup method is abstract in class Subst. There are two concrete forms of sub- stitutions which differ in how they implement this method. One form is deﬁned by the emptySubst value, the other is deﬁned by the extend method in class Subst. The next data type describes type schemes, which consist of a type and a list of names of type variables which appear universally quantiﬁed in the type scheme. For instance, the type scheme ∀a∀b.a→b would be represented in the type checker as: TypeScheme(List(Tyvar("a"), Tyvar("b")), Arrow(Tyvar("a"), Tyvar("b"))) . The class deﬁnition of type schemes does not carry an extends clause; this means that type schemes extend directly class AnyRef. Even though there is only one pos- sible way to construct a type scheme, a case class representation was chosen since it offers convenient ways to decompose an instance of this type into its parts. case class TypeScheme(tyvars: List[Tyvar], tpe: Type) { def newInstance: Type = { (emptySubst /: tyvars) ((s, tv) => s.extend(tv, newTyvar())) (tpe) } } Type scheme objects come with a method newInstance, which returns the type con- tained in the scheme after all universally type variables have been renamed to fresh variables. The implementation of this method folds (with /:) the type scheme’s type variables with an operation which extends a given substitution s by renaming a given type variable tv to a fresh type variable. The resulting substitution renames all type variables of the scheme to fresh ones. This substitution is then applied to the type part of the type scheme. The last type we need in the type inferencer is Env, a type for environments, which associate variable names with type schemes. They are represented by a type alias Env in module typeInfer: type Env = List[(String, TypeScheme)] There are two operations on environments. The lookup function returns the type scheme associated with a given name, or null if the name is not recorded in the environment. 1 The class inherits the function type as a mixin rather than as a direct superclass. This is because in the current Scala implementation, the Function1 type is a Java interface, which cannot be used as a direct superclass of some other class. 120 Hindley/Milner Type Inference def lookup(env: Env, x: String): TypeScheme = env match { case List() => null case (y, t) :: env1 => if (x == y) t else lookup(env1, x) } The gen function turns a given type into a type scheme, quantifying over all type variables that are free in the type, but not in the environment. def gen(env: Env, t: Type): TypeScheme = TypeScheme(tyvars(t) diff tyvars(env), t) The set of free type variables of a type is simply the set of all type variables which oc- cur in the type. It is represented here as a list of type variables, which is constructed as follows. def tyvars(t: Type): List[Tyvar] = t match { case tv @ Tyvar(a) => List(tv) case Arrow(t1, t2) => tyvars(t1) union tyvars(t2) case Tycon(k, ts) => (List[Tyvar]() /: ts) ((tvs, t) => tvs union tyvars(t)) } Note that the syntax tv @ ... in the ﬁrst pattern introduces a variable which is bound to the pattern that follows. Note also that the explicit type parameter [Tyvar] in the expression of the third clause is needed to make local type inference work. The set of free type variables of a type scheme is the set of free type variables of its type component, excluding any quantiﬁed type variables: def tyvars(ts: TypeScheme): List[Tyvar] = tyvars(ts.tpe) diff ts.tyvars Finally, the set of free type variables of an environment is the union of the free type variables of all type schemes recorded in it. def tyvars(env: Env): List[Tyvar] = (List[Tyvar]() /: env) ((tvs, nt) => tvs union tyvars(nt._2)) A central operation of Hindley/Milner type checking is uniﬁcation, which computes a substitution to make two given types equal (such a substitution is called a uniﬁer). Function mgu computes the most general uniﬁer of two given types t and u under a pre-existing substitution s. That is, it returns the most general substitution s0 which extends s, and which makes s0(t) and s0(u) equal types. def mgu(t: Type, u: Type, s: Subst): Subst = (s(t), s(u)) match { case (Tyvar(a), Tyvar(b)) if (a == b) => 121 s case (Tyvar(a), _) if !(tyvars(u) contains a) => s.extend(Tyvar(a), u) case (_, Tyvar(a)) => mgu(u, t, s) case (Arrow(t1, t2), Arrow(u1, u2)) => mgu(t1, u1, mgu(t2, u2, s)) case (Tycon(k1, ts), Tycon(k2, us)) if (k1 == k2) => (s /: (ts zip us)) ((s, tu) => mgu(tu._1, tu._2, s)) case _ => throw new TypeError("cannot unify " + s(t) + " with " + s(u)) } The mgu function throws a TypeError exception if no uniﬁer substitution exists. This can happen because the two types have different type constructors at correspond- ing places, or because a type variable is uniﬁed with a type that contains the type variable itself. Such exceptions are modeled here as instances of case classes that inherit from the predeﬁned Exception class. case class TypeError(s: String) extends Exception(s) {} The main task of the type checker is implemented by function tp. This function takes as parameters an environment env, a term e, a proto-type t, and a pre-existing substitution s. The function yields a substitution s0 that extends s and that turns s0(env) ` e : s0(t) into a derivable type judgment according to the derivation rules of the Hindley/Milner type system [Mil78]. A TypeError exception is thrown if no such substitution exists. def tp(env: Env, e: Term, t: Type, s: Subst): Subst = { current = e e match { case Var(x) => val u = lookup(env, x) if (u == null) throw new TypeError("undefined: " + x) else mgu(u.newInstance, t, s) case Lam(x, e1) => val a, b = newTyvar() val s1 = mgu(t, Arrow(a, b), s) val env1 = {x, TypeScheme(List(), a)} :: env tp(env1, e1, b, s1) case App(e1, e2) => val a = newTyvar() val s1 = tp(env, e1, Arrow(a, t), s) tp(env, e2, a, s1) 122 Hindley/Milner Type Inference case Let(x, e1, e2) => val a = newTyvar() val s1 = tp(env, e1, a, s) tp({x, gen(env, s1(a))} :: env, e2, t, s1) } } var current: Term = null To aid error diagnostics, the tp function stores the currently analyzed sub-term in variable current. Thus, if type checking is aborted with a TypeError exception, this variable will contain the subterm that caused the problem. The last function of the type inference module, typeOf, is a simpliﬁed facade for tp. It computes the type of a given term e in a given environment env. It does so by creating a fresh type variable a, computing a typing substitution that makes env ` e : a into a derivable type judgment, and returning the result of applying the substitution to a. def typeOf(env: Env, e: Term): Type = { val a = newTyvar() tp(env, e, a, emptySubst)(a) } }// end typeInfer To apply the type inferencer, it is convenient to have a predeﬁned environment that contains bindings for commonly used constants. The module predefined deﬁnes an environment env that contains bindings for the types of booleans, numbers and lists together with some primitive operations over them. It also deﬁnes a ﬁxed point operator fix, which can be used to represent recursion. object predefined { val booleanType = Tycon("Boolean", List()) val intType = Tycon("Int", List()) def listType(t: Type) = Tycon("List", List(t)) private def gen(t: Type): typeInfer.TypeScheme = typeInfer.gen(List(), t) private val a = typeInfer.newTyvar() val env = List( {"true", gen(booleanType)}, {"false", gen(booleanType)}, {"if", gen(Arrow(booleanType, Arrow(a, Arrow(a, a))))}, {"zero", gen(intType)}, {"succ", gen(Arrow(intType, intType))}, {"nil", gen(listType(a))}, {"cons", gen(Arrow(a, Arrow(listType(a), listType(a))))}, {"isEmpty", gen(Arrow(listType(a), booleanType))}, 123 {"head", gen(Arrow(listType(a), a))}, {"tail", gen(Arrow(listType(a), listType(a)))}, {"fix", gen(Arrow(Arrow(a, a), a))} ) } Here’s an example how the type inferencer can be used. Let’s deﬁne a function showType which returns the type of a given term computed in the predeﬁned en- vironment Predefined.env: object testInfer { def showType(e: Term): String = try { typeInfer.typeOf(predefined.env, e).toString } catch { case typeInfer.TypeError(msg) => "\n cannot type: " + typeInfer.current + "\n reason: " + msg } Then the application > testInfer.showType(Lam("x", App(App(Var("cons"), Var("x")), Var("nil")))) would give the response > (a6->List[a6]) Exercise 16.0.1 Extend the Mini-ML type inferencer with a letrec construct which allows the deﬁnition of recursive functions. Syntax: letrec ident "=" term in term . The typing of letrec is as for let, except that the deﬁned identiﬁer is visible in the deﬁning expression. Using letrec, the length function for lists can now be deﬁned as follows. letrec length = \xs. if (isEmpty xs) zero (succ (length (tail xs))) in ... Chapter 17 Abstractions for Concurrency This section reviews common concurrent programming patterns and shows how they can be implemented in Scala. 17.1 Signals and Monitors Example 17.1.1 The monitor provides the basic means for mutual exclusion of pro- cesses in Scala. Every instance of class AnyRef can be used as a monitor by calling one or more of the methods below. def synchronized[A] (e: => A): A def wait() def wait(msec: Long) def notify() def notifyAll() The synchronized method executes its argument computation e in mutual exclu- sive mode – at any one time, only one thread can execute a synchronized argument of a given monitor. Threads can suspend inside a monitor by waiting on a signal. Threads that call the wait method wait until a notify method of the same object is called subsequently by some other thread. Calls to notify with no threads waiting for the signal are ignored. There is also a timed form of wait, which blocks only as long as no signal was re- ceived or the speciﬁed amount of time (given in milliseconds) has elapsed. Fur- thermore, there is a notifyAll method which unblocks all threads which wait for the signal. These methods, as well as class Monitor are primitive in Scala; they are implemented in terms of the underlying runtime system. 126 Abstractions for Concurrency Typically, a thread waits for some condition to be established. If the condition does not hold at the time of the wait call, the thread blocks until some other thread has established the condition. It is the responsibility of this other thread to wake up waiting processes by issuing a notify or notifyAll. Note however, that there is no guarantee that a waiting process gets to run immediately after the call to notify is is- sued. It could be that other processes get to run ﬁrst which invalidate the condition again. Therefore, the correct form of waiting for a condition C uses a while loop: while (!C) wait() As an example of how monitors are used, here is is an implementation of a bounded buffer class. class BoundedBuffer[A](N: Int) { var in = 0, out = 0, n = 0 val elems = new Array[A](N) def put(x: A) = synchronized { while (n >= N) wait() elems(in) = x ; in = (in + 1) % N ; n = n + 1 if (n == 1) notifyAll() } def get: A = synchronized { while (n == 0) wait() val x = elems(out) ; out = (out + 1) % N ; n = n - 1 if (n == N - 1) notifyAll() x } } And here is a program using a bounded buffer to communicate between a producer and a consumer process. import scala.concurrent.ops._ ... val buf = new BoundedBuffer[String](10) spawn { while (true){ val s = produceString ; buf.put(s) } } spawn { while (true){ val s = buf.get ; consumeString(s) } } } The spawn method spawns a new thread which executes the expression given in the parameter. It is deﬁned in object concurrent.ops as follows. def spawn(p: => Unit) { val t = new Thread() { override def run() = p } t.start() 17.2 SyncVars 127 } 17.2 SyncVars A synchronized variable (or syncvar for short) offers get and put operations to read and set the variable. get operations block until the variable has been deﬁned. An unset operation resets the variable to undeﬁned state. Here’s the standard implementation of synchronized variables. package scala.concurrent class SyncVar[A] { private var isDefined: Boolean = false private var value: A = _ def get = synchronized { while (!isDefined) wait() value } def set(x: A) = synchronized { value = x; isDefined = true; notifyAll() } def isSet: Boolean = synchronized { isDefined } def unset = synchronized { isDefined = false } } 17.3 Futures A future is a value which is computed in parallel to some other client thread, to be used by the client thread at some future time. Futures are used in order to make good use of parallel processing resources. A typical usage is: import scala.concurrent.ops._ ... val x = future(someLengthyComputation) anotherLengthyComputation val y = f(x()) + g(x()) The future method is deﬁned in object scala.concurrent.ops as follows. def future[A](p: => A): Unit => A = { 128 Abstractions for Concurrency val result = new SyncVar[A] fork { result.set(p) } (() => result.get) } The future method gets as parameter a computation p to be performed. The type of the computation is arbitrary; it is represented by future’s type parameter a. The future method deﬁnes a guard result, which takes a parameter representing the result of the computation. It then forks off a new thread that computes the result and invokes the result guard when it is ﬁnished. In parallel to this thread, the func- tion returns an anonymous function of type a. When called, this functions waits on the result guard to be invoked, and, once this happens returns the result argument. At the same time, the function reinvokes the result guard with the same argument, so that future invocations of the function can return the result immediately. 17.4 Parallel Computations The next example presents a function par which takes a pair of computations as parameters and which returns the results of the computations in another pair. The two computations are performed in parallel. The function is deﬁned in object scala.concurrent.ops as follows. def par[A, B](xp: => A, yp: => B): (A, B) = { val y = new SyncVar[B] spawn { y set yp } (xp, y.get) } Deﬁned in the same place is a function replicate which performs a number of replicates of a computation in parallel. Each replication instance is passed an inte- ger number which identiﬁes it. def replicate(start: Int, end: Int)(p: Int => Unit) { if (start == end) () else if (start + 1 == end) p(start) else { val mid = (start + end) / 2 spawn { replicate(start, mid)(p) } replicate(mid, end)(p) } } 17.5 Semaphores 129 The next function uses replicate to perform parallel computations on all elements of an array. def parMap[A,B](f: A => B, xs: Array[A]): Array[B] = { val results = new Array[B](xs.length) replicate(0, xs.length) { i => results(i) = f(xs(i)) } results } 17.5 Semaphores A common mechanism for process synchronization is a lock (or: semaphore). A lock offers two atomic actions: acquire and release. Here’s the implementation of a lock in Scala: package scala.concurrent class Lock { var available = true def acquire = synchronized { while (!available) wait() available = false } def release = synchronized { available = true notify() } } 17.6 Readers/Writers A more complex form of synchronization distinguishes between readers which ac- cess a common resource without modifying it and writers which can both access and modify it. To synchronize readers and writers we need to implement opera- tions startRead, startWrite, endRead, endWrite, such that: • there can be multiple concurrent readers, • there can only be one writer at one time, • pending write requests have priority over pending read requests, but don’t preempt ongoing read operations. The following implementation of a readers/writers lock is based on the mailbox concept (see Section 17.10). 130 Abstractions for Concurrency import scala.concurrent._ class ReadersWriters { val m = new MailBox private case class Writers(n: Int), Readers(n: Int) { m send this } Writers(0); Readers(0) def startRead = m receive { case Writers(n) if n == 0 => m receive { case Readers(n) => Writers(0); Readers(n+1) } } def startWrite = m receive { case Writers(n) => Writers(n+1) m receive { case Readers(n) if n == 0 => } } def endRead = m receive { case Readers(n) => Readers(n-1) } def endWrite = m receive { case Writers(n) => Writers(n-1); if (n == 0) Readers(0) } } 17.7 Asynchronous Channels A fundamental way of interprocess communication is the asynchronous channel. Its implementation makes use the following simple class for linked lists: class LinkedList[A] { var elem: A = _ var next: LinkedList[A] = null } To facilitate insertion and deletion of elements into linked lists, every reference into a linked list points to the node which precedes the node which conceptually forms the top of the list. Empty linked lists start with a dummy node, whose successor is null. The channel class uses a linked list to store data that has been sent but not read yet. At the opposite end, threads that wish to read from an empty channel, register their presence by incrementing the nreaders ﬁeld and waiting to be notiﬁed. package scala.concurrent 17.8 Synchronous Channels 131 class Channel[A] { class LinkedList[A] { var elem: A = _ var next: LinkedList[A] = null } private var written = new LinkedList[A] private var lastWritten = written private var nreaders = 0 def write(x: A) = synchronized { lastWritten.elem = x lastWritten.next = new LinkedList[A] lastWritten = lastWritten.next if (nreaders > 0) notify() } def read: A = synchronized { if (written.next == null){ nreaders = nreaders + 1; wait(); nreaders = nreaders - 1 } val x = written.elem written = written.next x } } 17.8 Synchronous Channels Here’s an implementation of synchronous channels, where the sender of a message blocks until that message has been received. Synchronous channels only need a single variable to store messages in transit, but three signals are used to coordinate reader and writer processes. package scala.concurrent class SyncChannel[A] { private var data: A = _ private var reading = false private var writing = false def write(x: A) = synchronized { while (writing) wait() data = x writing = true 132 Abstractions for Concurrency if (reading) notifyAll() else while (!reading) wait() } def read: A = synchronized { while (reading) wait() reading = true while (!writing) wait() val x = data writing = false reading = false notifyAll() x } } 17.9 Workers Here’s an implementation of a compute server in Scala. The server implements a future method which evaluates a given expression in parallel with its caller. Unlike the implementation in Section 17.3 the server computes futures only with a prede- ﬁned number of threads. A possible implementation of the server could run each thread on a separate processor, and could hence avoid the overhead inherent in context-switching several threads on a single processor. import scala.concurrent._, scala.concurrent.ops._ class ComputeServer(n: Int) { private abstract class Job { type T def task: T def ret(x: T) } private val openJobs = new Channel[Job]() private def processor(i: Int) { while (true){ val job = openJobs.read job.ret(job.task) } } 17.9 Workers 133 def future[A](p: => A): () => A = { val reply = new SyncVar[A]() openJobs.write{ new Job { type T = A def task = p def ret(x: A) = reply.set(x) } } () => reply.get } spawn(replicate(0, n) { processor }) } Expressions to be computed (i.e. arguments to calls of future) are written to the openJobs channel. A job is an object with • An abstract type t which describes the result of the compute job. • A parameterless task method of type t which denotes the expression to be computed. •A return method which consumes the result once it is computed. The compute server creates n processor processes as part of its initialization. Every such process repeatedly consumes an open job, evaluates the job’s task method and passes the result on to the job’s return method. The polymorphic future method creates a new job where the return method is implemented by a guard named reply and inserts this job into the set of open jobs by calling the isOpen guard. It then waits until the corresponding reply guard is called. The example demonstrates the use of abstract types. The abstract type t keeps track of the result type of a job, which can vary between different jobs. Without abstract types it would be impossible to implement the same class to the user in a statically type-safe way, without relying on dynamic type tests and type casts. Here is some code which uses the compute server to evaluate the expression 41 + 1. object Test with Executable { val server = new ComputeServer(1) val f = server.future(41 + 1) println(f()) } 134 Abstractions for Concurrency 17.10 Mailboxes Mailboxes are high-level, ﬂexible constructs for process synchronization and com- munication. They allow sending and receiving of messages. A message in this con- text is an arbitrary object. There is a special message TIMEOUT which is used to signal a time-out. case object TIMEOUT Mailboxes implement the following signature. class MailBox { def send(msg: Any) def receive[A](f: PartialFunction[Any, A]): A def receiveWithin[A](msec: Long)(f: PartialFunction[Any, A]): A } The state of a mailbox consists of a multi-set of messages. Messages are added to the mailbox with the send method. Messages are removed using the receive method, which is passed a message processor f as argument, which is a partial function from messages to some arbitrary result type. Typically, this function is implemented as a pattern matching expression. The receive method blocks until there is a message in the mailbox for which its message processor is deﬁned. The matching message is then removed from the mailbox and the blocked thread is restarted by applying the message processor to the message. Both sent messages and receivers are ordered in time. A receiver r is applied to a matching message m only if there is no other {message, receiver} pair which precedes m,r in the partial ordering on pairs that orders each component in time. As a simple example of how mailboxes are used, consider a one-place buffer: class OnePlaceBuffer { private val m = new MailBox // An internal mailbox private case class Empty, Full(x: Int) // Types of messages we deal with m send Empty // Initialization def write(x: Int) { m receive { case Empty => m send Full(x) } } def read: Int = m receive { case Full(x) => m send Empty; x } } Here’s how the mailbox class can be implemented: class MailBox { private abstract class Receiver extends Signal { def isDefined(msg: Any): Boolean var msg = null 17.10 Mailboxes 135 } We deﬁne an internal class for receivers with a test method isDefined, which indi- cates whether the receiver is deﬁned for a given message. The receiver inherits from class Signal a notify method which is used to wake up a receiver thread. When the receiver thread is woken up, the message it needs to be applied to is stored in the msg variable of Receiver. private val sent = new LinkedList[Any] private var lastSent = sent private val receivers = new LinkedList[Receiver] private var lastReceiver = receivers The mailbox class maintains two linked lists, one for sent but unconsumed mes- sages, the other for waiting receivers. def send(msg: Any) = synchronized { var r = receivers, r1 = r.next while (r1 != null && !r1.elem.isDefined(msg)) { r = r1; r1 = r1.next } if (r1 != null){ r.next = r1.next; r1.elem.msg = msg; r1.elem.notify } else { lastSent = insert(lastSent, msg) } } The send method ﬁrst checks whether a waiting receiver is applicable to the sent message. If yes, the receiver is notiﬁed. Otherwise, the message is appended to the linked list of sent messages. def receive[A](f: PartialFunction[Any, A]): A = { val msg: Any = synchronized { var s = sent, s1 = s.next while (s1 != null && !f.isDefinedAt(s1.elem)) { s = s1; s1 = s1.next } if (s1 != null){ s.next = s1.next; s1.elem } else { val r = insert(lastReceiver, new Receiver { def isDefined(msg: Any) = f.isDefinedAt(msg) }) lastReceiver = r r.elem.wait() r.elem.msg 136 Abstractions for Concurrency } } f(msg) } The receive method ﬁrst checks whether the message processor function f can be applied to a message that has already been sent but that was not yet consumed. If yes, the thread continues immediately by applying f to the message. Otherwise, a new receiver is created and linked into the receivers list, and the thread waits for a notiﬁcation on this receiver. Once the thread is woken up again, it continues by applying f to the message that was stored in the receiver. The insert method on linked lists is deﬁned as follows. def insert(l: LinkedList[A], x: A): LinkedList[A] = { l.next = new LinkedList[A] l.next.elem = x l.next.next = l.next l } The mailbox class also offers a method receiveWithin which blocks for only a speci- ﬁed maximal amount of time. If no message is received within the speciﬁed time in- terval (given in milliseconds), the message processor argument f will be unblocked with the special TIMEOUT message. The implementation of receiveWithin is quite similar to receive: def receiveWithin[A](msec: Long)(f: PartialFunction[Any, A]): A = { val msg: Any = synchronized { var s = sent, s1 = s.next while (s1 != null && !f.isDefinedAt(s1.elem)) { s = s1; s1 = s1.next } if (s1 != null){ s.next = s1.next; s1.elem } else { val r = insert(lastReceiver, new Receiver { def isDefined(msg: Any) = f.isDefinedAt(msg) }) lastReceiver = r r.elem.wait(msec) if (r.elem.msg == null) r.elem.msg = TIMEOUT r.elem.msg } } f(msg) } 17.11 Actors 137 }// end MailBox The only differences are the timed call to wait, and the statement following it. 17.11 Actors Chapter3 sketched as a program example the implementation of an electronic auc- tion service. This service was based on high-level actor processes that work by in- specting messages in their mailbox using pattern matching. A reﬁned and opti- mized implementation of actors is found in the scala.actors package. We now give a sketch of a simpliﬁed version of the actors library. The code below is different from the implementation in the scala.actors package, so it should be seen as an example how a simple version of actors could be imple- mented. It is not a description how actors are actually deﬁned and implemented in the standard Scala library. For the latter, please refer to the Scala API documenta- tion. A simpliﬁed actor is just a thread whose communication primitives are those of a mailbox. Such an actor can be deﬁned as a mixin composition extension of Java’s standard Thread class with the MailBox class. We also override the run method of the Thread class, so that it executes the behavior of the actor that is deﬁned by its act method. The ! method simply calls the send method of the MailBox class: abstract class Actor extends Thread with MailBox { def act(): Unit override def run(): Unit = act() def !(msg: Any) = send(msg) } Bibliography [ASS96] Harold Abelson, Gerald Jay Sussman, and Julie Sussman. The Structure and Interpretation of Computer Programs, 2nd edition. MIT Press, Cambridge, Massachusetts, 1996. [Mil78] Robin Milner. A Theory of Type Polymorphism in Programming. Journal of Computer and System Sciences, 17:348–375, Dec 1978.