Introduction to Scala : Clueda

An Introduction to Scala
Andreas Neumann, Clueda

a.neumann@clueda.com

Overview
‣ A very short history of Scala

‣ Using Scala - Basics: Compile and run

‣ Introduction to the core concepts : vals, classes, objects, traits, control
structures and functions

‣ Some intermediate concepts in Scala: Pattern Matching, Option, XML,
Futures

‣ Patterns in Scala

‣ Functional Style

About Scala
‣ A JVM Languge

‣ mutltiparadigmatic

‣ functional object orientated

The father of Scala - Martin Odersky
‣ Professor EPFL

‣ Java-Compiler

‣ Java 1.4 Generics

‣ PIZZA

‣ Scala

‣ Co-Founder Typesafe

The company behind Scala -
‣ Training

‣ Consulting

‣ Scala

‣ Actor

‣ Play

Who is using Scala
‣ Xerox

‣ Twitter

‣ Foursquare

‣ Sony

‣ Siemens

‣ clueda

‣ …

Nice features
‣ Can be integrated in any Java based architecture

‣ support concurrent and asynchronous operations out of the box in many
ﬂavors (Futures,Parallel Collections, Akka)

‣ great for embedded DSLs

Introduction
‣ Run code / compile Code

‣ Syntax (some tips)

‣ declare variables

‣ functions

‣ clases, traits and objects

Run code, compile sources
‣ Scala can be used in many diﬀerent environments:

‣ compile and run on a JVM
‣ run as a script
‣ in an interactive console, called REPL

Scala - compiler / scalac
‣ the compiler is called
scalac

‣ The sources are
compiled to Java byte
code

‣ *.class

‣ run class with scala
<name>

Scripting with Scala - Unix
‣ Scala can be used as scripting
language

‣ change mode to excuteable or run
by sh
#!/bin/sh
exec scala "$0" "$@"
!#
!
object HelloWorld {
def main(args: Array[String]) {
println("Hello, world! " +
args.toList)
}
}

HelloWorld.main(args)

Scripting with Scala - Windows
‣ Resembles UNIX use a batch
script instead

‣ *.bat

‣ run
::#!!
@echo off!
call scala %0 %*!
goto :eof!
::!#!
!
object HelloWorld {!
def main(args: Array[String]) {!
println("Hello, world! " +!
args.toList)!
}!
}!
!
HelloWorld.main(args)!

Scala - Interpreter / REPL
‣ part of the scala
installation

‣ Start by typing scala
the shell

‣ :q or CTRL + d to quit
neumann@mac ~> scala
!
!
Welcome to Scala version 2.11.1 (Java HotSpot(TM)
64-Bit Server VM, Java 1.7.0_55).
Type in expressions to have them evaluated.
Type :help for more information.
!
scala> println("Hello World")
Hello World
!
scala> :q
!
!
neumann@mac ~>

Online REPL
‣ https://codebrew.io/

Syntax
‣ No need for „;“ , one expression per line

‣ Still possible to add ; if for example you want several expression within one
line

‣ The dot on method invocations can be dropped. It is best practice to do so
with inﬁx invocation but not with postﬁx operations.
"Hallo Welt !".split(" ")!
//res0: Array[String] = Array(Hallo, Welt, !)!
!
scala> "Hallo Welt !" split " " !
//res2: Array[String] = Array(Hallo, Welt, !)
scala> List(1,2.3).map(_ * 3).head!
//res3: Double = 3.0!
!
scala> ( List(1,2.3) map (_ * 3) ).head!
//res4: Double = 3.0

val, vars
‣ val creates a Value which is not changeable ( like final modifier in Java)

‣ var creates a Variable , which can be reassigned different values

val x = 42!
//x: Int = 42!
!
var y = 99!
//y: Int = 99!
!
y = 1!
y: Int = 1!
!
x = 1!
error: reassignment to val!
x = 1!
^
Example: val and var

Types and type inferrence
‣ Types are introduced after : which is written behind the var/val

!
‣ Giving the type explicitly is optional as the type inference can infer the type.
It’s considered good style to add the type information nonetheless.
s : String = "ein String"
val a = "Hallo"!
//a: java.lang.String = Hallo!
!
val b = 1!
//b: Int = 1!
!
val c = 3.5!
//c: Double = 3.5!
!
val d = List(1,2.0)!
//d: List[Double] = List(1.0, 2.0)

deﬁne methods
‣ methods are introduced with
def

‣ optionally there can be a list
of parameters enclosed by
parentheses

‣ then the body of the function

‣ methods returning a value
put a between name
arguments and the body

‣ the result of the last
expression evaluated within
the body is the return value
def write(aString: String):Unit= {!
println(aString) !
}!
!
write("Hallo ihr alle da draußen!")!
//Hallo ihr alle da draußen!
def add(x: Int, y:Int) : Int = { !
x + y!
}!
!
add(40,2)!
//res0: Int = 42

Collections
‣ Scala has a big Collections library

‣ Collections provide similar interfaces as far as possible

‣ Most collections come in up to four ﬂavors:

‣ basic (mostly = immutable)

‣ immutable

‣ mutable

‣ parallel

Scala Collections Example : List (1 / 2)
val a = List("a","b","c")!
// a: List[java.lang.String] = List(a, b, c)!
val b = List(1,2,3)!
//b: List[Int] = List(1, 2, 3)!
!
a.head!
// java.lang.String = a!
a.tail!
// List[java.lang.String] = List(b, c)!
!
0 :: b!
// List[Int] = List(0, 1, 2, 3)!
a ++ b!
// List[Any] = List(a, b, c, 1, 2, 3)!
a zip b!
// List[(java.lang.String, Int)] = List((a,1), (b,2), (c,3))!
a.sliding(2).toList!
// List[List[String]] = List(List(a, b), List(b, c))!

Scala Collections : Map
val counting = Map(1 -> "eins", 2 -> "zwei", 3 -> "drei")!
// counting: scala.collection.immutable.Map[Int,java.lang.String] =!
// Map((1,eins), (2,zwei), (3,drei))!
!
counting(2)!
//java.lang.String = zwei!
!
counting.get(2)!
//Option[java.lang.String] = Some(zwei)!
!
counting get 99!
// Option[java.lang.String] = None!

Classes
‣ Classes are introduced by the keyword class

‣ Optionally each class has constructor elements in parentheses

‣ optionally there is a class body

‣ Things to look out for

‣ Constructor elements prepended with the keyword val automatically get a
getter method with the same name as the val (uniform access principle)

‣ Constructor elements prepended with the keyword var get a getter method and
a setter method with the same name as the var (uniform access principle)

‣ Every expression within the body gets evaluated and called on object creation
time

Example: A scala class
‣ Instances are created with new <ClassName>
class Document(val title: String, val author: String, yearInt: Int) {!
!
val year = yearInt.toString!
!
def shortCitation: String = author + " : " + title + ". " + year!
}!
!
val scalaBook = !
! new Document("Programming In Scala","Martin Odersky",2011)!
!
println(scalaBook.title)!
println(scalaBook.year)!

Scala Objects
‣ Objects are created using the keyword object

‣ They have NO constructor

‣ Works roughly like a java class with static methods

‣ Calling members ObjectName.member or ObjectName.method
‣ Singleton-Object

Scala Object - Example
object DeepThought {!
!
! val theMeaningOfLife =!
"The meaning of life: 42"!
!
! def speak {!
println(theMeaningOfLife)!
}!
!
}!
!
!
DeepThought.speak!
//The meaning of life: 42

Companion Object
‣ Widely used pattern in Scala

‣ A object with the same name as a class

‣ Native Constructor Pattern

Case Class
‣ Introduce using the keywords case class

‣ Like a „normal class“

‣ Deﬁnes a companion object automatically with apply, unapply and some
other methods

‣ All constructor elements have getter methods as if prepended by val

‣ Adds hashCode and equals based on the constructor Elements

‣ Other classes must not inherit from case classes

Case Class - Example
case class Book(title: String, pages :Int)!
//defined class Book!
!
val book = Book("Necronomicon",1000)!
//book: Book = Book(Necronomicon,1000)!
!
println( book.title )!
//Necronomicon!
!
book == Book("Necronomicon",1000)!
//Boolean = true!
!
scala> book.hashCode!
//-364191203!
!
scala> Book("Necronomicon",1000).hashCode!
//-364191203

Trait
‣ introduced with keyword trait

‣ roughly comparable to a java interface

‣ allows an eﬀect resembling multiple inheritance without the dangers (i.e.
diamond of death)

‣ small building blocks

‣ like ruby mixins

Trait - Example
‣ Two „traits“ are deﬁned as
Traits: Edible and
ExoticTaste

‣ Two classes are deﬁned:
Cake, which implements
edible and ChiliChoc
implements Edible and
ExoticTaste
trait Edible {!
! def taste: String!
! def eat = println(taste)!
}!
!
trait ExoticTaste {!
! def eat: Unit!
!
! def describeTaste = {!
! ! eat!
! ! println("It tastes exotic")!
! }!
}!
!
!
case class Cake() extends Edible {!
! def taste = "sweet"!
}!
!
case class ChilliChoc(taste: String) !
! ! ! ! ! ! extends Edible with ExoticTaste

Trait - Example : Usage
val cake = new Cake()!
cake.eat!
!
val chilliChoc = ChilliChoc("sweet and hot")!
!
chilliChoc.eat!
chilliChoc.describeTaste
scala> val cake = new Cake()
cake: Cake = Cake()
!
scala> cake.eat
sweet
!
!
scala> val chilliChoc = ChilliChoc("sweet and hot")
chilliChoc: ChilliChoc = ChilliChoc(sweet and hot)
!
scala> chilliChoc.eat
sweet and hot
!
scala> chilliChoc.describeTaste
sweet and hot
It tastes exotic

control structures
‣ if, else

‣ while

‣ foreach, map

‣ for-comprehensions

Control structures
‣ Control structures like while, if and else work as in Java or C

‣ In contrast to Java, control structures are functions, i.e. they return a value

‣ while always returns returns Unit
val x = if ( 1 < 2 ) true!
//x: AnyVal = true

functional control structures
‣ some Examples

‣map : Apply function to each given element , keep result (like looping and
collecting the results)

!
‣foreach: Apply function to each given element , drop result
List(1,2,3,4) map (x => x + 1)!
//res1: List[Int] = List(2, 3, 4, 5)
List(1,2,3) foreach ( x => println(s"And a $x") )!
!
//And a 1!
//And a 2!
//And a 3

matching
‣ Pattern matching

‣ keyword match

‣ A group of cases introduced with the keyword case

‣ Switch on Steroids

‣ Pattern Guards allow better control ﬂow

‣Case Classes and Extractor Patterns for easy deconstruction and
extraction of input

Matching Example
case class Book( title: String, pages: Int, year: Int)!
!
val books = List( !
! Book("Programming Scala", 883, 2012),!
! Book("Programming Pearl", 1104, 2000),!
! Book("Necronomicon",666,666),!
! "Ein String“, 5, 42!
)!
!
val bookComments = books map {!
! case Book("Programming Scala", pages, year) => !
! ! s"New Scala Book by Martin Odersky from $year"!
! case Book(title, pages,year) => !
! ! s"$title $pages $year"!
! case x: Int if x > 10 => !
! ! "an integer bigger than 10"!
! case _ => !
! ! "Something else"!
}
books: List[Any] = List(Book(Programming Scala,883,2012), Book(Programming Pearl,1104,2000), Book(Necronomicon,666,666), Ein String, 5,
42)
bookComments: List[String] = List(New Scala Book by Martin Odersky from 2012, Programming Pearl 1104 2000, Necronomicon 666 666, Something
else, Something else, an integer bigger than 10)

For-Comprehensions
‣ program with a DSL that looks a lot like pseudocode

‣ will be translated to map, ﬁlter, ﬂatMap and reduce operations by the
compiler
def isEven(x: Int) = x % 2 == 0!
val integers = for {!
x <- 1 to 99!
if isEven(x)!
if x % 5 == 0!
} yield x!
//integers: scala.collection.immutable.IndexedSeq[Int] = !
Vector(10, 20, 30, 40, 50, 60, 70, 80, 90)
~ Translated ~!
(1 to 99) filter isEven filter ( _ % 5 == 0)

Some language features
‣ Strings: Interpolation and MultilineStrings

‣ Option[Type]

‣ Future[X]

‣ XML erzeugen und bearbeiten

‣ Reguläre Ausdrücke

‣ Parallele Collections

‣ Implicits

‣ String concatenation works with „+“

!
‣ String interpolation: prepending String with „s“ marking Variable with $
!
‣ Complex Expressions are enclosed within ${ }
String Interpolation
"hello" + " " + "world"!
//res2: String = hello world
val w = "world"!
s"hello $w"!
//res3: String = hello world
val w = "world"!
s"$w has length:${w.length}"!
//res4: String = world has length:5!

String Concatenation/Interpolation - Example
val names = List("Roger","Felix", "Bene")!
!
for (name <- names) println("Hello" + name)!
!
//HelloRoger!
//HelloFelix!
//HelloBene
!
for (name <- names) println(s"Hello $name")!
!
!
//Hello Roger!
//Hello Felix!
//Hello Bene

Multiline String
‣ Multiline String is created by using """""" instead of
""
!
!
!
!
‣ allows embedding " in Strings and gets rid of
double Escapes
"""This!
| is a!
| multiline String!
| """!
!
res6: String =!
"This!
is a!
multiline String!
"!
"""Hello "World"! """!
//res12: String = "Hello "World"! "

Combined example Strings
!
for (name <- names) println(!
s"""Hello $name your name!
has length:${ name.size }!
and reads backwards as:"${ name.reverse}"!
"""!
)
Hello Roger your name!
has length:5!
and reads backwards as:"regoR"!
!
Hello Felix your name!
has length:5!
and reads backwards as:"xileF"!
!
Hello Bene your name!
has length:4!
and reads backwards as:"eneB"

String format - The f interpolator
‣prepend String with f ““
‣ Syntax like C printf
def euroToDollar(euro: Double): Double = !
! ! euro * 1.352065!
!
val euro = List(1,2.5,5.12)!
!
euro map euroToDollar foreach { d =>!
println(f"Got $d%2.2f ($d)")!
}
Got 1,35 (1.352065)
Got 3,38 (3.3801625)
Got 6,92 (6.9225728)

The Option-Type
‣ Marks functions that may return a result but also may not return a result

‣ Comes in two ﬂavors: Some and None
!
!
!
‣ like maybe in Haskell

‣ A way to get around checking for null all the time
val x : Option[_] = None!
//x: Option[_] = None!
!
val x : Option[_] = Some("Hello World!")!
//x: Option[_] = Some(Hello World!)!

Option / List Comparison
‣ Option behaves like a list with one element

!
!
‣ An empty Option is called None. None behaves like an empty list.
List( 1 ) map (i => i + 0.5 ) !
//List[Double] = List(1.5)
val y : List[Int] = List()!
//y: List[Int] = List()!
!
y map (i => i+ 0.5)!
//List[Double] = List()
Some( 1 ) map (i => i + 0.5 )!
//Option[Double] = Some(1.5)
// Like an empty List!
val x : Option[Int] = None !
//x: Option[Int] = None!
x map (i => i+ 0.5)!
// Option[Double] = None

Option-Type Beispiel: Option vs. null
bigBangPHD("Leonard")!
//res0: java.lang.String = Ph.D.!
!
bigBangPHD("Howard")!
java.util.NoSuchElementException: key
not found: Howard!
! at scala.collection.MapLike
$class.default(MapLike.scala:223)!
! at scala.collection.immutable.Map
$Map3.default(Map.scala:132)
val bigBangPHD = Map(!
"Leonard" -> "Ph.D.",!
"Sheldon" -> "Ph.D.,Sc.D",!
"Rajesh" -> "Ph.D"!
)!
!
val friends = List("Leonard","Sheldon","Rajesh","Howard")
bigBangPHD get "Leonard"!
//res1: Option[java.lang.String] !
= Some(Ph.D.)!
!
bigBangPHD.get("Sheldon")!
= Some(Ph.D., Sc.D)!
!
bigBangPHD.get("Howard")!
= None

Option -Type :Examples 1
• Used widely throughout Scala
• many builtin methods to handle Option
// Liste mit Options erzeugen !
friends map (bigBangPHD.get(_))!
friends map bigBangPHD.get!
//List[Option[java.lang.String]] = !
//List(Some(Ph.D.), Some(Ph.D.,Sc.D), Some(Ph.D), None)!
!
!
// flatten entfernt None und „entpackt“ Some(thing)!
friends map bigBangPHD.get flatten!
friends flatMap (f => bigBangPHD.get(f))!
//res5: List[java.lang.String] = List(Ph.D., Ph.D.,Sc.D, Ph.D)!
!
// for comprehensions wenden Operationen nur auf Some() an und verwerfen None!
for {!
! person <- friends!
! phd <- bigBangPHD get person!
} yield s"$person has a $phd"!
//List[java.lang.String] = !
//List(Leonard has a Ph.D., Sheldon has a Ph.D.,Sc.D, Rajesh has a Ph.D)

Option -Type : Examples 2‚
// getOrElse erlaubt es einen Standardrückgabewert für None anzugeben, ansonsten wird
Some(thing) „ausgepackt“!
friends!
! .map( n =>(n,bigBangPHD.get(n)) ) // creates Tuple!
! .map{ case (n,d) => !
! ! ! ! n + " " + d.getOrElse("Sheldon tells me you only have a master's degree.") !
! }!
!
//res7: List[java.lang.String] = !
//List(Leonard Ph.D.,!
//Sheldon Ph.D.,Sc.D,!
//Rajesh Ph.D,!
//Howard Sheldon tells me you only have a master's degree.)!
!
// Option Types besitzen Extraktoren für Pattern Matching!
friends map bigBangPHD.get zip friends map {!
case (Some(phd), name ) => name + " : " + phd!
case (None, name) => name + " is just an engineer"!
}!
!
//res10: List[java.lang.String] = List(Leonard : Ph.D.,!
//Sheldon : Ph.D.,Sc.D,!
//Rajesh : Ph.D,!
//Howard is just an engineer)!

Futures
‣ Are a way to abstract over asynchronous computation

‣ non blocking

‣ can be used much like Option

‣ used in many popular Scala libraries

Futures - Plumbing
‣ Import com.scala.concurrent._ for future helpers

‣ Every future needs an ExcecutionContext
import scala.concurrent._!
import ExecutionContext.Implicits.global

Using Futures - Example
val urls = !
List("http://www.clueda.de","http://www.neumann.biz","http://www.an-it.com")!
!
def takeTime = { code to measure time }!
def extractURLs(data: String) : Iterator[String] = …!
def printURLs(data: String) : Unit = extractURLs(data) foreach println!
def printLinks = urls map ( url => printURLs( getData(url) ) )!
!
takeTime( printLinks )!
takeTime( future { printLinks } )!
scala> takeTime( printLinks )
Url -> /favicon.gif
Url -> /stylesheets/refinery/style.css
Url -> /stylesheets/style.css?1380179036
…
res9: (String, List[Unit]) =
(
took 2.109 s,
List((), (), ())
)
takeTime( future { printLinks } )
res10: (String, scala.concurrent.Future[List[Unit]]) =
(
took 0.0 s,
scala.concurrent.impl.Promise$DefaultPromise@..
)
!
scala> Url -> /favicon.gif
Url -> /stylesheets/flexslider.css?1349423712
…

Futures - Getting the result
‣ To get the result of a Future you have to block and wait

!
‣ This is usually bad

‣ Awaiting the result should happen as late as possible as it negates the
beneﬁts one gets using futures
import scala.concurrent.duration._!
!
takeTime( Await.result( Future(printLinks), 10 seconds ))
scala> takeTime( Await.result( Future(printLinks), 10 seconds ))
warning: there were 1 feature warning(s); re-run with -feature for details
Url -> /favicon.gif
Url -> /stylesheets/flexslider.css?1349423712
…
res30: (String, List[Unit]) = (took 1.976 s,List((), (), ()))

Futures - Composing
‣ As futures are Monads ( said it, done! ) they can be composed

‣ The futures run asynchronously and will not wait for each other but await will
wait till the last of the futures has completed or the timeout is reached.
def composedFutures: Future[(Unit,Unit,Unit)] = {!
! val f1 = Future( getAndPrintLinks("http://www.an-it.com") )!
! val f2 = Future( getAndPrintLinks("http://www.neumann.biz") )!
! val f3 = Future( getAndPrintLinks("http://www.clueda.com") )!
!
! for ( d1 <- f1 ; d2 <- f2 ; d3 <- f3) yield (d1,d2,d3)!
}
takeTime { Await.result(composedFutures,10 seconds) }
warning: there were 1 feature warning(s); re-run with -feature for details
Url -> /stylesheets/an-it.css?1339665275
Url -> mobile_stylesheets/mobile.css
Url -> /
Url -> ethnologie-studium
res21: (String, (Unit, Unit, Unit)) = (took 0.834 s,((),(),()))

XML in Scala
‣ XML is a ﬁrst class language citizen as string is in Java or Ruby

‣ It’s possible to embed XML in Scala source code and get syntax highlighting

Scala - XML
‣ Xml can be written within scala sources

‣ IDE s provide syntax-highlighting (Eclipse, Netbeans, IntelliJ)

‣ Code can be embedded using { } within XML literals

Emit XML - Example
case class Book( title: String, pages: Int, year: Int) {!
!
def toXML = !
<book>!
<title>{title}</title>!
<pages>{pages toString}</pages>!
<year>{year toString}</year>!
</book>!
!
}!
!
val books = List( !
Book("Programming Scala", 883, 2012),!
Book("Programming Perl", 1104, 2000),!
Book("Necronomicon",666,666) !
)!
!
for ( book <- books) {!
println(book.toXML)!
}!

Emitted XML
<book>!
<title>Programming Scala</title>!
<pages>883</pages>!
<year>2012</year>!
</book>!
<book>!
<title>Programming Perl</title>!
<pages>1104</pages>!
<year>2000</year>!
</book>!
<book>!
<title>Necronomicon</title>!
<pages>666</pages>!
<year>666</year>!
</book>!

Processing XML
‣ Scala provides an internal DSL inﬂuence providing a XPath like syntax (
instead of // and instead of / )

‣ <xml></xml> "tag" : Shallow -Match

‣ <xml></xml> "tag" : Deep -Match

‣ <xml attribute=„wert“></xml> "@attribut" : Deep -Match on a XML
attribute

‣ (<xml></xml> "tag").text : Extracts the text value of an xml node

processing XML - Example
def toXML = !
<book>!
<pages>{pages}</pages>!
<year>{year}</year>!
</book>!
!
implicit def intToString(in : Int) : String = in.toString!
}!
!
object Book {!
def fromXML(bookXML: scala.xml.NodeSeq) : Book= {!
val title = (bookXML "title").text!
val pages = (bookXML "pages").text.toInt!
val year = (bookXML "year").text.toInt!
new Book(title, pages, year)!
}!
}!

processing XML - Result
val books = !
<books>!
<book>!
<title>Programming Scala</title>!
<pages>883</pages>!
<year>2012</year>!
</book>!
<book>!
<title>Programming Perl</title>!
<pages>1104</pages>!
<year>2000</year>!
</book>!
<book>!
<title>Necronomicon</title>!
<pages>666</pages>!
<year>666</year>!
</book>!
</books>!
!
val booksInstances = (books „book") map Book.fromXML!
val booksPages = (books "pages").map(_.text.toInt)!
booksInstances: scala.collection.immutable.Seq[Book] =
List(Book(Programming Scala,883,2012), Book(Programming Perl,1104,2000), Book(Necronomicon,666,666))
booksPages: scala.collection.immutable.Seq[Int] = List(883, 1104, 666)

Regular Expressions
‣ Creating a regular Expression:.r aus einem String erzeugen:
!
!
!
‣ Uses Java-Regex-Engine to create a NFA

‣ Regex-Object also implement extractors for pattern matching
// Using the Constructor!
new scala.util.matching.Regex("hrefs?=s?"([^"]+)"")!
//Changing a string to a regex with the .r method!
"hrefs?=s?"([^"]+)"".r!
// Using """ , no need to escape " and double escaping of !
"""hrefs?=s?"([^"]+)"""".r !

Regex - Usage
import scala.io.Source!
!
val html = (Source fromURL "http://www.clueda.com").getLines mkString ""!
!
val urlExtractor = """hrefs?=s?"([^"]+)"""".r!
!
for ( urlExtractor(url) <- urlExtractor findAllIn html ) { !
! println(s"Url -> $url")!
}

ﬁrst-order-functions / anonymous functions
‣ functions have a type like Integer or String

‣ They can be arguments to function and passed around as values
val y = (x: Int) => x * x!
//y: (Int) => Int = !
!
y apply 5!
// Int = 25!
!
y(5)!
// Int = 25!
!
!
val add = (x: Int, y: Int) => x + y!
// add: (Int, Int) => Int = !
!
add(1,2)!
// Int = 3

Implicits
‣ are introduced using the keyword implicit

‣ trigger an automatic transformation

‣ not stackable

‣ shorter, more readable

‣ may introduce „magic“

‣ Pimp my library Pattern: Locally scopeﬁed monkey patching

Implicits: Example
def toXML = !
<book>!
<pages>{pages}</pages>!
<year>{year}</year>!
</book>!
!
implicit def intToString(in : Int) : String = in.toString!
}
‣ no more need to manually transform year to string when using xml

‣ will also work for all other integers in scope of Book

Parallel Collections
‣ Asynchronous, parallel processing to take advantage of multicore processors

‣ .par transforms a Collection to it’s parallel counterpart

‣ .seq transforms a parallel Collection to a sequential one

‣ Parallel is implemented as a trait => can be used to create own par
collections

‣ Also works for Map

Parallel Collections - Example
// Sequential!
(1 to 10) foreach println
// Parallel!
(1 to 10).par foreach println

Parallele Collections - Examples II
// Unordered!
val tenTimes = (1 to 10).par map (_ * 10) !
tenTimes foreach println
// Unordered!
val tenTimes = (1 to 10).par map (_ * 10) !
tenTimes foreach println!
!
//Ordered!
//.seq transforms a parallel collection to a sequential one!
tenTimes.seq foreach println
scala> tenTimes.seq foreach println
10
20
30
40
50
60
70
80
90
100
scala> tenTimes foreach println
10
80
90
60
30
70
100
20
40
50

Build your own control structures
‣ Curriﬁed functions can be used to build control structures
object ControlStructures {!
def unless( test: => Boolean)(action: => Any) = !
if (! test) action!
!
def times( n: Int )(action: => Unit) {!
(1 to n) foreach { _ => action}!
}!
}
scala> import ControlStructures._!
//import ControlStructures._!
!
scala> times(2) { println("Hoorray :)")}!
Hoorray :)!
Hoorray :)!
!
scala> unless (5 < 10) { println("Math stopped working.") }!
// Any = ()!
!
scala> val ifNot = unless (2 + 2 != 4) { "Math still works." }!
// Any = Math still works.!

Scala - Patterns
‣ Structural Typing

‣ Pimp-My-Library-Pattern

Structural Typing
‣ Classed are described by methods and return types they provide

‣ Works like duck typing but the checking happens in compile time, not run
time
class Cowboy { def shout = "Yehaaw !" }!
class Pirate { def shout = "Arrrgh !" }!
!
def sayHelloTo( person : { def shout: String} ) = !
! s"Me : Hello!n $person shouts ${person.shout}"
val johnWayne = new Cowboy
!
sayHelloTo(johnWayne)
scala> sayHelloTo(johnWayne)
res4: String =
Me : Hello!
Cowboy@185f8f75 shouts Yehaaw !
val guybrush = new Pirate
!
sayHelloTo(guybrush)
scala> sayHelloTo(guybrush)
res5: String =
Me : Hello!
Pirate@29c356d3 shouts Arrrgh !

Pimp-My-Library-Pattern
‣ Add new functions to existing libraries without changing the code

‣ Like monkey patching

‣ type safe

‣ scoped

Pimp-My-Library-Pattern : Example Source
object PimpString {!
!
class WeatherString(s: String) {!
def ☀ = { println(s"$s sunny!") }!
def ☁ = "Dont't forget your ☂!“!
}!
!
implicit class ♔(name : String) {!
def hail = s"Hail to king $name"!
}!
!
implicit def pimpString(in: String) : WeatherString = !
! new WeatherString(in)!
}!

Pimp-My-Library-Pattern : Example Usage
scala> import PimpString._
import PimpString._
!
scala> "Monday is" ☀
Monday is sunny!
!
scala> "???".☁
res8: String = Dont't forget your ☂
scala> val anotherKing = ♔("Louis")
anotherKing: PimpString.♔ = PimpString$$u2654@12359094
!
scala> val aKing = implicitly[♔]("George")
aKing: PimpString.♔ = PimpString$$u2654@5081371
!
scala> aKing.hail
res10: String = Hail to king George
‣ Use with caution !
scala> val guys = List("James", "Louis", "Franz-Ferdinand")
guys: List[String] = List(James, Louis, Franz-Ferdinand)
!
scala> guys map (_.hail)
res13: List[String] = List(Hail to king James, Hail to king Louis, Hail to king
Franz-Ferdinand)

Scala - imperative, object oriented, functional

Scala -imperative, object oriented, functional -
Rules of the thumb
‣ functional if possible

‣ Sometimes imperative is better and faster

‣ start out with val and immutable collections,switch to var or mutable
collections if needed

‣ Use object orientation to encapsulate side eﬀects and imperative code

Advantage of the functional approach
‣ short

‣ no side eﬀect -> easier to reason about

‣ composable

Advantage of the imperative approach
‣ familar

‣ Eventually everything will be iterative after being translated to machine code

Imperative vs. functional, Examples
Imperative Functional
var i = 0!
while (i < args.length) {!
if ( i != 0 )!
print(" ")!
print( args(i) )!
i += 1!
}!
println()!
var x = 1!
var sum = 0!
while (x <= 9999) {!
sum += x!
x += 1!
}!
(1 to 9999) foldLeft(0)(_ + _)
(1 to 9999) sum
println( !
args reduceOption ( (acc,arg ) => !
acc + " " + arg!
)!
)!
println( args mkString " " )

Imperative vs. functional, Examples 2
Imperative Functional
var i = null!
var data = gettingData()!
!
if (data != null && data.size > 0) !
! i = data(0)!
else !
! i = 42! val i = !
! gettingData().headOption getOrElse 42
val i = !
! if (data != null && data.size > 0)!
! ! data(0)!
! else!
! ! 42

Literatur:
‣ Wampler, D., & Payne, A. (2009). Programming Scala. Sebastopol, CA:
O'Reilly.

‣ Odersky, M., Spoon, L., & Venners, B. (2008). Programming in Scala.
Mountain View, Calif: Artima.

‣ Malayeri, M. “Pimp My Library” Is An Aﬀront To Pimpers Of The World,
Everywhere

‣ http://www.scala-lang.org

‣ http://www.an-it.com

Introduction to Scala : Clueda

Recommended

More Related Content

What's hot (20)

Similar to Introduction to Scala : Clueda (20)

More from Andreas Neumann (11)

Recently uploaded (20)

Introduction to Scala : Clueda