Existential types in Scala - scala

Please find below a short example which puzzles me.
I must concede that I have some difficulties to manipulate existential types in Scala.
How should I solve the type mismatch line 56 ?
proposer is OK type _$1 while proposers is of type _$1 <: Individual
Thanks in advance,
Maxime.
class Individual(n: String) {
protected val name=n
var preferred: Individual = this
override def toString(): String=name
}
class Man(n: String) extends Individual(n) { }
class Woman(n: String) extends Individual(n) { }
class Marriage(m: Man, w: Woman){
private val man=m
private val woman=w
def this(w: Woman, m: Man) = this(m,w)
override def toString(): String = man+"--"+woman
}
class Matching(){
private var list: List[Marriage] = Nil
def add(m: Marriage): Unit = { list = m ::list }
override def toString(): String= {
var s: String = ""
for (elm<-list) s=s+elm+" "
return s
}
}
object Test{
protected var male = true
def main(args: Array[String]): Unit = {
val al = new Man("Al")
val bob = new Man("Bob")
val alice = new Woman("Alice")
val barbara = new Woman("Barbara")
al.preferred = alice
bob.preferred = barbara
alice.preferred = bob
barbara.preferred = al
val men = Set(al, bob)
val women = Set(alice, barbara)
val m = new Matching()
//var proposers=women
var proposers: Set[_ <:Individual] = Set[Individual]()
if (male) proposers = men
else proposers = women
while (!proposers.isEmpty) {
for(proposer <- proposers) {
val proposer=proposers.head
if (proposer.isInstanceOf[Man])
m.add(new Marriage(
proposer.asInstanceOf[Man],
proposer.preferred.asInstanceOf[Woman]
))
else
m.add(new Marriage(
proposer.asInstanceOf[Woman],
proposer.preferred.asInstanceOf[Man]
))
proposers-=proposer//There is an error here
}
}
println(m)
}
}

This code is messy. It's poorly formatted, it mixes tabs and spaces, and it uses mutability even in the most trivial of places where a functional solution requires little thought.
This code also won't scale internationally to countries where same-sex marriage is a possibility.
Working from the top down...
I suspect you'll never want to directly instantiate an Individual, only ever a Man or a Woman. So a algebraic data type makes more sense, this is done with a sealed trait and case class subtypes.
I'll also drop the preferred property, as it can lead to circular references. Dealing with this in immutable data is beyond the level I'm willing to go in this answer.
sealed trait Individual {
def name: String
override def toString(): String=name
}
//as it's a case class, `name` becomes a val,
//which implements the abstract `def name` from the trait
case class Man(name: String) extends Individual
case class Woman(name: String) extends Individual
Marriage can also be a case class, and let's drop the clumsy duplication of class parameters into vals - it's just pointless boilerplate. This is also a good time to move the auxiliary constructor to a factory method in the companion object:
case class Marriage(man: Man, woman: Woman) {
override def toString(): String = man + "--" + woman
}
object Marriage {
def apply(w: Woman, m: Man) = new Marriage(m,w)
}
Matching is almost pointless, an entire class just to wrap a List? This kind of thing made sense in pre-Generics Java, but not any more. I'll keep it anyway (for now) so I can fix up that toString implementation, which is painfully mutable and uses return for no good reason:
case class Matching(){
private var list: List[Marriage] = Nil
def add(m: Marriage): Unit = { list ::= m }
override def toString() = list.mkString(" ")
}
Finally, the "meat" of the problem. Comments are inline, but you'll note that I don't need (or use) Matching. It's replaced in its entirety by the final println
object Test{
//better name, and a val (because it never changes)
protected val menPropose = true
def main(args: Array[String]): Unit = {
// `new` not required for case classes
val al = Man("Al")
val bob = Man("Bob")
val alice = Woman("Alice")
val barbara = Woman("Barbara")
// remember how preference was removed from `Individual`?
val mprefs = Map( al -> alice, bob -> barbara )
val fprefs = Map( alice -> bob, barbara -> al )
val men = Set(al, bob)
val women = Set(alice, barbara)
// nicely immutable, and using the returned value from if/else
val proposers = if (menPropose) men else women
// no while loop, name shadowing, or mutability.
// just a simple for-comprehension
val marriages = for(proposer <- proposers) yield {
//pattern-matching beats `isInstanceOf`... every time
proposer match {
case m: Man => Marriage(m, mprefs(m))
case f: Woman => Marriage(f, fprefs(f))
}
}
println(marriages mkString " ")
}
}
There's more that can be done here, way more. What of same-sex relationships? What if two or more people share the same preference? What if someone has no preference?
I could also encode the type of someone's preference into Individual instances. But that's getting a bit more advanced.

Related

DSL in scala using case classes

My use case has case classes something like
case class Address(name:String,pincode:String){
override def toString =name +"=" +pincode
}
case class Department(name:String){
override def toString =name
}
case class emp(address:Address,department:Department)
I want to create a DSL like below.Can anyone share the links about how to create a DSL and any suggestions to achieve the below.
emp.withAddress("abc","12222").withDepartment("HR")
Update:
Actual use case class may have more fields close to 20. I want to avoid redudancy of code
I created a DSL using reflection so that we don't need to add every field to it.
Disclamer: This DSL is extremely weakly typed and I did it just for fun. I don't really think this is a good approach in Scala.
scala> create an Employee where "homeAddress" is Address("a", "b") and "department" is Department("c") and that_s it
res0: Employee = Employee(a=b,null,c)
scala> create an Employee where "workAddress" is Address("w", "x") and "homeAddress" is Address("y", "z") and that_s it
res1: Employee = Employee(y=z,w=x,null)
scala> create a Customer where "address" is Address("a", "b") and "age" is 900 and that_s it
res0: Customer = Customer(a=b,900)
The last example is the equivalent of writing:
create.a(Customer).where("address").is(Address("a", "b")).and("age").is(900).and(that_s).it
A way of writing DSLs in Scala and avoid parentheses and the dot is by following this pattern:
object.method(parameter).method(parameter)...
Here is the source:
// DSL
object create {
def an(t: Employee.type) = new ModelDSL(Employee(null, null, null))
def a(t: Customer.type) = new ModelDSL(Customer(null, 0))
}
object that_s
class ModelDSL[T](model: T) {
def where(field: String): ValueDSL[ModelDSL2[T], Any] = new ValueDSL(value => {
val f = model.getClass.getDeclaredField(field)
f.setAccessible(true)
f.set(model, value)
new ModelDSL2[T](model)
})
def and(t: that_s.type) = new { def it = model }
}
class ModelDSL2[T](model: T) {
def and(field: String) = new ModelDSL(model).where(field)
def and(t: that_s.type) = new { def it = model }
}
class ValueDSL[T, V](callback: V => T) {
def is(value: V): T = callback(value)
}
// Models
case class Employee(homeAddress: Address, workAddress: Address, department: Department)
case class Customer(address: Address, age: Int)
case class Address(name: String, pincode: String) {
override def toString = name + "=" + pincode
}
case class Department(name: String) {
override def toString = name
}
I really don't think you need the builder pattern in Scala. Just give your case class reasonable defaults and use the copy method.
i.e.:
employee.copy(address = Address("abc","12222"),
department = Department("HR"))
You could also use an immutable builder:
case class EmployeeBuilder(address:Address = Address("", ""),department:Department = Department("")) {
def build = emp(address, department)
def withAddress(address: Address) = copy(address = address)
def withDepartment(department: Department) = copy(department = department)
}
object EmployeeBuilder {
def withAddress(address: Address) = EmployeeBuilder().copy(address = address)
def withDepartment(department: Department) = EmployeeBuilder().copy(department = department)
}
You could do
object emp {
def builder = new Builder(None, None)
case class Builder(address: Option[Address], department: Option[Department]) {
def withDepartment(name:String) = {
val dept = Department(name)
this.copy(department = Some(dept))
}
def withAddress(name:String, pincode:String) = {
val addr = Address(name, pincode)
this.copy(address = Some(addr))
}
def build = (address, department) match {
case (Some(a), Some(d)) => new emp(a, d)
case (None, _) => throw new IllegalStateException("Address not provided")
case _ => throw new IllegalStateException("Department not provided")
}
}
}
and use it as emp.builder.withAddress("abc","12222").withDepartment("HR").build().
You don't need optional fields, copy, or the builder pattern (exactly), if you are willing to have the build always take the arguments in a particular order:
case class emp(address:Address,department:Department, id: Long)
object emp {
def withAddress(name: String, pincode: String): WithDepartment =
new WithDepartment(Address(name, pincode))
final class WithDepartment(private val address: Address)
extends AnyVal {
def withDepartment(name: String): WithId =
new WithId(address, Department(name))
}
final class WithId(address: Address, department: Department) {
def withId(id: Long): emp = emp(address, department, id)
}
}
emp.withAddress("abc","12222").withDepartment("HR").withId(1)
The idea here is that each emp parameter gets its own class which provides a method to get you to the next class, until the final one gives you an emp object. It's like currying but at the type level. As you can see I've added an extra parameter just as an example of how to extend the pattern past the first two parameters.
The nice thing about this approach is that, even if you're part-way through the build, the type you have so far will guide you to the next step. So if you have a WithDepartment so far, you know that the next argument you need to supply is a department name.
If you want to avoid modifying the origin classes you can use implicit class, e.g.
implicit class EmpExtensions(emp: emp) {
def withAddress(name: String, pincode: String) {
//code omitted
}
// code omitted
}
then import EmpExtensions wherever you need these methods

Scala reflect string to singleton object

I'm looking for a way to convert a Scala singleton object given as a string (for example: package1.Main) to the actual instance of Main, so that I can invoke methods on it.
Example of the problem:
package x {
object Main extends App {
val objectPath: String = io.StdIn.readLine("Give an object: ") // user enters: x.B
// how to convert the objectPath (String) to a variable that references singleton B?
val b1: A = magicallyConvert1(objectPath)
b1.hi()
val b2: B.type = magicallyConvert2(objectPath)
b2.extra()
}
trait A {
def hi() = {}
}
object B extends A {
def extra() = {}
}
}
How can the magicallyConvert1 and magicallyConvert2 functions be implemented?
For a normal class, this can be done using something like:
val b: A = Class.forName("x.B").newInstance().asInstanceOf[A]
But I found a solution for singletons, using Java reflections:
A singleton is accesible in Java under the name:
package.SingletonName$.MODULE$
So you have to append "$.MODULE$", which is a static field.
So we can use standard Java reflections to get it.
So the solution is:
def magicallyConvert1(objectPath: String) = {
val clz = Class.forName(objectPath + "$")
val field = clz.getField("MODULE$")
val b: A = field.get(null).asInstanceOf[A]
b
}
def magicallyConvert2(objectPath: String) = {
val clz = Class.forName(objectPath + "$")
val field = clz.getField("MODULE$")
val b: B.type = field.get(null).asInstanceOf[B.type]
b
}
But it would be interesting to still see a solution with Scala-Reflect en Scala-Meta.
take a look at scalameta http://scalameta.org it does what you want and more

Extending instantiated objects by importing: Bad Idea?

I have list of many City objects, and at some point after their creation, they will be labelled, so a Label field will be set for each one.
I want to do this in a functional, stateless way, i.e. without setting a default and using var. I'm really just curious as to what the best practice is.
I've got this working example, but I feel pretty funny about it. Notice, in particular, that I'm import-ing the existing City objects, which happens when I call City.addLabel(label).
case class LabelledCity(importCity: CityLike, label: Label) extends CityLike with Labelled
{
val cityState = importCity.cityState
val income = importCity.income
}
case class City(string: String) extends CityLike {
val split = string.split(" --- ")
val cityState = split(0)
val income = split(1).toInt
}
abstract class CityLike {
def addLabel(l: Label): LabelledCity = { new LabelledCity(this, l) }
val cityState: String
val income: Int
}
This all feels a little wrong, or like I'm missing something simple... am I?
Here is the immutable way to archive this (I have dropped class bodies for a sake of simplicity):
case class City(string: String, label: Option[String] = None)
val unknown = City("somestring")
// unknown: City = City(somestring,None)
val washington = unknown.copy(label = Some("Washington"))
// washington: City = City(somestring,Some(Washington))

How to initialize a trait variable that is a val

I have MyObject and MyTrait:
class MyObject(private val myname: String = "") extends MyTrait {
_name = myname
def foo(myname : String) {
_name = myname
}
}
trait MyTrait {
protected var _name: String = _
def name = _name
}
This works fine as this
val myObject = new MyObject("abc")
println(myObject.name)
myObject.foo("def")
println(myObject.name)
prints
abc
def
as expected.
Problem now is that I want MyTrait._name to be a val instead of a var. But there is no way I can manage to get this to compile. Any hints appreciated.
Regards, Oliver
Here is an answer that uses the very latest cutting-edge naming conventions from Rex Kerr and Martin Odersky!
Read it on the scala-debate list. And you thought they sit around working on "higher kinds" and computing with unboxed ints.
There is a PR for the style changes, but this convention will have to wait a bit.
Doc Martin says: That does look promising. I have to experiment with it a little.
So be careful with this stuff; it's experimental and probably chemically unstable.
class MyObject(override protected val initialName: String = "") extends MyTrait {
private var myName: String = initialName
def name_=(newName: String) {
myName = newName
}
override def name = myName
}
trait MyTrait {
protected val initialName: String = "default"
def name = initialName
}
object Test extends App {
val myObject = new MyObject("abc")
println(myObject.name)
myObject.name = "def"
println(myObject.name)
}
The style guide has a section on brevity but is itself not brief. I'm sure there are answers on SO about "prefer def over val in traits" by Daniel Sobral. And don't forget to consult the one-question FAQ when you encounter init-order problems.

Mixing in a trait dynamically

Having a trait
trait Persisted {
def id: Long
}
how do I implement a method that accepts an instance of any case class and returns its copy with the trait mixed in?
The signature of the method looks like:
def toPersisted[T](instance: T, id: Long): T with Persisted
This can be done with macros (that are officially a part of Scala since 2.10.0-M3). Here's a gist example of what you are looking for.
1) My macro generates a local class that inherits from the provided case class and Persisted, much like new T with Persisted would do. Then it caches its argument (to prevent multiple evaluations) and creates an instance of the created class.
2) How did I know what trees to generate? I have a simple app, parse.exe that prints the AST that results from parsing input code. So I just invoked parse class Person$Persisted1(first: String, last: String) extends Person(first, last) with Persisted, noted the output and reproduced it in my macro. parse.exe is a wrapper for scalac -Xprint:parser -Yshow-trees -Ystop-after:parser. There are different ways to explore ASTs, read more in "Metaprogramming in Scala 2.10".
3) Macro expansions can be sanity-checked if you provide -Ymacro-debug-lite as an argument to scalac. In that case all expansions will be printed out, and you'll be able to detect codegen errors faster.
edit. Updated the example for 2.10.0-M7
It is not possible to achieve what you want using vanilla scala. The problem is that the mixins like the following:
scala> class Foo
defined class Foo
scala> trait Bar
defined trait Bar
scala> val fooWithBar = new Foo with Bar
fooWithBar: Foo with Bar = $anon$1#10ef717
create a Foo with Bar mixed in, but it is not done at runtime. The compiler simply generates a new anonymous class:
scala> fooWithBar.getClass
res3: java.lang.Class[_ <: Foo] = class $anon$1
See Dynamic mixin in Scala - is it possible? for more info.
What you are trying to do is known as record concatenation, something that Scala's type system does not support. (Fwiw, there exist type systems - such as this and this - that provide this feature.)
I think type classes might fit your use case, but I cannot tell for sure as the question doesn't provide sufficient information on what problem you are trying to solve.
Update
You can find an up to date working solution, which utilizes a Toolboxes API of Scala 2.10.0-RC1 as part of SORM project.
The following solution is based on the Scala 2.10.0-M3 reflection API and Scala Interpreter. It dynamically creates and caches classes inheriting from the original case classes with the trait mixed in. Thanks to caching at maximum this solution should dynamically create only one class for each original case class and reuse it later.
Since the new reflection API isn't that much disclosed nor is it stable and there are no tutorials on it yet this solution may involve some stupid repitative actions and quirks.
The following code was tested with Scala 2.10.0-M3.
1. Persisted.scala
The trait to be mixed in. Please note that I've changed it a bit due to updates in my program
trait Persisted {
def key: String
}
2. PersistedEnabler.scala
The actual worker object
import tools.nsc.interpreter.IMain
import tools.nsc._
import reflect.mirror._
object PersistedEnabler {
def toPersisted[T <: AnyRef](instance: T, key: String)
(implicit instanceTag: TypeTag[T]): T with Persisted = {
val args = {
val valuesMap = propertyValuesMap(instance)
key ::
methodParams(constructors(instanceTag.tpe).head.typeSignature)
.map(_.name.decoded.trim)
.map(valuesMap(_))
}
persistedClass(instanceTag)
.getConstructors.head
.newInstance(args.asInstanceOf[List[Object]]: _*)
.asInstanceOf[T with Persisted]
}
private val persistedClassCache =
collection.mutable.Map[TypeTag[_], Class[_]]()
private def persistedClass[T](tag: TypeTag[T]): Class[T with Persisted] = {
if (persistedClassCache.contains(tag))
persistedClassCache(tag).asInstanceOf[Class[T with Persisted]]
else {
val name = generateName()
val code = {
val sourceParams =
methodParams(constructors(tag.tpe).head.typeSignature)
val newParamsList = {
def paramDeclaration(s: Symbol): String =
s.name.decoded + ": " + s.typeSignature.toString
"val key: String" :: sourceParams.map(paramDeclaration) mkString ", "
}
val sourceParamsList =
sourceParams.map(_.name.decoded).mkString(", ")
val copyMethodParamsList =
sourceParams.map(s => s.name.decoded + ": " + s.typeSignature.toString + " = " + s.name.decoded).mkString(", ")
val copyInstantiationParamsList =
"key" :: sourceParams.map(_.name.decoded) mkString ", "
"""
class """ + name + """(""" + newParamsList + """)
extends """ + tag.sym.fullName + """(""" + sourceParamsList + """)
with """ + typeTag[Persisted].sym.fullName + """ {
override def copy(""" + copyMethodParamsList + """) =
new """ + name + """(""" + copyInstantiationParamsList + """)
}
"""
}
interpreter.compileString(code)
val c =
interpreter.classLoader.findClass(name)
.asInstanceOf[Class[T with Persisted]]
interpreter.reset()
persistedClassCache(tag) = c
c
}
}
private lazy val interpreter = {
val settings = new Settings()
settings.usejavacp.value = true
new IMain(settings, new NewLinePrintWriter(new ConsoleWriter, true))
}
private var generateNameCounter = 0l
private def generateName() = synchronized {
generateNameCounter += 1
"PersistedAnonymous" + generateNameCounter.toString
}
// REFLECTION HELPERS
private def propertyNames(t: Type) =
t.members.filter(m => !m.isMethod && m.isTerm).map(_.name.decoded.trim)
private def propertyValuesMap[T <: AnyRef](instance: T) = {
val t = typeOfInstance(instance)
propertyNames(t)
.map(n => n -> invoke(instance, t.member(newTermName(n)))())
.toMap
}
private type MethodType = {def params: List[Symbol]; def resultType: Type}
private def methodParams(t: Type): List[Symbol] =
t.asInstanceOf[MethodType].params
private def methodResultType(t: Type): Type =
t.asInstanceOf[MethodType].resultType
private def constructors(t: Type): Iterable[Symbol] =
t.members.filter(_.kind == "constructor")
private def fullyQualifiedName(s: Symbol): String = {
def symbolsTree(s: Symbol): List[Symbol] =
if (s.enclosingTopLevelClass != s)
s :: symbolsTree(s.enclosingTopLevelClass)
else if (s.enclosingPackageClass != s)
s :: symbolsTree(s.enclosingPackageClass)
else
Nil
symbolsTree(s)
.reverseMap(_.name.decoded)
.drop(1)
.mkString(".")
}
}
3. Sandbox.scala
The test app
import PersistedEnabler._
object Sandbox extends App {
case class Artist(name: String, genres: Set[Genre])
case class Genre(name: String)
val artist = Artist("Nirvana", Set(Genre("rock"), Genre("grunge")))
val persisted = toPersisted(artist, "some-key")
assert(persisted.isInstanceOf[Persisted])
assert(persisted.isInstanceOf[Artist])
assert(persisted.key == "some-key")
assert(persisted.name == "Nirvana")
assert(persisted == artist) // an interesting and useful effect
val copy = persisted.copy(name = "Puddle of Mudd")
assert(copy.isInstanceOf[Persisted])
assert(copy.isInstanceOf[Artist])
// the only problem: compiler thinks that `copy` does not implement `Persisted`, so to access `key` we have to specify it manually:
assert(copy.asInstanceOf[Artist with Persisted].key == "some-key")
assert(copy.name == "Puddle of Mudd")
assert(copy != persisted)
}
While it's not possible to compose an object AFTER it's creation, you can have very wide tests to determine if the object is of a specific composition using type aliases and definition structs:
type Persisted = { def id: Long }
class Person {
def id: Long = 5
def name = "dude"
}
def persist(obj: Persisted) = {
obj.id
}
persist(new Person)
Any object with a def id:Long will qualify as Persisted.
Achieving what I THINK you are trying to do is possible with implicit conversions:
object Persistable {
type Compatible = { def id: Long }
implicit def obj2persistable(obj: Compatible) = new Persistable(obj)
}
class Persistable(val obj: Persistable.Compatible) {
def persist() = println("Persisting: " + obj.id)
}
import Persistable.obj2persistable
new Person().persist()