Either

Value Members

1. final def !=(arg0: Any): Boolean

   

   Test two objects for inequality.

   Test two objects for inequality.

   returns

   true if !(this == that), false otherwise.

   Definition Classes

   AnyRef → Any

2. final def ##(): Int

   

   Equivalent to x.hashCode except for boxed numeric types and null.

   Equivalent to x.hashCode except for boxed numeric types and null. For numerics, it returns a hash value which is consistent with value equality: if two value type instances compare as true, then ## will produce the same hash value for each of them. For null returns a hashcode where null.hashCode throws a NullPointerException.

   returns

   a hash value consistent with ==

   Definition Classes

   AnyRef → Any

3. final def ==(arg0: Any): Boolean

   

   The expression x == that is equivalent to if (x eq null) that eq null else x.equals(that).

   The expression x == that is equivalent to if (x eq null) that eq null else x.equals(that).

   arg0

   the object to compare against this object for equality.

   returns

   true if the receiver object is equivalent to the argument; false otherwise.

   Definition Classes

   AnyRef → Any

4. object LeftBias

   

   This object contains utilities for defining an environment in which Either instances are "left-biased".

   This object contains utilities for defining an environment in which Either instances are "left-biased".

   Left-biased means instances of Either[A,B] can be operated upon via methods that render thme quite analogous to an Option[A], except that failures are represented by a Right containing information about the problem rathar than by uninformative None. In particular, a left-biased Either can be used directly in a for comprehension. Unlike a LeftProjection, a left-biased Either supports all features of a for comprehension, including pattern matching, filtering, and variable assignment.

   Left-biasing decorates vanilla Either[A,B] with the following API.

   def flatMap[BB >: B, Z]( f : A => Either[Z,BB] ) : Either[Z,BB]
   def map[Z]( f : A => Z )                         : Either[Z,B]
   def withFilter( p : A => Boolean )               : Either[A,B]
   def exists( f : A => Boolean )                   : Boolean
   def forall( f : A => Boolean )                   : Boolean
   def foreach[U]( f : A => U )                     : Any
   def get                                          : A
   def getOrElse[AA >: A ]( or : =>AA )             : AA
   def toOption                                     : Option[A]
   def toSeq                                        : collection.Seq[A]
   def xget                                         : B
   def xgetOrElse[BB>:B]( or : =>BB )               : BB
   def xmap[Z]( f : B => Z )                        : Either[A,Z]
   def replaceIfEmpty[BB>:B]( replacement : =>BB )  : Either[A,BB]
   def isEmpty                                      : Boolean
   def isLeftBiased                                 : Boolean
   def isRightBiased                                : Boolean
   def conformsToBias                               : Boolean

   For the full documentation, plase see Ops.

   To enable left-biasing, you first define a bias, typically via the Either.LeftBias.withEmptyToken factory.

   // will impart a left-bias suitable to Either[A,String]
   val LeftBias = Either.LeftBias.withEmptyToken("EMPTY")
   import LeftBias._

   The specified "empty token" will define the value in the Right object that will be used to indicate when a filter or pattern-match of a left-side object fails.

   If you don't wish to specify an empty token, you may use the more convenient

   import Either.LeftBias._

   However, if you do not specify an empty token, filter operations or pattern match failures that would leave the left-side empty will result in the raising of a java.util.NoSuchElementException.

   You may also create a left-biased environment within a trait, class, object, or package by having the entity extend the trait LeftBias or the abstract base class LeftBias.Base.

   A Detailed example

   Consider a hypothetical process that must read some binary data from a URL, parse the data into a Record, ensure that record's year is no older than 2011, and then operate upon the data via one of several external services, depending on the opCode, ultimately producing a Result. Things that can go wrong include network errors on loading, parsing problems, failure of the timestamp condition, and failures with the external service.

   With Option, one might define a process like:

   case class Record( opCode : Int, data : Seq[Byte], year : Int )
   trait Result
   
   def loadURL( url : String )                   : Option[Seq[Byte]] = ???
   def parseRecord( bytes : Seq[Byte] )          : Option[Record] = ???
   def process( opCode : Int, data : Seq[Byte] ) : Option[Result] = ???
   
   def processURL( url : String ) : Option[Result] = {
     for {
       rawBytes                     <- loadURL( url )
       Record( opCode, data, year ) <- parseRecord( rawBytes )
       result                       <- process( opCode, data ) if year >= 2011
     } yield result
   }

   That's fine, but if anything goes wrong, clients will receive a value None, and have no specific information about what the problem was. Instead of option, we might try to use Either, defining error codes for things that can go wrong. Unconventionally (since we are considering left-biased Either here), we'll let good values arrive as Left values and embed our error codes in Right objects.

   object Error {
     case object Empty extends Error;
     case class IO( message : String, stackTrace : List[StackTraceElement] = Nil ) extends Error;
     case class NoParse( message : String, stackTrace : List[StackTraceElement] = Nil ) extends Error;
     case object BadYear extends Error;
     case class ProcessorFailure( message : String, stackTrace : List[StackTraceElement] = Nil ) extends Error;
   }
   sealed trait Error;
   
   case class Record( opCode : Int, data : Seq[Byte], year : Int )
   trait Result
   
   def loadURL( url : String )                   : Either[Seq[Byte],Error] = ??? // fails w/Error.IO
   def parseRecord( bytes : Seq[Byte] )          : Either[Record,Error]    = ??? // fails w/Error.NoParse
   def process( opCode : Int, data : Seq[Byte] ) : Either[Result,Error]    = ??? // fails w/Error.ProcessorFailure
   
   // we try to use the left projection but the
   // for comprehension fails to compile
   def processURL( url : String ) : Either[Result,Error] = {
     for {
       rawBytes                     <- loadURL( url ).left
       Record( opCode, data, year ) <- parseRecord( rawBytes ).left
       result                       <- process( opCode, data ).left if year >= 2011
     } yield result
   }

   Unfortunately, this fails to compile, because scala.util.Either.LeftProjection does not support pattern-matching (used to extract opCode, data, and year) or filtering (which we perform based on year).

   Instead, we can left-bias the Either:

   object Error {
     case object Empty extends Error;
     case class IO( message : String, stackTrace : List[StackTraceElement] = Nil ) extends Error;
     case class NoParse( message : String, stackTrace : List[StackTraceElement] = Nil ) extends Error;
     case object BadYear extends Error;
     case class ProcessorFailure( message : String, stackTrace : List[StackTraceElement] = Nil ) extends Error;
   }
   sealed trait Error;
   
   // left-bias Either
   val LeftBias = Either.LeftBias.withEmptyToken( Error.Empty )
   import LeftBias._
   
   case class Record( opCode : Int, data : Seq[Byte], year : Int )
   trait Result
   
   def loadURL( url : String )                   : Either[Seq[Byte],Error] = ??? // fails w/Error.IO
   def parseRecord( bytes : Seq[Byte] )          : Either[Record,Error]    = ??? // fails w/Error.NoParse
   def process( opCode : Int, data : Seq[Byte] ) : Either[Result,Error]    = ??? // fails w/Error.ProcessorFailure
   
   // use Either directly, rather than a projection, in the for comprehension
   def processURL( url : String ) : Either[Result,Error] = {
     for {
       rawBytes                     <- loadURL( url )
       Record( opCode, data, year ) <- parseRecord( rawBytes )
       result                       <- process( opCode, data ) if year >= 2011
     } yield result
   }

   There is a problem: Error.Empty, would be the result of the year filter failing, when it should be Error.BadYear.

   The most general and straightforward wat to fix this is to replace empty tokens as they arrive with more informative errors. For this purpose, biased Either offers a method called LeftBias.Ops.replaceIfEmpty:

   // use Either directly, rather than a projection, in the for comprehension
   def processURL( url : String ) : Either[Result,Error] = {
     val processed = {
       for {
         rawBytes                     <- loadURL( url )
         Record( opCode, data, year ) <- parseRecord( rawBytes )
         result                       <- process( opCode, data ) if year >= 2011
       } yield result
     }
     processed.replaceIfEmpty( Error.BadYear )
   }

   The scope of the left-bias is the scope of the import. An alternative way to ensure empty tokens yield meaningful information is to define separate biases for separate contexts.

   val ProcessURLBias       = Either.LeftBias.withEmptyToken( Error.BadYear )
   val EnableHyperdriveBias = Either.LeftBias.withEmptyToken( Error.SpaceTimeRupture )
   ...
   
   def processURL( url : String ) : Either[Result,Int] = {
     import ProcessURLBias._
     for {
       rawBytes                     <- loadURL( url )
       Record( opCode, data, year ) <- parseRecord( rawBytes )
       result                       <- process( opCode, data ) if year >= 2011
     } yield result
   }

5. object RightBias

   

   This object contains utilities for defining an environment in which Either instances are "right-biased".

   This object contains utilities for defining an environment in which Either instances are "right-biased".

   Right-biased means instances of Either[A,B] can be operated upon via methods that render thme quite analogous to an Option[A], except that failures are represented by a Left containing information about the problem rathar than by uninformative None. In particular, a right-biased Either can be used directly in a for comprehension. Unlike a RightProjection, a right-biased Either supports all features of a for comprehension, including pattern matching, filtering, and variable assignment.

   Right-biasing decorates vanilla Either[A,B] with the following API.

   def flatMap[AA >: A, Z]( f : B => Either[AA,Z] ) : Either[AA,Z]
   def map[Z]( f : B => Z )                         : Either[A,Z]
   def withFilter( p : B => Boolean )               : Either[A,B]
   def exists( f : B => Boolean )                   : Boolean
   def forall( f : B => Boolean )                   : Boolean
   def foreach[U]( f : B => U )                     : Any
   def get                                          : B
   def getOrElse[BB>:B]( or : =>BB )                : BB
   def toOption                                     : Option[B]
   def toSeq                                        : collection.Seq[B]
   def xget                                         : A
   def xgetOrElse[AA>:A]( or : =>AA )               : AA
   def xmap[Z]( f : A => Z )                        : Either[Z,B]
   def replaceIfEmpty[AA>:A]( replacement : =>AA )  : Either[AA,B]
   def isEmpty                                      : Boolean
   def isLeftBiased                                 : Boolean
   def isRightBiased                                : Boolean
   def conformsToBias                               : Boolean

   For the full documentation, plase see Ops.

   To enable right-biasing, you first define a bias, typically via the Either.RightBias.withEmptyToken factory.

   // will impart a right-bias suitable to Either[String,B]
   val RightBias = Either.RightBias.withEmptyToken("EMPTY")
   import RightBias._

   The specified "empty token" will define the value in the Left object that will be used to indicate when a filter or pattern-match of a right-side object fails.

   If you don't wish to specify an empty token, you may use the more convenient

   import Either.RightBias._

   However, if you do not specify an empty token, filter operations or pattern match failures that would leave the right-side empty will result in the raising of a java.util.NoSuchElementException.

   You may also create a right-biased environment within a trait, class, object, or package by having the entity extend the trait RightBias or the abstract base class RightBias.Base.

   A Detailed example

   Consider a hypothetical process that must read some binary data from a URL, parse the data into a Record, ensure that record's year is no older than 2011, and then operate upon the data via one of several external services, depending on the opCode, ultimately producing a Result. Things that can go wrong include network errors on loading, parsing problems, failure of the timestamp condition, and failures with the external service.

   With Option, one might define a process like:

   case class Record( opCode : Int, data : Seq[Byte], year : Int )
   trait Result
   
   def loadURL( url : String )                   : Option[Seq[Byte]] = ???
   def parseRecord( bytes : Seq[Byte] )          : Option[Record]    = ???
   def process( opCode : Int, data : Seq[Byte] ) : Option[Result]    = ???
   
   def processURL( url : String ) : Option[Result] = {
     for {
       rawBytes                     <- loadURL( url )
       Record( opCode, data, year ) <- parseRecord( rawBytes )
       result                       <- process( opCode, data ) if year >= 2011
     } yield result
   }

   That's fine, but if anything goes wrong, clients will receive a value None, and have no specific information about what the problem was. Instead of option, we might try to use Either, defining error codes for things that can go wrong. Conventionally, we'll let good values arrive as Right values and embed our error codes in Left objects.

   object Error {
     case object Empty extends Error;
     case class IO( message : String, stackTrace : List[StackTraceElement] = Nil ) extends Error;
     case class NoParse( message : String, stackTrace : List[StackTraceElement] = Nil ) extends Error;
     case object BadYear extends Error;
     case class ProcessorFailure( message : String, stackTrace : List[StackTraceElement] = Nil ) extends Error;
   }
   sealed trait Error;
   
   case class Record( opCode : Int, data : Seq[Byte], year : Int )
   trait Result
   
   def loadURL( url : String )                   : Either[Error,Seq[Byte]] = ??? // fails w/Error.IO
   def parseRecord( bytes : Seq[Byte] )          : Either[Error,Record]    = ??? // fails w/Error.NoParse
   def process( opCode : Int, data : Seq[Byte] ) : Either[Error,Result]    = ??? // fails w/Error.ProcessorFailure
   
   // we try to use the right projection but the
   // for comprehension fails to compile
   def processURL( url : String ) : Either[Error,Result] = {
     for {
       rawBytes                     <- loadURL( url ).right
       Record( opCode, data, year ) <- parseRecord( rawBytes ).right
       result                       <- process( opCode, data ).right if year >= 2011
     } yield result
   }

   Unfortunately, this fails to compile, because scala.util.Either.RightProjection does not support pattern-matching (used to extract opCode, data, and year) or filtering (which we perform based on year).

   Instead, we can right-bias the Either:

   object Error {
     case object Empty extends Error;
     case class IO( message : String, stackTrace : List[StackTraceElement] = Nil ) extends Error;
     case class NoParse( message : String, stackTrace : List[StackTraceElement] = Nil ) extends Error;
     case object BadYear extends Error;
     case class ProcessorFailure( message : String, stackTrace : List[StackTraceElement] = Nil ) extends Error;
   }
   sealed trait Error;
   
   // right-bias Either
   val RightBias = Either.RightBias.withEmptyToken( Error.Empty )
   import RightBias._
   
   case class Record( opCode : Int, data : Seq[Byte], year : Int )
   trait Result
   
   def loadURL( url : String )                   : Either[Error,Seq[Byte]] = ??? // fails w/Error.IO
   def parseRecord( bytes : Seq[Byte] )          : Either[Error,Record]    = ??? // fails w/Error.NoParse
   def process( opCode : Int, data : Seq[Byte] ) : Either[Error,Result]    = ??? // fails w/Error.ProcessorFailure
   
   // use Either directly, rather than a projection, in the for comprehension
   def processURL( url : String ) : Either[Error,Result] = {
     for {
       rawBytes                     <- loadURL( url )
       Record( opCode, data, year ) <- parseRecord( rawBytes )
       result                       <- process( opCode, data ) if year >= 2011
     } yield result
   }

   There is a problem: Error.Empty, would be the result of the year filter failing, when it should be Error.BadYear.

   The most general and straightforward wat to fix this is to replace empty tokens as they arrive with more informative errors. For this purpose, biased Either offers a method called RightBias.Ops.replaceIfEmpty:

   // use Either directly, rather than a projection, in the for comprehension
   def processURL( url : String ) : Either[Int,Result] = {
     val processed = {
       for {
         rawBytes                     <- loadURL( url )
         Record( opCode, data, year ) <- parseRecord( rawBytes )
         result                       <- process( opCode, data ) if year >= 2011
       } yield result
     }
     processed.replaceIfEmpty( Error.BadYear )
   }

   The scope of the right-bias is the scope of the import. An alternative way to ensure empty tokens yield meaningful information is to define separate biases for separate contexts.

   val ProcessURLBias       = Either.RightBias.withEmptyToken( Error.BadYear )
   val EnableHyperdriveBias = Either.RightBias.withEmptyToken( Error.SpaceTimeRupture )
   ...
   
   def processURL( url : String ) : Either[Int,Result] = {
     import ProcessURLBias._
     for {
       rawBytes                     <- loadURL( url )
       Record( opCode, data, year ) <- parseRecord( rawBytes )
       result                       <- process( opCode, data ) if year >= 2011
     } yield result
   }

6. final def asInstanceOf[T0]: T0

   

   Cast the receiver object to be of type T0.

   Cast the receiver object to be of type T0.

   Note that the success of a cast at runtime is modulo Scala's erasure semantics. Therefore the expression 1.asInstanceOf[String] will throw a ClassCastException at runtime, while the expression List(1).asInstanceOf[List[String]] will not. In the latter example, because the type argument is erased as part of compilation it is not possible to check whether the contents of the list are of the requested type.

   returns

   the receiver object.

   Definition Classes

   Any

   Exceptions thrown

   ClassCastException if the receiver object is not an instance of the erasure of type T0.

7. def clone(): AnyRef

   

   Create a copy of the receiver object.

   Create a copy of the receiver object.

   The default implementation of the clone method is platform dependent.

   returns

   a copy of the receiver object.

   Attributes

   protected[java.lang]

   Definition Classes

   AnyRef

   Annotations

   @throws( ... )

   Note

   not specified by SLS as a member of AnyRef

8. def cond[A, B](test: Boolean, right: ⇒ B, left: ⇒ A): Either[A, B]

   

   If the condition is satisfied, return the given B in Right, otherwise, return the given A in Left.

   If the condition is satisfied, return the given B in Right, otherwise, return the given A in Left.

   val userInput: String = ...
   Either.cond(
     userInput.forall(_.isDigit) && userInput.size == 10,
     PhoneNumber(userInput),
     "The input (%s) does not look like a phone number".format(userInput)

9. final def eq(arg0: AnyRef): Boolean

   

   Tests whether the argument (arg0) is a reference to the receiver object (this).

   Tests whether the argument (arg0) is a reference to the receiver object (this).

   The eq method implements an equivalence relation on non-null instances of AnyRef, and has three additional properties:

   • It is consistent: for any non-null instances x and y of type AnyRef, multiple invocations of x.eq(y) consistently returns true or consistently returns false.
   • For any non-null instance x of type AnyRef, x.eq(null) and null.eq(x) returns false.
   • null.eq(null) returns true.

   When overriding the equals or hashCode methods, it is important to ensure that their behavior is consistent with reference equality. Therefore, if two objects are references to each other (o1 eq o2), they should be equal to each other (o1 == o2) and they should hash to the same value (o1.hashCode == o2.hashCode).

   returns

   true if the argument is a reference to the receiver object; false otherwise.

   Definition Classes

   AnyRef

10. def equals(arg0: Any): Boolean

    

    The equality method for reference types.

    The equality method for reference types. Default implementation delegates to eq.

    See also equals in scala.Any.

    returns

    true if the receiver object is equivalent to the argument; false otherwise.

    Definition Classes

    AnyRef → Any

11. def finalize(): Unit

    

    Called by the garbage collector on the receiver object when there are no more references to the object.

    Called by the garbage collector on the receiver object when there are no more references to the object.

    The details of when and if the finalize method is invoked, as well as the interaction between finalize and non-local returns and exceptions, are all platform dependent.

    Attributes

    protected[java.lang]

    Definition Classes

    AnyRef

    Annotations

    @throws( classOf[java.lang.Throwable] )

    Note

    not specified by SLS as a member of AnyRef

12. final def getClass(): Class[_]

    

    A representation that corresponds to the dynamic class of the receiver object.

    A representation that corresponds to the dynamic class of the receiver object.

    The nature of the representation is platform dependent.

    returns

    a representation that corresponds to the dynamic class of the receiver object.

    Definition Classes

    AnyRef → Any

    Note

    not specified by SLS as a member of AnyRef

13. def hashCode(): Int

    

    The hashCode method for reference types.

    The hashCode method for reference types. See hashCode in scala.Any.

    returns

    the hash code value for this object.

    Definition Classes

    AnyRef → Any

14. final def isInstanceOf[T0]: Boolean

    

    Test whether the dynamic type of the receiver object is T0.

    Test whether the dynamic type of the receiver object is T0.

    Note that the result of the test is modulo Scala's erasure semantics. Therefore the expression 1.isInstanceOf[String] will return false, while the expression List(1).isInstanceOf[List[String]] will return true. In the latter example, because the type argument is erased as part of compilation it is not possible to check whether the contents of the list are of the specified type.

    returns

    true if the receiver object is an instance of erasure of type T0; false otherwise.

    Definition Classes

    Any

15. final def ne(arg0: AnyRef): Boolean

    

    Equivalent to !(this eq that).

    Equivalent to !(this eq that).

    returns

    true if the argument is not a reference to the receiver object; false otherwise.

    Definition Classes

    AnyRef

16. final def notify(): Unit

    

    Wakes up a single thread that is waiting on the receiver object's monitor.

    Wakes up a single thread that is waiting on the receiver object's monitor.

    Definition Classes

    AnyRef

    Note

    not specified by SLS as a member of AnyRef

17. final def notifyAll(): Unit

    

    Wakes up all threads that are waiting on the receiver object's monitor.

    Wakes up all threads that are waiting on the receiver object's monitor.

    Definition Classes

    AnyRef

    Note

    not specified by SLS as a member of AnyRef

18. final def synchronized[T0](arg0: ⇒ T0): T0

    

    Definition Classes

    AnyRef

19. def toString(): String

    

    Creates a String representation of this object.

    Creates a String representation of this object. The default representation is platform dependent. On the java platform it is the concatenation of the class name, "@", and the object's hashcode in hexadecimal.

    returns

    a String representation of the object.

    Definition Classes

    AnyRef → Any

20. final def wait(): Unit

    

    Definition Classes

    AnyRef

    Annotations

    @throws( ... )

21. final def wait(arg0: Long, arg1: Int): Unit

    

    Definition Classes

    AnyRef

    Annotations

    @throws( ... )

22. final def wait(arg0: Long): Unit

    

    Definition Classes

    AnyRef

    Annotations

    @throws( ... )