Share
Embed
Transcript
82 slides
-
THIS TALK: WHAT AND WHY What: The talk I wished
I attended before banging my head against this ֜: ᝒ㴼ͯΡڹ抑͡ර͞ͼΑͭ͡͵ͩ; -
THIS TALK: WHAT AND WHY What: The talk I wished
I attended before banging my head against this Why: Because I still remember how it was before knowing it ֜ඳ: ᎣΡڹ΄ͩ;ΨΔͶ憝͞ͼ͚Ρ͡Ο -
THE PROBLEM val x: Future[List[Int]] = ??? futureList.map(list => list.map(f))
^ ^ |________________| 2 maps 1 function -
FUNCTOR OF FUTURE val futureF = new Functor[Future] { def
map[A, B](fa: Future[A])(f: A => B): Future[B] = fa.map(f) } -
futureList.map(f) // not really valid scala | // but you
get the point | Functor[Future[List]].map(futureList)(f) -
IN PRACTICE // create a `Functor[Future[List]]` val futureListF = Functor[Future].compose(Functor[List])
val data: Future[List[Int]] = Future(List(1, 2, 3)) // only one map! futureListF.map(data)(_ + 1) // Future(List(2, 3, 4)) 䋚檭΄πЄϖ -
ABOUT FLATTENING List(1, 2, 3).map(_ + 1) // List(2, 3,
4) List(1, 2, 3).map(n => List.fill(n)(n)) // List(List(1), List(2, 2), List(3, 3, 3)) List(1, 2, 3).map(n => List.fill(n)(n)).flatten // List(1, 2, 2, 3, 3, 3) فৼ͵ϷφϕΨϢ϶ϐϕͯΡ -
FLATMAP flatten ∘ map = flatMap so these are the
same List(1, 2, 3).map(n => List.fill(n)(n)).flatten List(1, 2, 3).flatMap(n => List.fill(n)(n)) -
IN OTHER WORDS when life gives you F[F[A]] you probably
wanted flatMap e.g. val f: Future[Future[Int]] = Future(42).map(x => Future(24)) val g: Future[Int] = Future(42).flatMap(x => Future(24)) F[F[A]] Ψ憎͵Ο flatMap ;͜͠ -
THE 'M' WORD trait Monad[F[_]] { def pure[A](a: A): F[A]
def map[A, B](fa: F[A])(f: A => B): F[B] def flatMap[A, B](fa: F[A])(f: A => F[B]): F[B] } ̿ϯ̀ͽতΔΡίϹ -
THE 'M' WORD trait Monad[F[_]] { def pure[A](a: A): F[A]
def map[A, B](fa: F[A])(f: A => B): F[B] def flatMap[A, B](fa: F[A])(f: A => F[B]): F[B] } -
A LESS CONTRIVED EXAMPLE case class User(name: String) case class
Address(city: String) Θͭ͜䋚አጱֺ -
A LESS CONTRIVED EXAMPLE def getUser(name: String): Future[User] def getAddress(user:
User): Future[Address] val getCity: Future[String] = getUser("Gabriele").flatMap( gab => getAddress(gab).map( address => address.city ) ) -
LET'S COMPREHEND THIS val getCity: Future[String] = for { gab
<- getUser("Gabriele") address <- getAddress(gab) } yield address.city for ٖ۱ᤒ懿ͽ䨗ͧΡ -
LESSONS monads allow sequential execution monads can squash F[F[A]] into
F[A] ϯϗϖ΅᭑ེ䋚ᤈΨݢᚆ;ͯΡ ϯϗϖ΅ `F[F[A]]` Ψ `F[A]` ͺΌͯͩ;͢ͽͣΡ -
BACK TO THE REAL WORLD def getUser(name: String): Future[User] //
<- really? def getAddress(user: User): Future[Address] 䋚ֺͽᘍ͞ͼΕΔͭΝ͜ -
BACK TO THE REAL WORLD def getUser(name: String): Future[Option[User]] //
better def getAddress(user: User): Future[Option[Address]] ΞΠᜉֺ͚ΠΔͭ͵ -
UH, OH... val city: Future[Option[String]] = for { gab <-
getUser("Gabriele") address <- getAddress(gab) // FAIL } yield address.city ०䤂ͭͼͭΔ͚Δͯ -
EVENTUALLY val city: Future[Option[String]] = for { gab <- getUser("Gabriele")
address <- getAddress(gab.get) // } yield address.get.city // get ֵ͵ΟύϮ -
OR... val city: Future[Option[String]] = for { maybeUser <- getUser("Gabriele")
maybeCity <- maybeUser match { case Some(user) => getAddress(user).map(_.map(_.city)) case None => Future.successful(None) } } yield maybeCity ;͚ͩ͜;΅... -
WHAT WE WOULD REALLY WANT val city: Future[Option[String]] = for
{ gab <- maybeUser <- getUser("Gabriele") address <- maybeAddress <- getAddress(gab) } yield address.city ཿͭ͡͵Θ΄ -
WHAT'S THE IMPOSSIBLE PART? // trivial def compose[F[_]: Functor, G[_]:
Functor]: Functor[F[G[_]]] = // impossible def compose[M[_]: Monad, N[_]: Monad]: Monad[M[N[_]]] = // (not valid scala, but you get the idea) 吖ΡϯϗϖΨݳ౮ͭͼϯϗϖΨ୵౮ͯΡͩ;΅ͽ͚ͣ -
flatMap FOR Future[Option[A]] val city: Future[Option[String]] = for { maybeUser
<- getUser("Gabriele") maybeCity <- maybeUser match { case Some(user) => getAddress(user).map(_.map(_.city)) case None => Future.successful(None) } } yield maybeCity Future[Option[String]] ΄͵Η΄ flatMap -
THE MONAD INTERFACE trait Monad[F[_]] { def pure[A](a: A): F[A]
def map[A, B](fa: F[A])(f: A => B): F[B] def flatMap[A, B](fa: F[A])(f: A => F[B]): F[B] } -
implicit val futOptMonad: Monad[FutOpt] = new Monad[FutOpt] { def pure[A](a:
A): FutOpt[A] = FutOpt(a.pure[Option].pure[Future]) def map[A, B](fa: FutOpt[A])(f: A => B): FutOpt[B] = FutOpt(fa.value.map(optA => optA.map(f))) def flatMap[A, B](fa: FutOpt[A])(f: A => FutOpt[B]): FutOpt[B] = FutOpt(fa.value.flatMap { case Some(a) => f(a).value case None => (None: Option[B]).pure[Future] }) // omitting tailRecM here } -
AND USE val f: FutOpt[String] = for { gab <-
FutOpt(getUser("Gabriele")) address <- FutOpt(getAddress(gab)) } yield address.city // ! val city: Future[Option[String]] = f.value ֵͼΕΡ -
implicit val listOptMonad: Monad[ListOpt] = new Monad[ListOpt] { def pure[A](a:
A): ListOpt[A] = ListOpt(a.pure[Option].pure[List]) def map[A, B](fa: ListOpt[A])(f: A => B): ListOpt[B] = ListOpt(fa.value.map(optA => optA.map(f))) def flatMap[A, B](fa: ListOpt[A])(f: A => ListOpt[B]): ListOpt[B] = ListOpt(fa.value.flatMap(opt => opt match { case Some(a) => f(a).value case None => (None: Option[B]).pure[List] })) // omitting tailRecM here } -
implicit val futOptMonad: Monad[FutOpt] = new Monad[FutOpt] { def pure[A](a:
A): FutOpt[A] = FutOpt(a.pure[Option].pure[Future]) def map[A, B](fa: FutOpt[A])(f: A => B): FutOpt[B] = FutOpt(fa.value.map(optA => optA.map(f))) def flatMap[A, B](fa: FutOpt[A])(f: A => FutOpt[B]): FutOpt[B] = FutOpt(fa.value.flatMap { case Some(a) => f(a).value case None => (None: Option[B]).pure[Future] }) // omitting tailRecM here } -
implicit val futOptMonad: Monad[FutOpt] = new Monad[FutOpt] { def pure[A](a:
=> A): FutOpt[A] = FutOpt(a.pure[Option].pure[Future]) def map[A, B](fa: FutOpt[A])(f: A => B): FutOpt[B] = FutOpt(fa.value.map(optA => optA.map(f))) def flatMap[A, B](fa: FutOpt[A])(f: A => FutOpt[B]): FutOpt[B] = FutOpt(fa.value.flatMap(opt => opt match { case Some(a) => f(a).value case None => (None: Option[B]).pure[Future] })) // omitting tailRecM here } -
MEET OptionT val f: OptionT[Future, String] = for { gab
<- OptionT(getUser("Gabriele")) address <- OptionT(getAddress(gab)) } yield address.city // ! val city: Future[Option[String]] = f.value -
ANOTHER EXAMPLE def getUser(id: String): Future[Option[User]] = ??? def getAge(user:
User): Future[Int] = ??? def getNickname(user: User): Option[String] = ??? val lameNickname: Future[Option[String]] = ??? // e.g. Success(Some("gabro27")) 㳨΄ֺ -
I KNOW THE TRICK! val lameNickname: OptionT[Future, String]] = for
{ user <- OptionT(getUser("123")) age <- OptionT(getAge(user)) // sorry, nope name <- OptionT(getNickname(user)) // sorry, neither } yield s"$name$age" ͜Δ͚͚ͥ͡ -
DO YOU EVEN LIFT, BRO? val lameNickname: OptionT[Future, String]] =
for { user <- OptionT(getUser("123")) age <- OptionT.liftF(getAge(user)) name <- OptionT.fromOption(getNickname(user)) } yield s"$name$age" lift (ᒶϕϹ) ;ͭ͡ͼ͚? -
EXAMPLE: UPDATING A USER > check user exists > check
it can be updated > update it ๅෛ΄ֺ -
THE NAIVE WAY def checkUserExists(id: String): Future[Option[User]] def checkCanBeUpdated(u: User):
Future[Boolean] def updateUserOnDb(u: User): Future[User] ύϮොဩ -
PROBLEMS def updateUser(u: User): Future[Option[User]] = checkUserExists("foo").flatMap { maybeUser =>
maybeUser match { case Some(user) => checkCanBeUpdated(user).flatMap { canBeUpdated => if (canBeUpdated) { updateUserOnDb(user).map(Some(_)) } else { Future.successful(None) } } case None => Future.successful(None) } } πЄϖ΄憎᭗ͭ͢䘂͚ -
MORE PROBLEMS (DETAILED ERRORS) case class MyError(msg: String) def updateUser(u:
User): Future[Either[MyError, User]] = checkUserExists("foo").flatMap { maybeUser => maybeUser match { case Some(user) => checkCanBeUpdated(user).flatMap { canBeUpdated => if (canBeUpdated) { updateUserOnDb(user).map(Right(_)) } else { Future.successful(Left(MyError("user cannot be updated"))) } } case None => Future.successful(Left(MyError("user does not exist"))) } } 抎ΕͻΟ͚ΔΔ -
HOW ABOUT case class MyError(msg: String) type ResultT[F[_], A] =
EitherT[F, MyError, A] type FutureResult[A] = ResultT[Future, A] ͩΟͿ͜ͽͭΝ͜͡ -
SOME HELPERS object FutureResult { def apply[A](a: A): FutureResult[A] =
apply(Future.successful(a)) def apply[A](fa: Future[A]): FutureResult[A] = EitherT.liftT(fa) def apply[A](e: Either[MyError, A]): FutureResult[A] = EitherT.fromEither(e) } ϥϸϞЄ樛හΨአ -
def checkUserExists(id: String): FutureResult[User] = FutureResult { if (id ===
"123") User("123").asRight else MyError("sorry, no user").asLeft } def checkCanBeUpdated(u: User): FutureResult[Unit] = ??? def updateUserOnDb(u: User): FutureResult[User] = ??? -
BETTER? def updateUser(user: User): FutureResult[User] = for { user <-
checkUserExists(user.id) _ <- checkCanBeUpdated(user) updatedUser <- updateUser(user) } yield updatedUser Ξͥ͵ͽͭΝ͜͡ -
TIP #1 stacking more than two monads gets bad really
quickly 2ͺզӤ΄ϯϗϖΨ坌ΕӤͨΡ;ᬔͥΡ -
EXAMPLE1 val effect: OptionT[EitherT[Task, String, ?], String] = for {
first <- readName.liftM[EitherT[?[_], String, ?]].liftM[OptionT] last <- readName.liftM[(EitherT[?[_], String, ?]].liftM[OptionT] name <- if ((first.length * last.length) < 20) OptionT.some[EitherT[Task, String, ?], String](s"$first $last") else OptionT.none[EitherT[Task, String, ?], String] _ <- (if (name == "Daniel Spiewak") EitherT.fromDisjunction[Task](\/.left[String, Unit]("your kind isn't welcome here")) else EitherT.fromDisjunction[Task](\/.right[String, Unit](()))).liftM[OptionT] _ <- log(s"successfully read in $name").liftM[EitherT[?[_], String, ?]].liftM[OptionT] } yield name 1 from djspiewak/emm -
TIP #2 keep your transformers for youself def publicApiMethod(x: String):
OptionT[Future, Int] = def publicApiMethod(x: String): Future[Option[Int]] = by the way val x: OptionT[Future, Int] = OptionT(Future(Option(42))) val y: Future[Option[Int]] = x.value // Future(Option(42)) ϯϗϖ䄜䟵ৼ΅ API ڊ͚ͫΞͯ͜Ρ -
TIP #3 ! Perf! Wrapping/unwrapping isn't cheap, so if you're
concerned about performance, consider benchmarking your code. ᯿͚πЄϖ΄ͽᚆΨ䶲ͯΡΟϦЀώϫЄμΨݐΡ -
TIP #4 Use them as a ""local optimization"". In case
your problem is not "local", consider alternative approaches. WHAT ELSE? ੴಅጱ๋晒۸አ͚Ρ -
FREE MONADS / TAGLESS FINAL > clearly separate structure and
interpretation > effects are separated from program definition http://typelevel.org/cats/datatypes/freemonad.html https://blog.scalac.io/exploring-tagless-final.html ᛔኧϯϗϖ΅ϤϺν϶ϭਧ嬝͡Ο֢አΨړ櫝ͯΡ -
EFF https://github.com/atnos-org/eff-cats "Extensible effects are an alternative to monad transformers
for computing with effects in a functional way" based on Freer Monads, More Extensible Effects by Oleg Kiselyov Eff ΅ϯϗϖ䄜䟵ৼ΄դ๊;ΡΘ΄ͽ̵֢አΨ樛හࣳጱ䜷͜