Haha, not exactly. You can replace
sj <- get let (a, sk) = runState something sj put sk with a <- something Also, you don't need do notation for single statements; "do return x" is just "return x" On Wed, Dec 22, 2010 at 7:21 PM, michael rice <nowg...@yahoo.com> wrote: > Thanks for the tip, Ozgur. It worked for me. Is this what you had in mind, > Ryan? > > Michael > > ============== > > import Control.Monad.State.Lazy > > import Control.Monad > import System.Random > > type GeneratorState = State StdGen > data Craps a = Roll a | Win a | Lose a deriving (Show) > > genRandomR :: Random a => (a,a) -> GeneratorState a > genRandomR = state . randomR > > rollDie :: GeneratorState Int > rollDie = genRandomR (1,6) > > roll2Dice :: GeneratorState Int > roll2Dice = liftM2 (+) rollDie rollDie > > f :: Craps [Int] -> GeneratorState (Craps [Int]) > f (Roll []) = do g0 <- get > let (throw1,g1) = runState roll2Dice g0 > put g1 > case throw1 of > 2 -> return (Lose [throw1]) > 3 -> return (Lose [throw1]) > 7 -> return (Win [throw1]) > 11 -> return (Win [throw1]) > _ -> do g1 <- get > let (throw2,g2) = runState roll2Dice g1 > put g2 > if throw2 == throw1 > then do return (Win [throw1,throw2]) > else > if throw2 == 7 > then do return (Lose [throw1,throw2]) > else do f (Roll [throw1,throw2]) > f (Roll z@(throw1:throws)) = do g0 <- get > let (throw,g1) = runState roll2Dice g0 > put g1 > if throw == throw1 > then do return (Win (z ++ [throw])) > else > if throw == 7 > then do return (Lose (z ++ [throw])) > else do f (Roll (z ++ [throw])) > > > > --- On *Wed, 12/22/10, Ozgur Akgun <ozgurak...@gmail.com>* wrote: > > > From: Ozgur Akgun <ozgurak...@gmail.com> > > Subject: Re: [Haskell-cafe] Why is Haskell flagging this? > To: "Ryan Ingram" <ryani.s...@gmail.com> > Cc: haskell-cafe@haskell.org, "Daniel Fischer" < > daniel.is.fisc...@googlemail.com> > Date: Wednesday, December 22, 2010, 7:37 PM > > > see also: > http://hackage.haskell.org/packages/archive/mtl/latest/doc/html/Control-Monad-State-Lazy.html#v:state > > On 22 December 2010 20:02, Ryan Ingram > <ryani.s...@gmail.com<http://mc/compose?to=ryani.s...@gmail.com> > > wrote: > > Interesting. In that case, > > state f = StateT $ \s -> Identity (f s) > > allows "state" to replace "State" in that code. > > > Ozgur > > -----Inline Attachment Follows----- > > > _______________________________________________ > Haskell-Cafe mailing list > Haskell-Cafe@haskell.org <http://mc/compose?to=haskell-c...@haskell.org> > http://www.haskell.org/mailman/listinfo/haskell-cafe > > >
_______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
