August 2013 - Haskell-Cafe

Re: [Haskell-cafe] sequence causing stack overflow on pretty small lists
by John Lato 28 Aug '13

28 Aug '13

IMHO it's perfectly reasonable to expect sequence/replicateM/mapM to be able to handle a list of ~1e6 elements in the Unescapable Monad (i.e. IO). All the alternate implementations in the world won't be as handy as Prelude.sequence, and no amount of documentation will prevent people from running into this headlong*. So unless there's a downside to upping the stack size limitation I'm unaware of, +1 to that suggestion from me. John [1] Most people are physically incapable of reading documents that explain why what they want to do won't work. Even if people did read the documentation, I suspect that the people most in need of the information would be the least likely to understand how it applies to their situation. On Tue, Aug 27, 2013 at 9:19 PM, John Alfred Nathanael Chee < cheecheeo(a)gmail.com> wrote: > This is somewhat related: http://ghc.haskell.org/trac/ghc/ticket/4219 > > This also solves the concrete problem you gave in your original post > (in reverse order): > > import Control.Monad > import System.Random > > sequencel :: Monad m => [m a] -> m [a] > sequencel = foldM (\tail m -> (\x -> return $ x : tail) =<< m) [] > > main :: IO () > main = print =<< sequencel (replicate 1000000 (randomIO :: IO Integer)) > > Following on Reid's point, maybe it's worth noting in the > documentation that replicateM, mapM, and sequence are not tail > recursive for Monads that define (>>=) as strict in the first > argument? > > On Tue, Aug 27, 2013 at 6:07 AM, Niklas Hambüchen <mail(a)nh2.me> wrote: > > On 27/08/13 20:37, Patrick Palka wrote: > >> You can use ContT to force the function to use heap instead of stack > >> space, e.g. runContT (replicateM 1000000 (lift randomIO)) return > > > > That is interesting, and works. > > > > Unfortunately its pure existence will not fix sequence, mapM etc. in > base. > > > > _______________________________________________ > > Libraries mailing list > > Libraries(a)haskell.org > > http://www.haskell.org/mailman/listinfo/libraries > > > > -- > Love in Jesus Christ, John Alfred Nathanael Chee > http://www.biblegateway.com/ > http://web.cecs.pdx.edu/~chee/ > > _______________________________________________ > Libraries mailing list > Libraries(a)haskell.org > http://www.haskell.org/mailman/listinfo/libraries >

2 1

Template Haskell: let statement in a splice put in the "main = do" part of a program?
by TP 27 Aug '13

27 Aug '13

Hi, I continue to test Template Haskell, and I have some difficulties to use a splice $() in a "do" contained in the "main" part of a program. Here is an example. I want to make a splice that does `let a="a"` in my code. $ cat MakeLetStatement.hs ---------------------------- {-# LANGUAGE TemplateHaskell #-} module MakeLetStatement where import Language.Haskell.TH makeLetStatement :: String -> ExpQ makeLetStatement s = return $ DoE $ [ LetS $ [ ValD (VarP $ mkName s) (NormalB $ LitE $ StringL s) [] ]] ---------------------------- $ cat test_MakeLetStatement.hs ---------------------------- {-# LANGUAGE TemplateHaskell #-} import MakeLetStatement main = do $(makeLetStatement "a") -- print a ---------------------------- Note I have commented "print a" because otherwise I obtain "Not in scope: `a'" that shows that `a` has not been defined correctly, but does not show whether my splice has been correctly expanded (I use --dump-splices GHC option, but it seems it is not working for splices in the "main = do" part). I obtain: $ runghc -ddump-splices test_MakeLetStatement.hs test_MakeLetStatement.hs:7:3: Illegal last statement of a 'do' block: let a = "a" (It should be an expression.) When splicing a TH expression: do let a = "a" In a stmt of a 'do' block: $(makeLetStatement "a") In the expression: do { $(makeLetStatement "a") } In an equation for `main': main = do { $(makeLetStatement "a") } That shows that my splice has been correctly expanded: we have `let a = "a"`. However, what happens is the same as in the following dummy script, we have in fact defined a "do" inside the first "do" (with DoE), and so we obtain an error because the last statement in a do block should be an expression. ---------------------------- main = do do let a = "a" print a ---------------------------- So my code does not work, without surprise, but in fact my problem is to transform a LetS statement: http://www.haskell.org/ghc/docs/latest/html/libraries/template-haskell-2.8.… that has type Stmt, in an ExpQ that seems to be the only thing that we can put in a splice. I have found that it can only be done by doE (or DoE) and compE (or CompE) according to http://www.haskell.org/ghc/docs/latest/html/libraries/template-haskell-2.8.… But doE is not a solution as we have seen above, and compE is to construct list comprehensions, which is a different thing. So, is there any solution to my problem? Thanks in advance, TP

3 4

Template Haskell
by Jose A. Lopes 27 Aug '13

27 Aug '13

Hi, Is it possible to retrieve all definitions contained in a module using Template Haskell ? Thanks, Jose -- Jose Antonio Lopes Ganeti Engineering Google Germany GmbH Dienerstr. 12, 80331, München Registergericht und -nummer: Hamburg, HRB 86891 Sitz der Gesellschaft: Hamburg Geschäftsführer: Graham Law, Christine Elizabeth Flores Steuernummer: 48/725/00206 Umsatzsteueridentifikationsnummer: DE813741370

2 2

Re: [Haskell-cafe] Extending Type Classes
by Henning Thielemann 27 Aug '13

27 Aug '13

The problem of refinement of type classes annoys me from time to time when I work on the NumericPrelude. It is an experimental type class hierarchy for mathematical types. Sometimes a new data type T shall be implemented and it turns out that you can implement only a part of all methods of a certain class. Then a natural step is to split the class into two classes A and B: 'A' contains the methods we can implement for T and 'B' contains the remaining methods and 'B' is a sub-class of 'A'. First, this means that all client code has to be rewritten. Second, code for instances becomes very lengthy, because over the time code tends to contain one instances for every method. However the many small instances actually carry information: Every instance has its specialised constraints. E.g. you would certainly try to use only Applicative constraints in an Applicative instance and not Monad constraints. However, if there is a way to define Applicative and Monad instances in one go, the Applicative instance may get Monad constraints.

2 1

Re: [Haskell-cafe] sequence causing stack overflow on pretty small lists
by Niklas Hambüchen 26 Aug '13

26 Aug '13

Maybe an unlimited stack size should be the default? As far as I understand, the only negative effect would be that some programming mistakes would not result in a stack overflow. However, I doubt the usefulness of that: * It already depends a lot on the optimisation level * If you do the same thing in a slightly different way, and you allocate on the heap instead of on the stack you will not get it either

1 0

definition of the term "combinator"
by damodar kulkarni 26 Aug '13

26 Aug '13

Hello, The word "combinator" is used several times in the Haskell community. e.g. parser combinator, combinator library etc. Is it exactly the same term that is used in the "combinatory logic" ? A combinator is a higher-order function that uses *only function application* and earlier defined combinators to define a result from its arguments. [1] It seems, the term combinator as in, say, "parser combinator", doesn't have much to do with the "*only function application*" requirement of the "combinatory logic", per se. If the above observation holds, is the term combinator as used in the Haskell community, properly defined? In other words: Where can I find a formal and precise definition of the term "combinator", as a term used by the Haskell community to describe "something"? Ref: http://en.wikipedia.org/wiki/Combinatory_logic Thanks and regards, -Damodar Kulkarni

4 4

TypeLits & Typeable
by Nicolas Trangez 26 Aug '13

26 Aug '13

Hello Cafe, I was playing around with TypeLits in combination with Typeable (using GHC 7.7.7.20130812 FWIW), but was surprised to find Symbols aren't Typeable, and as such the following doesn't work. Is this intentional, or am I missing something? Thanks, Nicolas {-# LANGUAGE DataKinds, KindSignatures, DeriveFunctor, DeriveDataTypeable #-} module Main where import Data.Typeable import GHC.TypeLits data NoSymbol n a b = NoSymbol a b deriving (Typeable) data WithSymbol (n :: Symbol) a b = WithSymbol a b deriving (Typeable) data Sym deriving (Typeable) main :: IO () main = do print $ typeOf (undefined :: NoSymbol Sym Int Int) let d = undefined :: WithSymbol "sym" Int Int {- print $ typeOf d No instance for (Typeable Symbol "sym") arising from a use of ‛typeOf’ -} return ()

3 2

Proposal: Hackage's packages should be seperated by buildable
by He-chien Tsai 26 Aug '13

26 Aug '13

I'm sick for checking whether package is obsolete or not. I think packages build failed long time ago should be collected and moved to another page until someone fix them, or hackage pages should have a filter for checking obsolete packages.

3 2

xmonad (+ mate) evince problem?
by Johannes Waldmann 26 Aug '13

26 Aug '13

Hi. I am using xmonad + mate (on fedora 19) and evince (PDF viewer) seems unresponsive: it reacts to my (mouse) input only after switching to another screen and back. Any hints? - Thanks, J. W.

1 1

Conduit : is it possible to write this function?
by Erik de Castro Lopo 26 Aug '13

26 Aug '13

Hi all Using the Conduit library is it possible to write the function: eitherSrc :: MonadResource m => Source m a -> Source m b -> Source m (Either a b) which combines two sources into new output source such that data being produced aysnchronously by the original two sources will be returned as either a Left or Right of tne new source? If so, how? Cheers, Erik -- ---------------------------------------------------------------------- Erik de Castro Lopo http://www.mega-nerd.com/

2 2