- Beginners - Haskell.org

Parse file with existentials
by Dmitriy Matrosov 13 Dec '16

13 Dec '16

> {-# LANGUAGE GADTs #-} > {-# LANGUAGE DataKinds #-} > {-# LANGUAGE RankNTypes #-} > {-# LANGUAGE KindSignatures #-} > {-# LANGUAGE StandaloneDeriving #-} > {-# LANGUAGE FlexibleInstances #-} > {-# LANGUAGE DeriveDataTypeable #-} > > import Prelude hiding (getLine) > import Data.Maybe > import Data.List > import Data.Typeable > import Control.Monad.Identity > import Control.Monad.Trans.Identity > import Control.Monad.Writer > import Control.Applicative > import System.FilePath Hi. I ask for an opinion about interface (implemented below) for parsing `rsync` filter files. The parser does not parse full syntax, i wrote it for determining rsync filter dependencies, when installing them using `shake`. So, i distinguish two kinds of lines: include of another filter file, which looks like . file and any other. I want to distinguish them at type level, so e.g. a record function for one constructor can't be applied to another, etc. I know, that i can prevent this at runtime by exporting only smart constructor, but i want a type check. > data RsyncFilterT = IncludeT | LineT > type IncludeT = 'IncludeT > type LineT = 'LineT > > -- Particular rsync filters distinguishable at type-level. > data RsyncFilter :: RsyncFilterT -> * where > Include :: {getInclude :: FilePath} -> RsyncFilter 'IncludeT > Line :: {getLine :: String} -> RsyncFilter 'LineT > deriving instance Show (RsyncFilter a) > deriving instance Typeable RsyncFilter For accessing records i use lenses redefined here. I redefine them with `Applicative` instead of `Functor` to make modify/set work even if value does not have a required record (by returning original (unmodified) value using `pure`). > type LensA a b = forall f. Applicative f => (b -> f b) -> a -> f a > > viewA :: LensA a b -> a -> b > viewA l = fromJust . getLast . getConst . l (Const . Last . Just) > viewAmaybe :: LensA a b -> a -> Maybe b > viewAmaybe l = getLast . getConst . l (Const . Last . Just) > > modifyA :: LensA a b -> (b -> b) -> a -> a > modifyA l f = runIdentity . l (Identity . f) > > modifyAA :: Applicative t => LensA a b -> (b -> t b) -> a -> t a > modifyAA l f = runIdentityT . l (IdentityT . f) > > setA :: LensA a b -> b -> a -> a > setA l s = modifyA l (const s) Here're lenses for `RsyncFilter` (its constructors are distinguishable at type-level, so i don't really need `Applicative` lenses here): > includeL :: LensA (RsyncFilter 'IncludeT) FilePath > includeL f z@Include {getInclude = x} = > fmap (\x' -> z{getInclude = x'}) (f x) > lineL :: LensA (RsyncFilter 'LineT) FilePath > lineL f z@Line {getLine = x} = > fmap (\x' -> z{getLine = x'}) (f x) The order of lines (may) matter, so i need to store all `RsyncFilter a` values in a list in original file order. But now the values are of different type. So.. i use existential container: > -- Generic container for any type of rsync filter. > data AnyFilter = forall (a :: RsyncFilterT). Typeable a => > AnyFilter (RsyncFilter a) > deriving instance Show AnyFilter > deriving instance Typeable AnyFilter And still i want to work on values of certain type to have some guarantees against misuse, so i need to cast `AnyFilter` back into `RsyncFilter` value: > -- Extract rsync filter from AnyFilter. > getFilter :: (forall (a :: RsyncFilterT). Typeable a => > RsyncFilter a -> b) -> AnyFilter -> b > getFilter f (AnyFilter x) = f x and here i also want to use lenses, but now the value may be of different type, that the lens expect, so i really need `Applicative` lenses here: > rsyncIncludeL' :: LensA AnyFilter (RsyncFilter 'IncludeT) > rsyncIncludeL' f z = maybe (pure z) (fmap AnyFilter . f) (getFilter cast z) > rsyncIncludeL :: LensA AnyFilter FilePath > rsyncIncludeL = rsyncIncludeL' . includeL > > rsyncLineL' :: LensA AnyFilter (RsyncFilter 'LineT) > rsyncLineL' f z = maybe (pure z) (fmap AnyFilter . f) > (getFilter cast z) > rsyncLineL :: LensA AnyFilter String > rsyncLineL = rsyncLineL' . lineL Then i define another Read/Show class just to be able to keep default Read/Show instances: > class Serialize a where > fromString :: String -> Maybe a > toString :: a -> String > > instance Serialize (RsyncFilter 'LineT) where > fromString = Just . Line > toString (Line xs) = xs > > -- RULE and PATTERN separator is space (`_` not supported). > -- Only short rule names without modifiers are supported. > instance Serialize (RsyncFilter 'IncludeT) where > fromString = go . break (== ' ') > where > go :: (String, String) -> Maybe (RsyncFilter 'IncludeT) > go (r, _ : x : xs) > | r == "." = Just (Include (x : xs)) > go _ = Nothing > toString (Include xs) = ". " ++ xs > > instance Serialize AnyFilter where > fromString x = > fmap AnyFilter (fromString x :: Maybe (RsyncFilter 'IncludeT)) > <|> fmap AnyFilter (fromString x :: Maybe (RsyncFilter 'LineT)) > toString x = fromMaybe "" $ > fmap toString (viewAmaybe rsyncIncludeL' x) > <|> fmap toString (viewAmaybe rsyncLineL' x) and a lens from String to AnyFilter, which effectively parses file and writes it back: > rsyncAnyL :: LensA String AnyFilter > rsyncAnyL f z = maybe (pure z) (fmap toString . f) (fromString z) And here is how i use this: > -- | Replace path prefix, if matched. > replacePrefix :: FilePath -> FilePath -> FilePath -> FilePath > replacePrefix old new x = maybe x (combine new . joinPath) $ > -- For ensuring that path prefix starts and ends at path > -- components (directories) boundaries, i first split them. > stripPrefix (splitDirectories old) (splitDirectories x) > > -- | Rewrite path in rsync inlcude line @line@ from source path > -- @srcdir@ to install path @prefix@ > -- > -- > usedIncludes srcdir prefix line > -- > -- and collect (rewritten) rsync include pathes in @Writer@ monad. > -- Other lines return as is. > usedIncludes :: FilePath -- ^ Source path. > -> FilePath -- ^ Install path. > -> String -- ^ Line from rsync filter file. > -> Writer [FilePath] String > usedIncludes srcdir prefix = > modifyAA (rsyncAnyL . rsyncIncludeL) $ \x -> do > let x' = replacePrefix srcdir prefix x > tell [x'] > return x' and then a shake rule: -- | Add file rule for instaling rsync filters with extension -- @ext@, rewriting source path @srcdir@ to install path @prefix@ -- in any rsync includes: -- -- > rsyncFilter ext srcdir prefix -- rsyncFilter :: String -- ^ Extension. -> FilePath -- ^ Install path. -> FilePath -- ^ Source path. -> Rules () rsyncFilter ext prefix srcdir = prefix ++ "//*" <.> ext %> \out -> do let src = replacePrefix prefix srcdir out ls <- readFileLines src let (rs, incs) = runWriter $ mapM (usedIncludes srcdir prefix) ls need incs putNormal $ "> Write " ++ out writeFileChanged out . unlines $ rs I probably won't think too much about this API, if i haven't read [Luke Palmer's post about existentials][1] . And now i doubt, did i fall into the same trap with existentials and does not see an obvious solution with functions? [1]: https://lukepalmer.wordpress.com/2010/01/24/haskell-antipattern-existential… -- Dmitriy Matrosov

1 0

Real World Haskell: Deprecated Libraries
by John Ky 27 Nov '16

27 Nov '16

Hello Haskellers, I have been working through the Real World Haskell chapter on Profiling and Optimisation <http://book.realworldhaskell.org/read/profiling-and-optimization.html>, and got to the section *Advanced techniques: fusion*, where I encountered some code using Data.Array.Vector from the deprecated library uvector. Does anyone know how to rewrite this to not use deprecated libraries? I have tried to use Data.Vector.Unboxed from vector, but my version consumes a lot of memory. The code I have is here: https://github.com/haskell-works/hw-tutorial-performance/blob/master/hw-tut… With that code, I get “1831 MB total memory in use”, which is terrible Build run steps I used were: stack build --executable-profiling --library-profiling --ghc-options="-fprof-auto -rtsopts -auto-all -caf-all -fforce-recomp" time /Users/jky/wrk/haskell-works/hw-tutorial-performance/.stack-work/install/x86_64-osx/lts-7.10/8.0.1/bin/hw-tutorial-performance-rwhe +RTS -sstderr -p -hy -RTS 1e7 I get very large numbers of allocations of Pair and Double. There is also a very large upfront allocation ARR_WORDS, which I feared would happen when using Data.Vector Any help appreciated. Cheers, -John

2 2

Setting up WXHaskell on the 8.0.1 platform?
by Cleverson Casarin Uliana 27 Nov '16

27 Nov '16

Hello all, I've sent this to the WXHaskell list, but no answers. Please what is the recommended way to set up WXWidgets and WXHaskell for the 8.0.1 Haskell Platform on Windows? I tried using the Achelanne installer, but it gives errors apparently related to packages versions incompatibilities. Thanks, Cleverson

2 1

pattern matching on a common element
by briand＠aracnet.com 25 Nov '16

25 Nov '16

Here's what I'm doing: data X = A1 String Double | A2 String Int | A3 String Double Int name c = case c of A1 name _ -> name A2 name _ -> name A3 name _ _ -> name I'm sure there's a better way...

5 10

Equivalence (or not) of lists
by Lawrence Wickert 13 Nov '16

13 Nov '16

Hello all, I am a rank beginner to functional languages. Working through Lipovaca's book, up to Chapter 3. Ok, setup this function in editor and compiled: length' :: (Num b) => [a] -> b length' [] = 0 length' (_:xs) = 1 + length' xs skippy@skippy:~$ ghci GHCi, version 7.10.3: http://www.haskell.org/ghc/ :? for help Prelude> :l baby [1 of 1] Compiling Main ( baby.hs, interpreted ) Ok, modules loaded: Main. *Main> length' [1,2,3] 3 *Main> 1:2:3:[] [1,2,3] *Main> length' 1:2:3:[] <interactive>:5:9: Could not deduce (Num [a0]) arising from the literal '1' from the context (Num a) bound by the inferred type of it :: Num a => [a] at <interactive>:5:1-16 The type variable 'a0' is ambiguous In the first argument of 'length'', namely '1' In the first argument of '(:)', namely 'length' 1' In the expression: length' 1 : 2 : 3 : [] *Main> Obviously, there is something I don't understand about the apparent non-equivalence of the lists [1,2,3] and 1:2:3:[]I am guessing that the solution is contained in that error message but I can't quite decipher it. Thanks for any help.

4 3

How to tell haskell that 'a' is 'a' ?
by Lai Boon Hui 13 Nov '16

13 Nov '16

Hi all, skips :: [a] -> [[a]] skips xs = go 1 [] where go :: Int -> [[a]] -> [[a]] *-- Compiles if i remove this line* go n acc | n > (length xs) = acc | n == 1 = xs : (go (n+1) acc) | n < 1 = [] | otherwise = (everyN n xs) : (go (n+1) acc) everyN :: Int -> [a] -> [a] everyN n xs = case (drop (n-1) xs) of y:ys -> y : (everyN n ys) [] -> [] How can i tell haskell compiler that a is the sub program if the same as a in the main program? -- Best Regards, Boon Hui

2 1

Request for Code Review: Dice Game Distribution
by Jan Brusch 12 Nov '16

12 Nov '16

Hi, I'm currently learning Haskell and as a first project I wrote a little helper for a dice game. It's very short. Nonetheless I would really appreciate a little Code Review or any other input. The Game: Each player writes down the numbers from 1 to 42. Each turn you roll 3 dice (6-sided). The rolling player can cross of any one number that he can construct from the three dice using the math operators +, -, *, /. First player to cross off all numbers wins. The Helper: Calculates the reachable numbers for each dice roll. It the aggregates a count for each number, from how many dice rolls it can be reached. E.g.: There are 216 possible dice rolls (6^3), the number 2 can be reached form 171 of them. The helper gives out the data in JSON format that can be read by NVD3 to be displayed in a browser. The graph helps you to make a decision on which number you should prefer to cross off with your current roll. The Code: http://lpaste.net/338358 The Graph: http://imgur.com/a/NSDT8 Looking forward to any feedback and thanks in advance Jan

1 0

Equivalence (or not) of lists
by Lawrence Wickert 12 Nov '16

12 Nov '16

Thanks, Sylvain. That works and I think I understand the difference now. Larry

1 0

OI UTCTime to String
by sasa bogicevic 10 Nov '16

10 Nov '16

Hi All, This is a small program {-# LANGUAGE OverloadedStrings #-} module Main where import Control.Monad.IO.Class (liftIO) import qualified Data.ByteString.Lazy.Char8 as L import Data.Time import Network (withSocketsDo) import Network.HTTP.Conduit createRequestData today = [("index:brKursneListe",""), ("index:year","2016"), ("index:inputCalendar1", today), ("index:vrsta","3"), ("index:prikaz","0"), ("index:buttonShow","Prikazi")] timeFromString s = parseTimeOrError True defaultTimeLocale "%d %b %Y %l:%M %p" s formatDateString time = formatTime defaultTimeLocale "%m/%d/%Y" time getDateString = getCurrentTime getFormatedDate = formatDateString $ timeFromString getDateString main = do print $ getFormatedDate And here is my error main.hs:25:54: error: • Couldn't match type ‘IO UTCTime’ with ‘[Char]’ Expected type: String Actual type: IO UTCTime • In the first argument of ‘timeFromString’, namely ‘getDateString’ In the second argument of ‘($)’, namely ‘timeFromString getDateString’ In the expression: formatDateString $ timeFromString getDateString Thanks!

4 4

Re: [Haskell-beginners] Constructor classes and type classes.
by PATRICK BROWNE 03 Nov '16

03 Nov '16

Imants Thanks for the sanity check on the instances. I am concerned about the use of the function immsucc' from the same paper <http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.36.9395>, which is defined in instances the Time class (but omitted from the class). The function immsucc' is first used without arguments, which to me implies that it 'might' a function passed to the higher-order closure function. It is then used as just an ordinary function with arguments. I have found a version of closure based on Craft of Function Programming. <https://github.com/buntine/Haskell--Craft-of-FP/blob/master/AbsTypes/Relati…> which does not require a function as an argument. So basically my question is: does the usage of immsucc' make sense in the code below?, do the types match up? Regards, Pat data TimeFacts t = Bef (t, t, TimeFacts t) | New class SetClass t => Time l t where new :: (t, t) -> l t bef :: (t, t, l t) -> l t succ , pred :: (t, l t) -> Set t immsucc , immpred :: (t, l t) -> Set t before, after :: (t, t, l t) -> Bool immedbefore, immedafter :: (t, t, l t) -> Bool immedafter (a, b, t) = immedbefore (b, a, t) after (a, b, t) = before (b, a, t) before (a, b, t) = memberInSet (b, succ (a, t)) succ (a, tl) = closure ([a], immsucc' , tl) pred (a, tl) = closure ([a], immpred' , tl) immsucc (a, tl) = immsucc' ([a], [], tl) immpred (a, tl) = immpred' ([a], [], tl) instance (SetClass t) => Time TimeFacts t where immsucc’ (a, s, New ) = s immsucc’ (a, s, Bef (e, f, tt)) = if memberInSet (e,a) then immsucc’ (a, insertInSet (f,s), tt) else immsucc’ (a, s,tt) On 3 November 2016 at 18:05, Imants Cekusins <imantc(a)gmail.com> wrote: > this builds. not sure if the instances are as intended > > > {-# LANGUAGE MultiParamTypeClasses, FlexibleInstances #-} > module SetClass where > > data TimeFacts t = Bef (t, t, TimeFacts t) | New > > class SetClass t where > class SetClass t => Time l t where > class Time l t => TimeE l t where > class Time l t => TimeTO l t where > > > instance SetClass (TimeFacts t) where -- OK > instance (SetClass t) => Time (TimeFacts t) t where > instance (SetClass t) => TimeE (TimeFacts t) t where > instance (SetClass t) => TimeTO (TimeFacts t) t where > > > > _______________________________________________ > Beginners mailing list > Beginners(a)haskell.org > http://mail.haskell.org/cgi-bin/mailman/listinfo/beginners > > -- This email originated from DIT. If you received this email in error, please delete it from your system. Please note that if you are not the named addressee, disclosing, copying, distributing or taking any action based on the contents of this email or attachments is prohibited. www.dit.ie Is ó ITBÁC a tháinig an ríomhphost seo. Má fuair tú an ríomhphost seo trí earráid, scrios de do chóras é le do thoil. Tabhair ar aird, mura tú an seolaí ainmnithe, go bhfuil dianchosc ar aon nochtadh, aon chóipeáil, aon dáileadh nó ar aon ghníomh a dhéanfar bunaithe ar an ábhar atá sa ríomhphost nó sna hiatáin seo. www.dit.ie Tá ITBÁC ag aistriú go Gráinseach Ghormáin – DIT is on the move to Grangegorman <http://www.dit.ie/grangegorman>

1 0