- Haskell-Cafe - Haskell.org

same function's type accepted in top level, but rejected in where clause
by Ömer Sinan Ağacan 08 Jul '13

08 Jul '13

Hi all, I came across an interesting type checker error, a function is accepted in top level, but rejected by type checker when moved to `where ...` clause. I moved required code to a new module for demonstration purposes: module Bug where fix :: (a -> a) -> a fix f = let x = f x in x data Fix f = Fix (f (Fix f)) unFix :: Fix f -> f (Fix f) unFix (Fix a) = a data ListF a l = NilF | ConsF a l instance Functor (ListF a) where fmap _ NilF = NilF fmap f (ConsF a l) = ConsF a (f l) fold :: Functor f => (f a -> a) -> Fix f -> a fold f a = f (fmap (fold f) (unFix a {- f (Fix f) -})) unfold :: Functor f => (a -> f a) -> a -> Fix f unfold f a = Fix (fmap (unfold f) (f a)) Now, after this code, type checker accept this function: iterateListF :: (a -> a) -> a -> Fix (ListF a) iterateListF fn e = unfold (foldFn fn) e foldFn :: (a -> a) -> a -> ListF a a foldFn fn a = ConsF a (fn a) But rejects this: iterateListF :: (a -> a) -> a -> Fix (ListF a) iterateListF fn e = unfold f e where f :: a -> ListF a a f a = ConsF a (fn a) With error: bug.hs:27:20: Couldn't match expected type `a1' with actual type `a' `a1' is a rigid type variable bound by the type signature for f :: a1 -> ListF a1 a1 at bug.hs:26:10 `a' is a rigid type variable bound by the type signature for iterateListF :: (a -> a) -> a -> Fix (ListF a) at bug.hs:23:17 In the return type of a call of `fn' In the second argument of `ConsF', namely `(fn a)' In the expression: ConsF a (fn a) Changing type variables in type of `f` to `x` fails with this error: bug.hs:28:20: Couldn't match expected type `x' with actual type `a' `x' is a rigid type variable bound by the type signature for f :: x -> ListF x x at bug.hs:27:10 `a' is a rigid type variable bound by the type signature for iterateListF :: (a -> a) -> a -> Fix (ListF a) at bug.hs:24:17 In the return type of a call of `fn' In the second argument of `ConsF', namely `(fn a)' In the expression: ConsF a (fn a) Failed, modules loaded: none. .. and this is strange because error message describes function as it is before changing `a` to `x`. Any ideas why this definition rejected? Is this a bug in GHC? --- Ömer Sinan Ağacan http://osa1.net

4 4

Catch multiple exceptions using 'Control.Exception'
by Nikita Karetnikov 07 Jul '13

07 Jul '13

I'm trying to update a package that uses 'Control.OldException' (works with GHC 6.10.4). Here is a relevant (and simplified) part: import Control.OldException -- | A predicate matching synchronous exceptions. syncExceptions :: Exception -> Maybe Exception syncExceptions (AsyncException _) = Nothing syncExceptions e = Just e throwAsync :: IO a throwAsync = throwIO $ AsyncException StackOverflow throwArith :: IO a throwArith = throwIO $ ArithException DivideByZero 'syncExceptions' is usually used like this: *Main> tryJust syncExceptions throwArith Left divide by zero *Main> tryJust syncExceptions throwAsync -- pass through *** Exception: stack overflow The above doesn't work with GHC 7.6.3 because 'Control.OldException' [1] was removed. And 'Control.Exception' doesn't have the 'Exception' type. Is there a way to adapt 'syncExceptions' to work with 'Control.Exception'? [1] http://hackage.haskell.org/packages/archive/base/4.1.0.0/doc/html/Control-O…

4 6

Parsec : parse either an interger or a double?
by Fredrik Karlsson 07 Jul '13

07 Jul '13

Dear list, Sorry for asking you another parsec question. I have a text file that is the output of a C/C++ program, involving time references in double format. Now, the program seems to truncate the output if possible, so a file that is one second long will produce this file: File type = "ooTextFile" Object class = "TextGrid" xmin = 0 xmax = 1 .... and a file that is 1.1 seconds long will be this: File type = "ooTextFile" Object class = "TextGrid" xmin = 0 xmax = 1.1 ... Now, how do I handle this case in parsec? I have tried this: start <- try (float haskellDef) <|> fmap ( / 1.0 ) $ integer haskelldef and similar solutions, but I cant get it to accept the 0. /Fredrik -- "Life is like a trumpet - if you don't put anything into it, you don't get anything out of it."

1 1

ANN: HaTeX 3.6. Now with Babel, TikZ and more!
by Daniel Díaz Casanueva 06 Jul '13

06 Jul '13

Hello everyone! It is time for a new release of HaTeX, the Haskell LaTeX library. http://hackage.haskell.org/package/HaTeX-3.6 == What is HaTeX? == HaTeX contains numerous types and functions that will assist you in the process of creating a LaTeX document within Haskell. You can also include your manually created LaTeX code in HaTeX, so you can choose the more convenient tool you want to use for each case. == What's new in this release? == A lot of changes have been made since the last release! The most important changes are explained in the blog post I just wrote: http://deltadiaz.blogspot.com.es/2013/06/hatex-36-texy-class-babel-fontenc-… However, the new feature I am more excited with is the TikZ script generator. Using a very intuitive interface, you will be able to generate high quality graphics for LaTeX documents. It is still far from be complete, but next releases will fill the gaps that you may find right now. Just to give you a taste, consider the following code: myFigure :: Figure myFigure = Scale 2 $ Figures [ RectangleFilled (0,0) 1 1 , Colored Green $ RectangleFilled (-1,1) 1 1 , Colored Red $ RectangleFilled ( 0,2) 1 1 , Colored Blue $ RectangleFilled ( 1,1) 1 1 ] With this simple construction you already have the following picture ready to be inserted in your LaTeX file output. http://daniel-diaz.github.com/projects/hatex/tikzsimple.png In the examples directory contained in the source distribution there is also an example (tikz.hs) that includes the plotting of a Haskell function (sine). However, the interface it uses may be less comfortable than the one shown in the example above. If you are a HaTeX user, don't miss this update, because it also contains some very important bug fixes. Any feedback gladly accepted. Good luck, Daniel Díaz.

8 14

Postdocs and Research Programmer for Compositional Learning via Generalized Automatic Differentiation
by Barak A. Pearlmutter 05 Jul '13

05 Jul '13

We are adding exact first-class derivative calculation operators (Automatic Differentiation or AD) to the lambda calculus, and embodying the combination in a production-quality fast system suitable for numeric computing in general, and compositional machine learning methods in particular. To the programming language community, we seek to contribute a way to make numeric software faster, more robust, and easier to write. To the machine learning community, in addition to the above practical benefits, we seek to contribute a system that embodies *compositionality*, in that gradient optimisation can be automatically and efficiently performed on systems themselves consisting of many components, even when such components may internally perform optimisation. (Examples of this include, say, optimisation of the rules of a multi-player game to cause the players actions to satisfy some desiderata, where the players themselves optimise their own strategy with using a simple model of the opponent which they optimise according to their opponent's behaviour; or multi-agent learning where one agent learns an internal model of another agent, where that internal model itself performs learning.) To this end, we are seeking two postdoctoral researchers and one research programmer with interest and experience in a cohert subset of: programming language theory, numerics, automatic differentiation, and machine learning. Inquiries to: Barak A. Pearlmutter <barak(a)cs.nuim.ie> Informal announcment with more details: http://www.bcl.hamilton.ie/~barak/ad-fp-positions.html, which will have a reference to the formal announcement when it becomes available. -- Barak A. Pearlmutter Hamilton Institute & Dept Computer Science NUI Maynooth

1 0

Basic Parsec float & integer parsing question
by Fredrik Karlsson 05 Jul '13

05 Jul '13

Dear list, Sorry for asking a simple parsec question, but both Parsec and Haskell is new to me, so please be gentle :-) I have this code: ------------------------------------------------------------------------------------------------ import Text.ParserCombinators.Parsec import Text.Parsec.Token import Text.ParserCombinators.Parsec.Char data VariableLine = VariableLine String String deriving Show data TierType = IntervalTier | PointTier deriving Show data Tier = Tier TierType String Float Float Integer data Label = Interval Float Float String data LabelFile = LabelFile Float Float symbol :: Parser Char symbol = oneOf "!#$%&|*+-/:<=>?@^_~" testString = "intervals [1]:\n xmin = 0 \n xmax = 0.028199999999999992 \n text = \"\"" headTS1 = "File type = \"ooTextFile\"\nObject class = \"TextGrid\"\n\nxmin =" header :: Parser LabelFile header = do headTS1 start <- float string "\nxmax = " end <- float string "\ntiers? <exists>\nsize = " integer char '\n' return $ LabelFile start end ------------------------------------------------------------------------------------------------ Loading it into ghci I get : Prelude> :l parsectest.hs [1 of 1] Compiling Main ( parsectest.hs, interpreted ) parsectest.hs:21:9: Couldn't match type `[]' with `Text.Parsec.Prim.ParsecT String () Data.Functor.Identity.Identity' Expected type: Text.Parsec.Prim.ParsecT String () Data.Functor.Identity.Identity Char Actual type: [Char] In a stmt of a 'do' block: headTS1 In the expression: do { headTS1; start <- float; string "\ \xmax = "; end <- float; .... } In an equation for `header': header = do { headTS1; start <- float; string "\ \xmax = "; .... } parsectest.hs:22:18: Couldn't match expected type `Text.Parsec.Prim.ParsecT String () Data.Functor.Identity.Identity Float' with actual type `GenTokenParser s0 u0 m0 -> Text.Parsec.Prim.ParsecT s0 u0 m0 Double' In a stmt of a 'do' block: start <- float In the expression: do { headTS1; start <- float; string "\ \xmax = "; end <- float; .... } In an equation for `header': header = do { headTS1; start <- float; string "\ \xmax = "; .... } parsectest.hs:24:16: Couldn't match expected type `Text.Parsec.Prim.ParsecT String () Data.Functor.Identity.Identity Float' with actual type `GenTokenParser s1 u1 m1 -> Text.Parsec.Prim.ParsecT s1 u1 m1 Double' In a stmt of a 'do' block: end <- float In the expression: do { headTS1; start <- float; string "\ \xmax = "; end <- float; .... } In an equation for `header': header = do { headTS1; start <- float; string "\ \xmax = "; .... } parsectest.hs:26:9: Couldn't match expected type `Text.Parsec.Prim.ParsecT String () Data.Functor.Identity.Identity a0' with actual type `GenTokenParser s2 u2 m2 -> Text.Parsec.Prim.ParsecT s2 u2 m2 Integer' In a stmt of a 'do' block: integer In the expression: do { headTS1; start <- float; string "\ \xmax = "; end <- float; .... } In an equation for `header': header = do { headTS1; start <- float; string "\ \xmax = "; .... } Failed, modules loaded: none. I'm sure I'm doing something really stupid here, but I need help to get through this problem. I've used the predefined "letter" parser at other places in the code, so I can't understand why "float" and "integer" does not work. /Fredrik -- "Life is like a trumpet - if you don't put anything into it, you don't get anything out of it."

3 2

Extending Type Classes
by Frantisek Farka 05 Jul '13

05 Jul '13

Hello all, I was looking for my master thesis topic and my supervisor suggested an idea of extending class system so it enables refactoring Type Class hierarchy without affecting client source code which is using refactored classes. One example is Functor - Applicative - Monad problem and corresponding proposal [1]. But this proposal instead of allowing the change through extending Type Classes forces client code to prepare for the new class layout and then switch the classes to the new layout. My goal is rather to allow direct changes in class hierarchy without affecting client source code. I have found different proposals approaching this problem on HaskellWiki, some of them are overlapping, some of them refer each other. The most promising to me seems Default superclass instances proposal [2]. This one is somehow implemented in the Strathclyde Haskell Enhancement (SHE) [3] but I haven't found much reference or user experience really. So the reason why I write this email is to ask you for some tips where above mentioned problem occurs in real source code. I would like to investigate some real examples before designing some ad hoc changes to the Type Classes system. Besides that I'd appreciate anyone who has used default superclass instances in SHE to share his experience. And last but not least I am always grateful for any comments and suggestions. Best wishes Frantisek [1] http://www.haskell.org/haskellwiki/Functor_hierarchy_proposal [2] http://hackage.haskell.org/trac/ghc/wiki/DefaultSuperclassInstances [3] https://personal.cis.strath.ac.uk/conor.mcbride/pub/she/

1 0

newbie question about "Functional dependencies conflict between instance declarations:".....
by Nicholls, Mark 05 Jul '13

05 Jul '13

Hello, I largely don't know what I'm doing or even trying to do, it is a voyage into the unknown....but....if I go... > {-# LANGUAGE MultiParamTypeClasses #-} > {-# LANGUAGE FunctionalDependencies #-} > {-# LANGUAGE FlexibleInstances #-} > {-# LANGUAGE UndecidableInstances #-} > class Foo x y | x -> y, y -> x > instance Foo Integer Integer That seems to work....and my head seems to say...your created some sort of binary relation between 2 types...and made <Integer,Integer> a member of it... Something like that anyway.... Then I go.... > data Bar > instance Foo Bar x Error!....but I'm think I understand this....I can't claim that <Bar,x> is a member of Foo and <Integer,Integer> is member of Foo and preserve my functional dependencies, because <Bar,Integer> is now a member of Foo.. Bad programmer....... So how I naively go.... > class NotAnInteger a > instance (NotAnInteger x) => Foo Bar x I haven't declared integer to be "NotAnInteger"....so (in a closed world)....this would seem to exclude the contradiction....but... Functional dependencies conflict between instance declarations: instance Foo Integer Integer -- Defined at liam1.lhs:7:12 instance NotAnInteger x => Foo Bar x -- Defined at liam1.lhs:13:12 So i) I clearly don't understand something about the type system. ii) I don't know how to restrict type variables in instance declarations....i.e. how do I use the notion of "Foo" across different combinations of types, without them colliding. CONFIDENTIALITY NOTICE This e-mail (and any attached files) is confidential and protected by copyright (and other intellectual property rights). If you are not the intended recipient please e-mail the sender and then delete the email and any attached files immediately. Any further use or dissemination is prohibited. While MTV Networks Europe has taken steps to ensure that this email and any attachments are virus free, it is your responsibility to ensure that this message and any attachments are virus free and do not affect your systems / data. Communicating by email is not 100% secure and carries risks such as delay, data corruption, non-delivery, wrongful interception and unauthorised amendment. If you communicate with us by e-mail, you acknowledge and assume these risks, and you agree to take appropriate measures to minimise these risks when e-mailing us. MTV Networks International, MTV Networks UK & Ireland, Greenhouse, Nickelodeon Viacom Consumer Products, VBSi, Viacom Brand Solutions International, Be Viacom, Viacom International Media Networks and VIMN and Comedy Central are all trading names of MTV Networks Europe. MTV Networks Europe is a partnership between MTV Networks Europe Inc. and Viacom Networks Europe Inc. Address for service in Great Britain is 17-29 Hawley Crescent, London, NW1 8TT.

3 4

Product-Categories
by Lyndon Maydwell 05 Jul '13

05 Jul '13

Hi Café. I'm wracking my brain trying to figure out a simple, reasonably general, implementation for a category instance for pairs of categories. So far I've looked at [1], which seems great, but doesn't use the built-in category instance, and [2], which I'm just not sure about. Ideally I'd like to be able to express something like - instance (Category a, Category b) => Category (Product a b) where id = Product id id Product o1 o2 . Product i1 i2 = Product (o1 . i1) (o2 . i2) However, it all falls apart when I actually try to define anything. Is this possible? If not, why not? As far as I can tell the issue boils down to not being able to translate "Category i o" to "Product (Fst i) (Fst o) (Snd i) (Snd o)" without breaking the kind expectation of the category instance. Please help me, I'm having a bad brain day :-) [1] - http://twanvl.nl/blog/haskell/categories-over-pairs-of-types [2] - http://hackage.haskell.org/packages/archive/categories/1.0.6/doc/html/Contr…

2 2

Automatically test different versions of single dependency
by Daniel Díaz Casanueva 04 Jul '13

04 Jul '13

Hello everyone. I would like to check what dependencies is one of my packages compatible with. For example, say I have a package called "foo" that depends on package "bar". Most likely, "foo" does not build with each version of "bar". What I want to do is try to build "foo" with each single version of "bar" (not manually). What is the best approach to this?

2 3