- Haskell-Cafe - Haskell.org

ANNOUNCE: Ajhc Haskell Compiler 0.8.0.7 Release
by Kiwamu Okabe 09 Jul '13

09 Jul '13

We are happy to announce Ajhc 0.8.0.7. You can program interrupt handler with Haskell language on this release. But not yet collect (big) patch sets, the changes will be merged to jhc. You can get Ajhc using "cabal install ajhc" command. The usage is found at Ajhc's project web site http://ajhc.metasepi.org/. The source code at https://github.com/ajhc/ajhc/tags. Welcome sending any bugs or your ideas to https://github.com/ajhc/ajhc/issues. ## An example of interrupt handler written with Haskell <https://github.com/ajhc/demo-cortex-m3/tree/master/stm32f3-discovery> The demo for Cortex-M4 has main context and intrrupt context. The main context waits time expire with polling counter. <https://github.com/ajhc/demo-cortex-m3/blob/master/stm32f3-discovery/hs_src…> The interrupt context is called from clock exception, and decrement counter. <https://github.com/ajhc/demo-cortex-m3/blob/master/stm32f3-discovery/hs_src…> ## Other changes * Guard StablePtr critical section. * Add _JHC_JGC_SAVING_MALLOC_HEAP option for getting smaller malloc heap. * Link forkIO to forkOS. Enjoy! :) - - - Metasepi team

5 6

Linux users needed for OpenGL extensions survey
by Brian Lewis 09 Jul '13

09 Jul '13

Hi, I'm doing a survey to find out how well various OpenGL extensions are supported, to know where to focus efforts on Haskell game software. If you run Linux on your desktop and want to help, here's how: 1.) Save the information like this: $ glxinfo > SOMENAME.glxinfo ... where SOMENAME is your name, or nickname, or the computer's name, whatever you like. If you don't have glxinfo, you might need to install mesa-utils on Ubuntu, or mesa-demos on Arch. 2.) Ensure the file contains information about OpenGL extensions supported by your graphics card. 3.) Attach it to me *off list*. If you have questions, please ask me *off list*. If there's interest, I'll make the survey results available. Thanks!

3 2

Announcing postgresql-libpq-0.8.2.3
by Leon Smith 09 Jul '13

09 Jul '13

I just fixed a fairly serious performance problem with postgresql-libpq's binding to PQescapeStringConn; in was exhibiting a non-linear slowdown when more strings are escaped and retained. https://github.com/lpsmith/postgresql-libpq/commit/adf32ff26cdeca0a12fa5965… If you are using postgresql-libpq's escapeStringConn, or a library that uses it (e.g. postgresql-simple, or persistent-postgresql), I do recommend upgrading. You may or may not see a performance improvement, depending on your particular use case, but if you do it can be quite substantial. It's not entirely clear to me what the root cause really is, but it certainly appears as though it's related to the (direct) use of mallocBytes, which was replaced with (indirect) calls to mallocForeignPtrBytes / mallocPlainForeignPtrBytes (through the bytestring package). In this case, it resulted in an asymptotic improvement in time complexity of some algorithms. Best, Leon

2 2

same function's type accepted in top level, but rejected in where clause
by Ömer Sinan Ağacan 09 Jul '13

09 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 08 Jul '13

08 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 06 Jul '13

06 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 06 Jul '13

06 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 06 Jul '13

06 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