- Haskell-Cafe - Haskell.org

Numerics & implementing different instances of the same class
by George Pollard 13 Dec '08

13 Dec '08

Is there a good way of doing this? My running example is Monoid: > class Monoid a where > operation :: a -> a -> a > identity :: a With the obvious examples on Num: > instance (Num a) => Monoid a where > operation = (+) > identity = 1 > > instance (Num a) => Monoid a where > operation = (*) > identity = 0 Of course, this won't work. I could introduce a newtype wrapper: > newtype (Num a) => MulNum a = MulNum a > newtype (Num a) => AddNum a = AddNum a > > instance (Num a) => Monoid (MulNum a) where > operation (MulNum x) (MulNum y) = MulNum (x * y) > identity = MulNum 1 > > instance (Num a) => Monoid (AddNum a) where ... -- etc However, when it comes to defining (e.g.) a Field class you have two Abelian groups over the same type, which won't work straight off: > class Field a where ... > instance (AbelianGroup a, AbelianGroup a) => Field a where ... Could try using the newtypes again: > > instance (AbelianGroup (x a), AbelianGroup (y a) => Field a where ... ... but this requires undecidable instances. I'm not even sure if it will do what I want. (For one thing it would also require an indication of which group distributes over the other, and this may restore decidability.) I'm beginning to think that the best way to do things would be to drop the newtype wrappers and include instead an additional parameter of a type-level Nat to allow multiple definitions per type. Is this a good way to do things? Has anyone else done something similar? I've taken a look at the Numeric Prelude but it seems to be doing things a bit differently. (e.g. there aren't constraints on Ring that require Monoid, etc) - George

3 2

wlan library
by Cetin Sert 13 Dec '08

13 Dec '08

Hi *^o^*, I am writing a network manager <http://sert.homedns.org/hnm/> as a replacement for some broken, already existing knetworkmanager for a friend's computer. I was looking for some haskell libraries that provide access to wlan cards but could not find any on hackage. Maybe I missed something. Currently my tool is using linux the commands: modprobe, ifconfig, iwconfig, iwlist, dhclient etc.. Can a library calling external stand-alone linux commands be considered good enough as is? Or should one definitely write a wrapper around a more low-level, stable, POSIX-compliant foreign library to access and configure, control WLAN cards? I'd love to hear your comments. Best Regards, Cetin Sert

2 1

Parsec and (then?) type-check
by Greg Fitzgerald 13 Dec '08

13 Dec '08

Parser gurus, When you write a parser with a library like Parsec, do you typically type-check while parsing, or afterward in a separate pass? The latter is more modular, but it means labeling every element in the AST with the parser position so that you can give good error messages. Do you find the added modularity worth the hassle or just pack type-checking into the parser pass? Thanks, Greg

2 1

Memoization-question
by Mattias Bengtsson 13 Dec '08

13 Dec '08

The program below computes (f 27) almost instantly but if i replace the definition of (f n) below with (f n = f (n - 1) * f (n -1)) then it takes around 12s to terminate. I realize this is because the original version caches results and only has to calculate, for example, (f 25) once instead of (i guess) four times. There is probably a good reason why this isn't caught by the compiler. But I'm interested in why. Anyone care to explain? > main = print (f 27) > > f 0 = 1 > f n = let f' = f (n-1) > in f' * f' (compiled with ghc --make -O2) Mattias

5 8

Data.Array.Storable vs GC
by David F. Place 12 Dec '08

12 Dec '08

If you have a giant unboxed array that will never become garbage, it would be nice to put it somewhere where the GC won't bother with it. Since Data.Array.Storable arrays are allocated in the C heap, I thought it would be a good choice. However, I am getting very poor performance due to the GC copying 6G in each run. The only explanation that I can think of is that it is copying my giant array. Am I wrong in my understanding that Data.Array.Storable arrays should not be copied by the GC? BTW, I am using GHC 6.10.1 on linux. Thanks for reading.

4 6

Multi-parameter Type Class
by Paul Keir 12 Dec '08

12 Dec '08

Hi all, I've been trying to refactor my tree conversion code to make better use of type classes; and I've discovered multi-parameter type classes and functional dependencies. I have a class with a function a2b, and I'd like "map" to be used when it's a list of type a. I've created a simple failing example: data Foo = Foo Bar deriving(Show) data Bar = Bar String deriving(Show) class ZOT a b | a -> b where zot :: a -> b instance ZOT Foo Integer where zot x = 17 instance ZOT Bar String where zot x = "Eighteen" instance ZOT [x] [y] where -- This bit zot xs = map zot xs -- fails main = do print $ zot $ Foo $ Bar "Blah" print $ zot $ Bar "Blah" print $ zot $ [Bar "Blah", Bar "Blah"] -- No map here please I know this would work if the third instance of zot explicitly took [Bar] and [String]. Can I not instead generalise for all the ADTs in my tree in the way I've outlined? Must I instantiate for the type of each list pair? Cheers, Paul

5 6

Re: problems with hs-plugins load-eval (possible bug)
by Alberto G. Corona 12 Dec '08

12 Dec '08

The duplicate definition error appears when I compile Main.hs and execute it. But when I run it with runghc the behaviour is different. It works well: > runghc Main.hs 3 Any idea? is this a bug of hs-plugins? it is just something expected?? 2008/12/6 Alberto G. Corona <agocorona(a)gmail.com> > I have a web server which load server extensions. these extensions > eval-uate configuration files that contains code (user-editable workflow > descriptions). The problem is that I need common definitions (inside > imported modules) for the extensions and for the configuration files. This > is not permitted by ha-plugins. > > The minimal code example are the files below. main loads eval.hs , that > evaluate a expression. The common definitions are in Include.hs. The error > is: > > *GHCi runtime linker: fatal error: I found a duplicate definition for > symbol > Include_sum1_srt > whilst processing object file > /home/magocoal/haskell/devel/votesWorkflow/src/unused/tests/Include.o > > This could be caused by: > * Loading two different object files which export the same symbol > * Specifying the same object file twice on the GHCi command line > * An incorrect `package.conf' entry, causing some object to be > loaded twice. > GHCi cannot safely continue in this situation. Exiting now. Sorry. > * > > > ** > Do you kno how to solve the problem while maintaining the functionality? > > -------Include.hs------- > module Include where > > sum [x,y]= x+y > > > ------Main.hs----- > module Main > where > > import Include > import System.Plugins > > main= do > s <-loadExec "eval.o" "mainc" > print s > > loadExec:: String-> String->IO String > loadExec file method = do > > mv <- load file ["."] [] method > case mv of > LoadSuccess mod v -> v :: IO String > LoadFailure msg -> return $ concat msg > > > > ------------Eval.hs-------- > > module Eval(mainc) where > import System.IO.Unsafe > import System.Eval.Haskell > > mainc= do i <- unsafeEval_ "sum1 [1,2]" ["Include"] [] []["."] :: IO > (Either [String] Int) > return $ show i > > > > >

2 2

Haskell project proposals reddit
by Don Stewart 12 Dec '08

12 Dec '08

I'd like to echo Jason's remarks earlier. http://www.reddit.com/r/haskell_proposals/ We've tried for a couple of years now to efficiently track 'wanted libraries' for the community, but never with much success. In particular, two approaches have been tried: * a wiki page * the 200{6,7,8} summer of code trac neither proved workable long term, likely because no one knew about them, they're harder to contribute to and other factors. Now I know there's a lot of spare Haskell capacity in this community. 3000 mailing list readers, but only 70 new libraries a week being uploaded to Hackage --- that's not how we take over the world! So this is your chance to contribute: * propose new libraries, and explain why you'd want them * vote on things you'd like to see * pick up tasks that sound interesting Let's try to make this work! -- Don

9 10

unsafeThawByteArray#
by Patrick Perry 12 Dec '08

12 Dec '08

I've noticed that there is no longer an "unsafeThawByteArray#" function in GHC.Prim. Looking through the darcs history, I found a patch with description: [project @ 2000-03-13 12:11:43 by simonmar] simonmar**20000313121144 Remove unsafeThawByteArray# primop (which was a no-op), and use unsafeCoerce# instead. ] Is the following code safe? data IntArray = IntArray !Int (ByteArray#) data STIntArray s = STIntArray !Int (MutableByteArray# s) unsafeThaw :: IntArray -> ST s (STIntArray s) unsafeThaw (IntArray n arr#) = ST $ \s1# -> let coerceArray :: State# s -> MutableByteArray# s coerceArray _ = unsafeCoerce# arr# marr# = coerceArray s1# marr = STIntArray n marr# in (# s1#, marr #) Thanks, Patrick

1 0

Re: [Haskell-cafe] A curios monad
by Andrew Coppin 12 Dec '08

12 Dec '08

Alberto G. Corona wrote: > but because keys and data must be of the same type, you are restricted > to instances of Eq and Ord. Itsn´n true? > > 2008/12/11 Andrew Coppin <andrewcoppin(a)btinternet.com > <mailto:andrewcoppin@btinternet.com>> > > Alberto G. Corona wrote: > > Well, you have not to use typeable, but instead you need to > use instances of Eq and Ord. This is because the values and > the keys must be of the same type. this is the tradeoff. > > > Not really, no. The *keys* need Eq and Ord, but the values I'm > storing don't. The values I'm storing don't need any specific > properties at all. That's what I like about it. > > No. Keys are of type "Key v", where "v" can be anything. It's a phantom type, so it's not used for anything (except knowing what to coerce the value back to when it's looked up). "Key v" is just an alias to Int, which *is* in Eq and Ord, as required. There are no constraints on "v" at all.

1 0

problems configuring hs-plugins 1.3.1 in windows
by Alberto G. Corona 12 Dec '08

12 Dec '08

I have problems compiling plugins 1.3.1 in windows at the configure phase configure: error C compiler cannot create executables I use ghc 6.8.2 I know that plugins is nor much tested under windows but I´m trying anyway. I think that this problem is very common also in Linux. I tried to find references to this error ut all of then are old (for plugins 1.2) and it is supposed to be solved. Do you have any workaround? By the way the reported compiler used is gcc Responder Responder a todos Reenviar

1 0

Re: [Haskell-cafe] A curios monad
by Andrew Coppin 12 Dec '08

12 Dec '08

Alberto G. Corona wrote: > it is safe to use unsafeCoherce in your context? I mean, I think you > need to use instances of Data.Typeable (or some with the same > functionality) to make it safe. Like in Data.Dynamic... The *only* way for you to generate a Key value is to use the function I provide - which conviniently tags it with the type of value it's supposed to retreive. That means the type cast is guaranteed to cast it back to the type it was in the first place. ...unless you can muddle keys from other state threads up. In that case, all bets are off. I looked at using Data.Dynamic, but that only works for a few types. I wanted something that will work for *any* type.

1 0

type error
by Cetin Sert 12 Dec '08

12 Dec '08

Hi, Why does this not function? Prelude> sequence [print 'a', print 2] 'a' 2 [(),()] *Prelude> let myprint = print* *Prelude> sequence [myprint 'a', myprint 2]* <interactive>:1:18: Couldn't match expected type `()' against inferred type `Char' In the first argument of `myprint', namely 'a' In the expression: myprint 'a' In the first argument of `sequence', namely `[myprint 'a', myprint 2]' Can providing some type annotations (interactively in ghci or in some .hs file) help solve the problem? Best Regards, CS

4 3

Can my type be allowed as return type in FFI?
by Mauricio 11 Dec '08

11 Dec '08

Hi, When I do: foreign import "nameOfFunction" nameOfFunction :: IO MyType I can get a function that return MyType only if it's a pointer or some of the C* type family. Is it possible to write a new MyType and make it allowed as a return type from foreign functions? Is changing the compiler the only way to do that? Thanks, Maurício

5 8

how to pipe in pure input with HSH? (Would be helpful for sending emails via sendmail/mailx via HSH interface)
by Thomas Hartman 11 Dec '08

11 Dec '08

{- I am trying to figure out a way to send email using HSH and I'm stumped. The problem is that there is no ShellCommand instance for a pure vanilla string, which can be piped into another ShellCommand. There is a ShellCommand String instance, but the string is a command to be executed. I defined PureInput wrapper around String in an attempt to get this working, but I'm stumped defining an fdInvoke method for it. class (Show a) => ShellCommand a where fdInvoke :: a -> System.Posix.Types.Fd -> System.Posix.Types.Fd -> [System.Posix.Types.Fd] -> IO () -> IO [InvokeResult] Can someone help out? Below is a stub. Thomas. -} import HSH newtype PureInput = PureInput { unpureinput :: String } deriving (Read,Show) -- This works fine, blah blah blah gets outputted demo1 = runIO $ ( ( ( ("echo blah blah blah") :: String) -|- ( "cat" :: String) ) :: PipeCommand String String ) -- This is what I want. Specify a pure input string, to be piped into another command. -- In this example it's cat, which isn't very usefuol. -- This would be useful, however, for sending email via sendmail or mailx, with variable input piped in. -- The result should be that "blah blah blah" is printed from cat, just as in demo1. demo2 = runIO $ (PureInput "blah blah blah") -|- ( "cat" :: String) -- How can/should this be done? instance ShellCommand PureInput where fdInvoke (PureInput justAString) ifd ofd closefd forkfunc = undefined

2 1

Retrospective on 2008?
by Don Stewart 11 Dec '08

11 Dec '08

I've been kicking the idea around of a 2008 retrospective. How did we do this year? After all, it has been a dramatic growth period (we're close to 600 people a day in #haskell now, for example, and many new faces!) Some ideas: * 10 best new libraries * 10 best new apps * 10 most influential blog posts * best new library by category Determing this is an interesting task. To start with, I scanned hackage. Some stats: * Hackage currently holds 914 applications and libraries. * 601 of them were released on Hackage this year: http://galois.com/~dons/tmp/new-releases-2008.txt Which is rather impressive, I have to say. We're getting better at the production of new libraries and apps (I imagine all the releases are feeding into new apps and libs in new areas not easily reachable before). The last couple of months, we're averaging around 10 uploads a day, in fact. You can see the upload growth here, http://galois.com/~dons/images/hackage-daily-graph.png We've fundamentally transformed the suitability of Haskell and GHC in a range of domains through the production of these apps and libraries. Well done everyone! Now, as for determining which new things make the top 10, a chance to apply some heuristics? (Most voted on? Most downloaded? Most depended upon by other things?). Any thoughts? Anyone have their own top 10 list? Or other key events of the year? -- Don

4 5

Projects that depend on the vty package?
by Corey O'Connor 11 Dec '08

11 Dec '08

Hello, For further development of the vty package I'm really only paying attention to the requirements that fall out of the Yi project. Are there any other projects that depend on the vty package? In addition, the vty project has it's own wiki: http://trac.haskell.org/vty/ Right now there isn't much information there but it is a great place to send bug reports or enhancement requests if you have them. Cheers, Corey O'Connor

4 5

Re: Haskell-Cafe Digest, Vol 64, Issue 17
by Artyom Shalkhakov 11 Dec '08

11 Dec '08

Hello, 2008/12/10 Ben <midfield(a)gmail.com>: > do you have any plans for iteratee based IO? I would like to use it in XHB (X Haskell Bindings). We use ByteStrings ATM, but I find the code hard to read. > have you seen takusen? Actually, I only took a glance or two. Regards, Artyom Shalkhakov.

1 0

Source code
by Andrew Coppin 11 Dec '08

11 Dec '08

The library documentation pages on Haskell.org used to have helpful links to the library source code. But they don't seem to be there any more. Anybody know why?

1 0

ANNOUNCE: haskell-src-exts 0.4.4
by Niklas Broberg 11 Dec '08

11 Dec '08

Fellow Haskelleers, it is my pleasure to announce the new release of the haskell-src-exts package, version 0.4.4: http://hackage.haskell.org/cgi-bin/hackage-scripts/package/haskell-src-exts… darcs get http://code.haskell.org/HSP/haskell-src-exts The new feature in this release is support for pragmas. haskell-src-exts-0.4.4 supports all pragmas supported by GHC, with the exception of option-level pragmas (LANGUAGE, OPTIONS_GHC etc) that appear before the module header. The reason these are not yet supported is simply time, there's no real difficulty involved in supporting them too and I will surely get there eventually. 0.4.4 is backwards incompatible with 0.4.3 for two constructors: * The Module constructor (:: Module) now has an extra argument of type 'Maybe WarningText', used for deprecated modules. * The ImportDecl constructor (:: ImportDecl) now has an extra argument of type Bool, stating whether the SOURCE pragma has been used for the import. The full list of pragmas supported by 0.4.4 is: SOURCE, RULES, DEPRECATED, WARNING, INLINE, NOINLINE, SPECIALISE, CORE, SCC, GENERATED and UNPACK. Cheers and Happy Haskelling, /Niklas

7 16

Re: Iteratee-based IO and lightweight monadic
by John Lato 11 Dec '08

11 Dec '08

> From: "Artyom Shalkhakov" <artyom.shalkhakov(a)gmail.com> > > Hello, > > Is anybody planning to use these shiny new ways for doing IO? > > I'm also interested in a fair comparison of ByteString/Binary > with iteratee-based IO. Has anybody done this? > I would like to use Iteratee-based IO, and have been doing some experiments with it recently for my problem domain (reading sound files). There has been a discussion recently on the haskell-art list about this specific topic. I posted some timing/memory data, and links to my test code, which you should be able to find in the archives if you're interested. I have found that, with my specific implementations and test case, Iteratee-based IO is slightly slower than the most performant ByteString/Binary code I used, however it uses significantly less memory and has nice composability. Also note that my ByteString-based reader actually reads strict ByteStrings and outputs a stream of UArr Doubles (from the uvector package) using unsafeInterleaveIO rather than using lazy ByteStrings directly. Other implementations made different choices, which I covered in some detail in the mentioned thread. Oleg has recently (after I did my tests) published a small library for reading TIFF files using Iteratees. This new code offers some significant advantages over what I was doing with Iteratees, and may offer performance benefits. More work for me... Unfortunately, neither approach (Iteratee nor ByteString) is yet competitive with hsndfile/libsndfile. However, an Iteratee layer that uses hsndfile to read could be implemented, giving the benefits of the Iteratee approach with high performance. I believe the hsndfile authors are interested in this. It should be very easy to implement.

1 0

Re: [Haskell-cafe] File name encodings
by Don Stewart 10 Dec '08

10 Dec '08

Oh, perhaps you want to 'decode' the string that dirOpenDialog returns. redcom: > Hi Don, > > must be doing something wrong. > > The messed up string originates from calling Graphics.UI.WX.dirOpenDialog > and selecting a directory with Umlauts. > > The encode function doesn't seem to help here as it's too far at the end > of the chain. > > Günther > > Am 10.12.2008, 02:45 Uhr, schrieb Don Stewart <dons(a)galois.com>: > > >redcom: > >>Hi, > >> > >>how can I convert a file name with for example umlauts (ü, ä ... ) into > >>UTF8, or some other format that I can actually use it? > >> > >>Günther > >> > > > >Using the utf8-string package to encode the file name string. > > > > e.g. > > > > x <- readFile (encode "ኃይሌ ገብረሥላሴ") > > > > http://hackage.haskell.org/cgi-bin/hackage-scripts/package/utf8-string > >

3 2

Associated data types
by Eyal Lotem 10 Dec '08

10 Dec '08

If we have associated type synonyms, is there still reason to have associated data types? For example, we can replace: class C b where data D b ... instance C Int where data D Int = D | E with: class C b where type D b ... data DInt = D | E instance C Int where type D Int = DInt Or perhaps allow, for convenience, this form (which would desugar to the above): class C b where type D b ... instance C Int where data D Int = D | E

4 4

Translating RWH into other languages
by Bulat Ziganshin 10 Dec '08

10 Dec '08

Hello Don, There is a team of people which want to translate RWH book into Russian. is it ok? can you help us by establishing "subversion" of RWH book on your site. if it will contain copy of the English book for the beginning and allow team members to further edit it - it will be great! -- Best regards, Bulat mailto:Bulat.Ziganshin@gmail.com

1 0

Does ghc use Language.Haskell.*?
by Maurício 10 Dec '08

10 Dec '08

Hi, I've just coded a small utility to prettyprint my code, based on Language.Haskell.Parser and friends. Much to my surprise it wasn't able to parse itself, complaining about a non-ASC character (í). Why can ghc read the program, but not the standard library parser? Does ghc use something else? Is it possible to use whatever ghc uses to build a prettyprinter? Thanks, Maurício

4 3

a haskell_proposals subreddit
by Jason Dusek 10 Dec '08

10 Dec '08

There is a subreddit for people to propose libraries: http://www.reddit.com/r/haskell_proposals/ The idea being, that Web 2.0 will help us to allocate our collective talents more efficiently when it comes to extensions (and perhaps clue us in when our pet project is something people really want). -- _jsn

3 6

File name encodings
by Günther Schmidt 10 Dec '08

10 Dec '08

Hi, how can I convert a file name with for example umlauts (ü, ä ... ) into UTF8, or some other format that I can actually use it? Günther -- On a German Win XP

2 2

Building/installing haskelldb and friends with ghc 6.10.1
by Laurent Giroud 10 Dec '08

10 Dec '08

Hi everyone, I have been trying to install haskelldb in the last few days and encountered a number of hurdles which raised a few questions and for which I'd appreciate some insight and advice. Note that I am posting here because they are not specifically haskelldb related. I am using Cabal to install haskelldb, haskelldb-hsql, haskelldb-hsql- mysql (and others) and discovered that the packages available on hackage do not build properly with ghc 6.10.1. The typical error being of the following kind: Could not find module `System.Time': it is a member of package old-time-1.0.0.1, which is hidden These ones are relatively easy to fix by finding the locally cached .tar.gz and modifying the .cabal file to include the corresponding packages (there were several of them which were discovered in successive attempts). But I finally stumbled on hsql-mysql due to it apparently not being actively maintained anymore (this seemingly simple conclusion actually required quite a lot of research). I read on haskell-cafe that Frederik Eaton had been apparently fixing these problems but could find nowhere a version exhibiting the will to compile gracefully, a darcs repository at http://code.haskell.org/ HSQL/ apparently containing patches from him also proved reluctant to build, although for different reasons. I wasn't courageous enough to pursue and try to fix them at that point of the adventure. Other searches in haskell cafe lead me to people advising to use HDBC instead, after which I contemplated suicide for a few seconds before coming back to my senses and cabal-installing haskelldb-hdbc which quickly led me to error messages of the exact same kind mentioned above. Strangely enough, hdbc 1.1.5 itself installs quite fine using Cabal, however, haskell-hdbc insists on depending on version 1.0.1. I haven't been able so far to find what configuration file to modify in order for it to try to build with the latest Cabalized version (I admit I didn't search much). I then discovered this page http://www.haskell.org/haskellwiki/Upgrading_packages which hints that some of the existing packages in Hackage need "a bit" of work to be usable with ghc 6.10.1 as well as ghc 6.8.x apparently. HSQL and haskelldb-HDBC seem indeed to be in the list of packages which would require some effort in order to be usable with the latest ghc versions (can't remember where I read that though). Pursuing any further would seem to require to delve deeper in the intricacies of the Cabal system and fix either HSQL or HaskellDB-HDBC. Since I really want to use HaskellDB I'll probably try these avenues, but I would appreciate comments at this point from persons knowledgeable in the evocated areas: Are the databases interface for haskelldb still officially maintained and where is this information available ? Is the information that this and that package are compatible (or not) with ghc 6.8.x or 6.10.1 available somewhere without compiling first hand the packages ? It would be immensely valuable to the haskell newcomer to have the incompatible packages signaled on http://hackage.haskell.org/packages/archive/pkg-list.html for example. What would you recommend at this point : fix hsql, or haskelldb-hdbc so it uses 1.1.5 instead of 1.1, try an alternative to haskelldb (are there any ?) ? Thanks in advance for your comments and advices. Best regards, Laurent -- Laurent Giroud ml(a)niaow.com

2 2

scanr producer?
by Louis Wasserman 10 Dec '08

10 Dec '08

Is there a change in strictness with the following transformation? It seems like scanr should be a good consumer and a good producer, intuitively. scanrFB :: (a -> b -> b) -> b -> [a] -> (b -> c -> c) -> c -> c scanrFB f x ls c n = snd (foldr (\ x (b, bs) -> let b' = f x b in (b', b' `c` bs)) (x, n) ls) {-# RULES "scanr" [~1] forall f x l . scanr f x l = build (scanrFB f x l); -# RULES} -- Louis Wasserman wasserman.louis(a)gmail.com

1 0

FFI, TH or GHCI, foreign bindings and the linker(s)
by Andrea Rossato 10 Dec '08

10 Dec '08

Hello, I know that this is somehow a recurring question, still the archives have not been helpful for finding a working solution. I'm writing the bindings[1] to bibutils[2], a set of utilities for converting from and to different bibliographic databases (MODS, bibtex, etc.). bibutilis uses a static library (which is not installed) to build a set of different binaries which are the installed utilities. Suppose I build the library in /tmp/bibutils. If I build the haskell bindings (hs-bibutils) with: runhaskell Setup.lhs configure --extra-include-dirs=/tmp/bibutils/lib --extra-lib-dirs=/tmp/bibutils/lib where libbibutils.a and the header files are located, then I can build a simple test[3] program with ghc --make test.hs and it works perfectly. When I tried to link citeproc-hs to hs-bibutils everything worked fine, too. But when I tried building pandoc, which uses the Template Haskell extension to generate some code at compilation time, I got a linker error when TH started its process: [...] [ 6 of 29] Compiling Text.Pandoc.ODT ( Text/Pandoc/ODT.hs, dist/build/Text/Pandoc/ODT.o ) Loading package ghc-prim ... linking ... done. [...] Loading package template-haskell ... linking ... done. ghc: /tmp/bibutils_3.43/lib/bibutils.o: unknown symbol `fields_add' Loading package hs-bibutils-0.1 ... linking ... ghc: unable to load package `hs-bibutils-0.1' So I went back to the test file with the original bindings. If I try to load it and run it on ghci I get the very same error: Prelude Main> main Loading package syb ... linking ... done. Loading package hs-bibutils-0.1 ... linking ... <interactive>: /tmp/bibutils/lib/bibutils.o: unknown symbol `fields_add' ghc: unable to load package `hs-bibutils-0.1' After searching the we I found this: http://article.gmane.org/gmane.comp.lang.haskell.cafe/23635 which states that GHCi cannot load static libraries (the same can be argued by reading the GHCi docs, indeed). After reading this thread: http://article.gmane.org/gmane.comp.lang.haskell.cafe/40412 I came to know that GHCi and GHC are not using the same linker, and, as far as I understand, TH uses the first one too. So I tried to build a dynamic library. ar -t /tmp/bibutils/lib/libbibutils.a to get the list of objects to link with: gcc -shared -o libbibutils.so *.o after compiling them with the -fPIC flag. But this library is not being loaded because of some undefined symbols: lib/libbibutils.so: undefined reference to `corps' lib/libbibutils.so: undefined reference to `asis' lib/libbibutils.so: undefined reference to `progname' These symbols refer to some variables defined as external entities, for instance: extern char progname[]; these entities are initialized by a stub.c file in the Haskell bindings: http://code.haskell.org/~arossato/hs-bibutils/cbits/stub.c Unfortunately my knowledge of such low level stuff is very very limited and I'm not seeing any way out. Is there one? Thanks, Andrea [1] http://code.haskell.org/~arossato/hs-bibutils/ [2] http://www.scripps.edu/~cdputnam/software/bibutils/ [3] the test.hs file import Text.Bibutils main :: IO () main = do init_globals "mods2bibtex" bibl <- bibl_init param <- bibl_initparams mods_in bibtex_out setFormatOpts param [bibout_brackets, bibout_uppercase] setBOM param setVerbose param bibl_read param bibl "/tmp/prova.biblio" mods_in bibl_write param bibl "-" bibtex_out return ()

1 1

Timber mailing list
by ryah dahl 10 Dec '08

10 Dec '08

Is there a user mailing list for the Timber language? (http://www.timber-lang.org/)

1 0

Exact parsing of decimal numbers
by Emil Axelsson 09 Dec '08

09 Dec '08

Hello, I don't know enough about the RealFrac class to answer this myself: Can I be sure that Numeric.readFloat :: ReadS Rational never exhibits any rounding errors? Thanks, / Emil

2 1

Cabal: defaultMainNoRead and source file location
by Mauricio 09 Dec '08

09 Dec '08

Hi, I've just seen this from Distribution.ModuleName (ghc 6.10): toFilePath $ ( simple "A.B.C" ) to which ghci answers: "A.B.C". Shouldn't it say "A/B/C"? The reason why I'm asking is that I've just created a Setup.hs with 'defaultMainNoRead', and 'Setup build' complains it can't find A.B.C. When I copy (move doesn't work) C.hs from <src_dir>/A/B to <src_dir>/A.B.C.hs, it doesn't complain, although it fails to generate my library saying: [1 of 1] Compiling A.B.C ( src/A/B/C.hs, dist/build/A/B/C.o) /usr/bin/ar: creating dist/build/libHSabc-0.a /usr/bin/ar: dist/build/A.B.C.o: No such file or directory Thanks for your tips, Maurício

2 2

How to define Show [MyType] ?
by Dmitri O.Kondratiev 09 Dec '08

09 Dec '08

I am trying to define instance Show[MyType] so show (x:xs :: MyType) would return a single string where substrings corresponding to list elements will be separated by "\n". This would allow pretty printing of MyType list in several lines instead of one, as default Show does for lists. For example: data ShipInfo = Ship { name :: String, kind :: String, canons :: Int } deriving Show s1 = Ship {name ="HMS Fly", kind = "sloop", canons=16} s2 = Ship {name ="HMS Surprise", kind = "frigate", canons=42} -- Yet when I try to define: instance (Show ShipInfo) => Show [ShipInfo] where show (x:xs) = "<" ++ show x ++ ">" ++ show xs -- I get this error: Illegal instance declaration for `Show [ShipInfo]' (The instance type must be of form (T a b c) where T is not a synonym, and a,b,c are distinct type variables) In the instance declaration for `Show [ShipInfo]' Failed, modules loaded: none. -- On the other hand this definition: instance (Show a) => Show [a] where show (x:xs) = "<" ++ show x ++ ">" ++ show xs -- Gives a different error: Duplicate instance declarations: instance (Show a) => Show [a] -- Defined at C:/wks/haskell-wks/ShowMatrix.hs:37:0 instance (Show a) => Show [a] -- Defined in GHC.Show Failed, modules loaded: none. -- Note the last error implicitly tells us that defining Show [a] is possible in principle! -- In fact it already defined in GHC.Show !!! -- How to define Show [MyType] ? Thanks! Dmitri

9 15

Data.Map add only if not exist and return true if added
by John Ky 09 Dec '08

09 Dec '08

Hi, I'm looking for a function for Data.Map that will insert a new key and value into the map if the key doesn't already exist in the map. When the key already exists, I don't want the value updated in the map. Additionally, I want to know whether the key/value was inserted or not so that I can return that fact as a boolean from my function. I can't work out which function in Data.Map is suitable for this task and I wanted to avoid two lookups. Thanks, -John

2 1

Why does "instance Ord Pat" causes <<loop>>
by Martin Hofmann 09 Dec '08

09 Dec '08

I am storing the TH data types 'Exp' and 'Pat' in Maps and Sets. As a first attempt to quickly get rid of typechecker's complaints I defined some naive instances of Ord for Exp and Pat. Now it took me about a week to realise, that 'instance Ord Pat' causes ghc to loop. Apparently, there is a reason, why Pat and Exp are not instances of Ord. What is so bad about it? If Pat and Exp should not be instances of Ord, maybe a note in the source code or haddock would be helpful. On the other hand, what would argue against a lexicographic ordering (apart from the inefficient implementation of the following one)? Following some literate code to reproduce the <<loop>> (or stack overflow in GHCi), by commenting and uncommenting the appropriate lines: > {-# OPTIONS_GHC -fglasgow-exts -fth #-} > module Test where > import Language.Haskell.TH > import Data.Set ------------------- naive Ord > instance Ord Exp > instance Ord Pat ------------------- lexicographic Ord instance Ord Exp where compare l r = compare (show l) (show r) instance Ord Pat where compare l r = compare (show l) (show r) ------------------- > mkVP s = VarP $ mkName s > mkVE s = VarE $ mkName s > rule1 = (,) [mkVP "x_14"] (mkVE "y_14") > rule2 = (,) [InfixP (mkVP "x1_15") '(:) (mkVP "x_16")] (InfixE (Just (mkVE "y1_15")) (ConE '(:)) (Just (mkVE "ys_16"))) > stack_overflow = fromList [rule1,rule2] Thanks, Martin

4 5

Reading showables
by John Ky 09 Dec '08

09 Dec '08

Hi, Is there a way to read Showables? main = do putStrLn $ show $ read Thanks -John

6 8

Dr Dobbs: Time to get good at functional programming
by Jeff Heard 09 Dec '08

09 Dec '08

Seen on Slashdot: http://www.ddj.com/development-tools/212201710;jsessionid=3MQLTTYJRPL3CQSND…

6 8

Building "plugins" from Hackage
by Sol 09 Dec '08

09 Dec '08

Hello List, when I try to install the package "plugins" with cabal i get the following error. cabal: dependencies conflict: ghc-6.8.3 requires Cabal ==1.2.4.0 however Cabal-1.2.4.0 was excluded because plugins-1.3.1 requires Cabal ==1.4.* Is there a way to resolve this? Any ideas? Sol.

2 1

Overlapping instances
by John Ky 09 Dec '08

09 Dec '08

Hi, I've got the following code which tries to implement a TShow class, which is equivalent to Show, except it is supposed to work on TVar types as well. import GHC.Conc createEngine :: String -> Int -> Int -> IO Engine createEngine name major minor = do tUsers <- newTVarIO [] return $ Engine { engineName = name , version = EngineVersion { major = major , minor = minor } , users = tUsers } class TShow a where tshow :: a -> IO String instance Show (TVar a) where show a = "%" instance (Show a) => TShow a where tshow a = return $ show a instance (Show a) => TShow (TVar a) where tshow ta = do a <- atomically (readTVar ta) return $ show a data User = User { userName :: String } deriving Show data EngineVersion = EngineVersion { major :: Int , minor :: Int } deriving Show data Engine = Engine { engineName :: String , version :: EngineVersion , users :: TVar [User] } instance TShow Engine where tshow a = do users <- atomically (readTVar (users a)) return $ "Engine { " ++ "engineName = " ++ show (engineName a) ++ ", " ++ "version = " ++ show (version a) ++ ", " ++ "users = %" ++ show users ++ " }" When I run it however, I get this: *Main> te <- createEngine "Hello" 1 2 *Main> s <- tshow te <interactive>:1:5: Overlapping instances for TShow Engine arising from a use of `tshow' at <interactive>:1:5-12 Matching instances: instance (Show a) => TShow a -- Defined at fxmain.hs:(26,0)-(27,27) instance TShow Engine -- Defined at fxmain.hs:(51,0)-(58,41) In a stmt of a 'do' expression: s <- tshow te I'm not seeing how instance (Show a) => TShow a in the above error message is applicable here since Engine is not an instance of Show. Why is it complaining? Thanks, -John

3 2

Problem with System.Random.randoms
by S. Günther 09 Dec '08

09 Dec '08

Hi, I have a small problem with System.Random.randoms. I need a rather large number of random numbers but the following program consumes a huge amount of memory. I terminated it when it used up more than 2 Gb: module Main where import System.Random n :: Int n = maxBound main = do g <- getStdGen print $ length $ take n $ ((randoms g)::[Int]) On the other hand using take n $ [1..] it runs in constant space. Am I doing something wrong? Or should I just abandon randoms and use the more primitive functions in System.Random? Thanks in advance S. Günther

3 2

Deriving something else?
by John Ky 08 Dec '08

08 Dec '08

Hi, I've defined a class and some instances, which I'm hoping would help me "show" values of types that may include transactional elements. class TShow a where tshow :: a -> IO String instance Show (TVar a) where show = "%" instance (Show a) => TShow a where tshow a = return $ show a instance (Show a) => TShow (TVar a) where tshow ta = do a <- readTVar ta return $ show a Having created a new class is it possible to do some magic so that it can be put it into a deriving clause? data Type = Type { field1 :: Int , field2 :: Int } deriving Show data AnotherType = AnotherType { field3 :: Int , field4 :: TVar Type } deriving *TShow* Thanks -John

2 1

Animated line art
by Andrew Coppin 08 Dec '08

08 Dec '08

So, the muse has taken me. I'm going to attempt to produce some animated mathematical drawings involving lines and curves. Gtk2hs has a Cairo binding that should make rendering the stuff fairly straight-forward. So no problems there. Now, what I *could* do is write a new Haskell program for each drawing I want, hard-coding in the necessary imperative loops for updating coordinates, looping over multiple lines, etc. But that wasn't be very Haskell, would it? ;-) It seems that the "correct" course of action is to design a DSL for declaratively describing animated line art. Does anybody have ideas about what such a thing might look like?

8 12

Is unsafePerformIO safe here?
by John Ky 08 Dec '08

08 Dec '08

Hi, Is the following safe? moo :: TVar Int moo = unsafePerformIO $ newTVarIO 1 I'm interested in writing a stateful application, and I wanted to start with writing some IO functions that did stuff on some state and then test them over long periods of time in GHCi. I was worried I might be depending on some guarantees that aren't actually there, like moo being discarded and recreated inbetween invocations of different functions. Thanks, -John

4 5

Haskell haikus
by Gwern Branwen 08 Dec '08

08 Dec '08

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 Hi everyone. So today I finally got around to something long on my todo list - a compilation of all the Haskell haikus I've seen around! It is at http://haskell.org/haskellwiki/Haiku But I'm afraid I only have 5, and Google doesn't turn up any more. So: does anybody have a haiku I missed? Or even better, is anyone feeling poetically inspired tonight? :) - -- gwern -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) iEYEAREKAAYFAkk5ttoACgkQvpDo5Pfl1oJ+ygCfdSSBmbgyoOwkG53rKahF2Su1 84UAoIOrxGwe3u+WwnKxvKulq1AT4IJS =+2kH -----END PGP SIGNATURE-----

11 14

Could FDs help usurp an ATs syntactic restriction?
by Nicolas Frisby 08 Dec '08

08 Dec '08

>From the error below, I'm inferring that the RHS of the associated type definition can only contain type variables from the instance head, not the instance context. I didn't explicitly see this restriction when reading the GHC/Type_families entry. Could perhaps the "a b -> bn" functional dependency of the TypeEq class lift this restriction for bn? This isn't my ball park, but that idea has my hopes up :). <haskell> {-# LANGUAGE TypeFamilies #-} import TypeEq -- Attempting to encapsulate TypeEq behind an associated type. class EQ a b where type BN a b instance TypeEq a b bn => EQ a b where type BN a b = bn </haskell> results in an error <ghci> /tmp/Test.hs:9:16: Not in scope: type variable `bn' Failed, modules loaded: none. </ghci>

2 2

ANN: permutation-0.2
by Patrick Perry 08 Dec '08

08 Dec '08

Hi Everyone, I've uploaded a new version of the permutation library to hackage. Here is the package description: This library includes data types for storing permutations. It implements pure and impure types, the latter which can be modified in-place. The main utility of the library is converting between the linear representation of a permutation to a sequence of swaps. This allows, for instance, applying a permutation or its inverse to an array with O(1) memory use. . Much of the interface for the library is based on the permutation functions in the GNU Scientific Library (GSL). Here's the url: http://hackage.haskell.org/cgi-bin/hackage-scripts/package/permutation Patrick

1 0

[Announce] TTTAS (with excuse for second attempt to post)
by S. Doaitse Swierstra 08 Dec '08

08 Dec '08

We are pleased to announce the availability of the package "TTTAS", which contains the code associated with our paper at the coming TLDI workshop: \@inproceedings{ BSV09, author = {Arthur Baars and S. Doaitse Swierstra and Marcos Viera}, title = {Typed Transformations of Typed Abstract Syntax}, booktitle = {TLDI '09: fourth ACM SIGPLAN Workshop on Types in Language Design and Implementation}, year = {2009}, location = {Savannah, Georgia, USA}, publisher = {ACM}, address = {New York, NY, USA}, } For more information see: http://www.cs.uu.nl/wiki/bin/view/Center/TTTAS Arthur Baars Marcos Viera Doaitse Swierstra

1 0

Dr Dobbs: Time to get good at functional programming
by Tim Wawrzynczak 08 Dec '08

08 Dec '08

Amen to that. People who haven't really given a fair look at functional langauges (Haskell in particular) seem to have a very poor conception of them. Again, this seems to especially be a problem with Haskell (i.e., the whole "monads are hard" thing)... If this is where people are getting their information from, it would definitely behoove the Haskell community to update its image to further Haskell and FP in general :) On Sun, Dec 7, 2008 at 12:11 PM, Don Stewart <dons(a)galois.com> wrote: > jason.dusek: > > Too bad they didn't pimp Haskell as practical. > > It looked like an archaic view of Haskell based on reading wikipedia, > imo. Perhaps we should take charge of the wikipedia page, if it is that > influential. > _______________________________________________ > Haskell-Cafe mailing list > Haskell-Cafe(a)haskell.org > http://www.haskell.org/mailman/listinfo/haskell-cafe >

1 0

Data.Map.fromListWith
by Paul Johnson 07 Dec '08

07 Dec '08

I've just been looking at the Data.Map function "fromListWith". According to the docs, it has the type: * fromListWith* :: Ord <http://www.haskell.org/ghc/docs/latest/html/libraries/base/Data-Ord.html#t%…> k => (a -> a -> a) -> [(k, a)] -> Map <http://www.haskell.org/ghc/docs/latest/html/libraries/containers/Data-Map.h…> k a I'd have thought that a better type would be * fromListWith* :: Ord <http://www.haskell.org/ghc/docs/latest/html/libraries/base/Data-Ord.html#t%…> k => (a -> b -> b) -> [(k, a)] -> Map <http://www.haskell.org/ghc/docs/latest/html/libraries/containers/Data-Map.h…> k b This wouldn't break any existing code, but would allow things like "fromListWith (:)" to do the Right Thing. Would this be a sensible change (along with the other "with" functions in the module). Paul.

3 3