- Haskell-Cafe - Haskell.org

A question about mfix
by Wei Hu 30 Jul '08

30 Jul '08

What's wrong about giving mfix the following general definition? > mfix :: (a -> m a) -> m a > mfix f = (mfix f) >>= f I know it diverges if (>>=) is strict on the first argument. My question is, is this definition correct for all lazy monads? The documentation (http://haskell.org/ghc/docs/latest/html/libraries/base/ Control-Monad-Fix.html#v%3Amfix) says "mfix f executes the action f only once, with the eventual output fed back as the input.". So my definition looks like a valid one, doesn't it? I haven't fully wrapped my head around this monadic fixed-point thing yet. So, if you can give an example showing how my definition differs from a standard monad, that'll be great.

4 5

Fw: patch applied (ghc): Remove the OpenGL family of libraries fromextralibs
by Claus Reinke 29 Jul '08

29 Jul '08

FYI: Haskell's OpenGL binding has just been dropped from GHC's extralibs, which means that it will no longer be kept in sync with GHC development, at least not by GHC HQ. GHC HQ has its hands full and -generally speaking - extralibs are to be replaced by H(L)P, the Haskell (Library) Platform: http://haskell.org/haskellwiki/Haskell_Platform so this part is understandable. But OpenGL has been dropped before H(L)P is ready to take over from extralibs, and at the recent GHC Irc meeting on the topic http://hackage.haskell.org/trac/ghc/attachment/wiki/IRC_Meetings/ghc-2008-0… the mood seemed to favour not including OpenGL&co in the initial versions of H(L)P. Sven put a lot of good work into these libraries, but he is often not around for a long time, and it would be a shame if these gems were orphaned and went out of sync in the meantime. While I haven't used OpenGL in a while, I've always hoped to come back to that, not to mention Sven's spatial audio binding additions. And if I'm not mistaken, there are other community members who are using these libs right now for research, possibly even prototyping in connection with a startup? So, if one of you wanted to step forward and offer to keep these Haskell bindings for OpenGL&co maintained, perhaps steward them into the H(L)P, now would probably be a good time. Just thought I'd forward this, for those not following cvs-ghc, Claus ----- Original Message ----- From: "Ian Lynagh" <igloo(a)earth.li> To: <cvs-ghc(a)haskell.org> Sent: Friday, July 25, 2008 12:11 AM Subject: patch applied (ghc): Remove the OpenGL family of libraries fromextralibs > Thu Jul 24 03:27:36 PDT 2008 Ian Lynagh <igloo(a)earth.li> > * Remove the OpenGL family of libraries from extralibs > > M ./libraries/Makefile -4 > M ./libraries/extra-packages -4 > > View patch online: > http://darcs.haskell.org/ghc/_darcs/patches/20080724102736-3fd76-f5601cb7a2… > > _______________________________________________ > Cvs-ghc mailing list > Cvs-ghc(a)haskell.org > http://www.haskell.org/mailman/listinfo/cvs-ghc

11 21

Problem compiling a CGI script that needs to write to file during its execution
by Jefferson Heard 29 Jul '08

29 Jul '08

Please ignore the obvious security holes, as this is not a script meant for public consumption, but some internal testing and prototyping. I would like to write the result of my computation out to a file inside of cgiMain, but the type of the monad inside cgiMain is this odd CGIT IO CGIResult. I tried using liftM on writeFile, but it then complained that "newanns" was a string instead of a list of strings, which I don't understand at all. Here's the code: DeleteAnnotation.hs: -------------------------------------------------------------------------------- import Network.CGI import Annotations import Graphics.Rendering.OpenGL.GL (GLfloat) import Control.Monad (liftM) import Data.List (filter) getInput' v = do x <- getInput v case x of Nothing -> fail "essential variable not found" Just y -> return y cgiMain :: String -> CGI CGIResult cgiMain anns_dot_txt = do ordnl <- (liftM read) $ getInput' "ordinal" let anns = (filter (notequal ordnl) . read $ anns_dot_txt) :: [Annotation] newanns = show anns output $ newanns writeFile "Annotations.txt" $ newanns notequal :: String -> Annotation -> Bool notequal ordnl ann = ordnl == ordinal ann main :: IO () main = do f <- readFile "Annotations.txt" runCGI (handleErrors (cgiMain f)) -------------------------------------------------------------------------------- $ ghc --make DeleteAnnotation DeleteAnnotation.hs:19:2: Couldn't match expected type `CGIT IO CGIResult' against inferred type `IO ()' In the expression: writeFile "Annotations.txt" $ newanns In the expression: do ordnl <- (liftM read) $ getInput' "ordinal" let anns = ... newanns = show anns output $ newanns writeFile "Annotations.txt" $ newanns In the definition of `cgiMain': cgiMain anns_dot_txt = do ordnl <- (liftM read) $ getInput' "ordinal" let anns = ... .... output $ newanns .... If I change writeFile "Annotations.txt" to (liftM (writeFile "Annotations.txt")): $ ghc --make DeleteAnnotation DeleteAnnotation.hs:19:42: Couldn't match expected type `String' against inferred type `Char' Expected type: [String] Inferred type: String In the second argument of `($)', namely `newanns' In the expression: (liftM (writeFile "Annotations.txt")) $ newanns

4 4

#haskell irc channel reaches 500 users
by Don Stewart 29 Jul '08

29 Jul '08

A small announcement :) 6 1/2 years after its inception, under the guiding hand of Shae Erisson (aka shapr), the #haskell IRC channel[1] on freenode has finally reached 500 users! To chart the growth, we can note that the channel was founded in late 2001, and had slow growth till 2006, reaching 200 users in January of that year. Since then growth in the user base has been far more rapid, reaching 300 users in Dec 2006, 400 users in August 2007, and 500 users by July 2008. This puts the channel at around the 12th largest community of the 7000 freenode channels. For comparision, a sample of the state of the other language communities, with comments comapred to their status a year ago: #php 720 #python 640 -- up from 5th place #perl 620 ##c++ 585 ##c 530 > #haskell 502 #rubyonrails 428 #ruby-lang 350 -- down from 420 max. ##javascript 308 #lisp 295 #erlang 178 #perl6 129 -- unchanged #scheme 148 -- unchanged #lua 120 #ocaml 70 -- unchanged You can see the growth of the channel over here: http://www.cse.unsw.edu.au/~dons/irc If you've not dropped by the channel yet, feel free to come and chat, and toss around some lambdas! :) Cheers, Don

1 0

ANN: Hayoo! beta 0.2
by Timo B. Hübel 29 Jul '08

29 Jul '08

Hello, we are pleased to announce the second beta release of Hayoo!, a Haskell API search engine providing advanced features like suggestions, find-as-you-type, fuzzy queries and much more. Visit Hayoo! here: http://holumbus.fh-wedel.de/hayoo The major change is the inclusion of all packages available on Hackage, i.e. the documentation of the latest versions of all packages is included in the index. Unfortunately we had to drop the direct links to the source code, as the documentation on Hackage is currently generated without source code. But as soon as this changes, we will include these links again. Additionally, we added some tweaks to the interface which make the browser history/the back button work (at least in Firefox). Please bear in mind that this is still a beta release and we are continuously working on further improvements. Any suggestions and feedback is highly welcomed. Cheers, Timo & Sebastian

4 4

Cabal and Strings and CPP
by Philip Weaver 29 Jul '08

29 Jul '08

Hello all, I'm trying to use CPP-defined strings in a Haskell module, like this: main :: IO () main = putStrLn FOO This of course will not work: ghc -DFOO="hello world" --make Main.hs -o test You'll get this error message: ./Main.hs:6:16: Not in scope: `hello' ./Main.hs:6:22: Not in scope: `world' Either of these will do what I want: ghc -DFOO="\"hello world\"" --make Main.hs -o test ghc -DFOO='"hello world"' --make Main.hs -o test # (that's double quotes inside single quotes) However, passing the same CPP definition via cabal does not work. runhaskell Setup.hs build --ghc-options=-DFOO="\"hello world\"" runhaskell Setup.hs build --ghc-options=-DFOO='"hello world"' With either of these commands, I get the same error message as above. This is understandable, since cabal has to evaluate the string before sending it to GHC, so I lose my escaped quotes. Any idea how I could change the Haskell module or the command line argument so that I get what I want? I've tried many combinations of quotes and escaped quotes with no luck. Thanks, - Phil

3 3

carry "state" around ....
by Galchin, Vasili 29 Jul '08

29 Jul '08

hello, Following is more of a criticism of Linux implementation of the Posix real-time extension of asynchronous I/O.... if not interesting please skip. The central data structure for Posix AIO is an aiocb. In any case, the Linux implementors added to the aiocb: /* Asynchronous I/O control block. */ struct aiocb { int aio_fildes; /* File desriptor. */ int aio_lio_opcode; /* Operation to be performed. */ int aio_reqprio; /* Request priority offset. */ volatile void *aio_buf; /* Location of buffer. */ size_t aio_nbytes; /* Length of transfer. */ struct sigevent aio_sigevent; /* Signal number and value. */ /* Internal members. */ <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< struct aiocb *__next_prio; int __abs_prio; int __policy; int __error_code; __ssize_t __return_value; #ifndef __USE_FILE_OFFSET64 __off_t aio_offset; /* File offset. */ char __pad[sizeof (__off64_t) - sizeof (__off_t)]; #else __off64_t aio_offset; /* File offset. */ #endif char __unused[32]; }; My viewpoint is that the above "Internal members" must be "carried" around in a Haskell program. Am I correct?? If I am correct, then the Linux implementation of Posix AIO is not portable to say Solaris? In hindsight, if I am correct, it seems that the Linux implementation of AIO should use the ordered pair (pid, fd) to reference the "internal" members and leave the "aiocb" "clean"? Very kind regards, Vasili

9 27

Data Types a la Carte - automatic injections (help!)
by Kenn Knowles 29 Jul '08

29 Jul '08

I have a question about Data Types a la Carte (http://www.cs.nott.ac.uk/~wss/Publications/DataTypesALaCarte.pdf) and more generally hacking smart coproduct injections into Haskell (all extensions are fair game). > {-# LANGUAGE MultiParamTypeClasses,TypeOperators,UndecidableInstances,IncoherentInstances,FlexibleInstances,FlexibleContexts #-} Here are the definitions from Wouter's paper, which cleverly manage the instances of (:<:) > data (f :+: g) a = Inl (f a) | Inr (g a) > infixr 6 :+: > instance (Functor f, Functor g) => Functor (f :+: g) where > fmap f (Inl x) = Inl (fmap f x) > fmap f (Inr x) = Inr (fmap f x) > class (Functor sub, Functor sup) => sub :<: sup where > inj :: sub a -> sup a > instance Functor f => (:<:) f f where > inj = id > instance (Functor f, Functor g) => (:<:) f (f :+: g) where > inj = Inl > instance (Functor f, Functor g, Functor h, (f :<: g)) => (:<:) f (h :+: g) where > inj = Inr . inj Now in my case, I want to be able to inject an arbitrary coproduct into a larger one, assuming the components are in the same order, so it is a subsequence of the larger sum, for example: > coprodInject1 :: (Functor f, Functor g, Functor h, Functor i) => (f :+: g :+: h) a -> (f :+: g :+: h :+: i) a > coprodInject1 = inj > coprodInject2 :: (Functor f, Functor g, Functor h, Functor i) => (f :+: g :+: i) a -> (f :+: g :+: h :+: i) a > coprodInject2 = inj > instance (Functor f, Functor g, Functor h, (g :<: h)) => (:<:) (f :+: g) (f :+: h) where > inj (Inl x) = Inl x > inj (Inr x) = Inr (inj x) The above works fine, and maybe I could manipulate my code to only have injections of those forms, but I want this next one to work too: > coprodInject3 :: (Functor f, Functor g, Functor h, Functor i) => (f :+: g :+: i :+: j) a -> (f :+: g :+: h :+: i :+: j) a > coprodInject3 = inj It fails because after "passing up" the h in the search, it can either use the reflexivity rule or decompose the sum on (i :+: j). The full error is: Overlapping instances for (i :+: j) :<: (i :+: j) arising from a use of `inj' ... Matching instances: instance [incoherent] (Functor f) => f :<: f -- Defined at ... instance [incoherent] (Functor f, Functor g, Functor h, g :<: h) => (f :+: g) :<: (f :+: h) -- Defined at ... In the expression: inj In the definition of `coprodInject3': coprodInject3 = inj What confuses me is that IncoherentInstances is on, but it is still rejected by GHC 6.8.3 seemingly for being incoherent. I haven't tried it with any other version. Am I missing something? Any suggestions or pointers?

3 2

ANN: RandomDotOrg-0.1
by Austin Seipp 29 Jul '08

29 Jul '08

Hi, I've just uploaded a package to hackage which is an interface to the random.org random number generator. For those who don't know, random.org provides random data through the use of atmospheric noise rather than a PRNG that would typically be invoked if you were to use the System.Random module or somesuch. As of current, it can only generate a list of random numbers. However, random.org also provides an interface to generate random strings, randomize an input list and a random sequence generator (the difference being that the sequence generator will not repeat itself, while the integer one may.) Would anybody find these interfaces useful? I think the random number generator alone is pretty nice, but I'm thinking of expanding it in the future with these other things since they're nice to have if you want randomness anyway. You can get it from [1] or just: $ cabal install RandomDotOrg The development repository is tracked using git; you can get yourself a clone by doing $ git clone git://github.com/thoughtpolice/randomdotorg.git Austin [1] http://hackage.haskell.org/cgi-bin/hackage-scripts/package/RandomDotOrg-0.1

2 1

ANN: Mueval 0.3.1, 0.4, 0.4.5, 0.4.6, 0.5
by Gwern Branwen 28 Jul '08

28 Jul '08

Hi everyone. So there've been a number of releases of Mueval lately. I figured I'd tell you about them. WHERE As always, you can 'cabal install mueval', or get it from Hackage <http://hackage.haskell.org/cgi-bin/hackage-scripts/package/mueval>, or get it from the new Darcs repository <http://code.haskell.org/mubot/>. Usage remains as before, but see below for numerous examples, or the file 'tests.sh'. WHAT'S INTERESTING Mueval is now even more capable. I've fixed the big security hole Spencer identified, so right now I know of no vulnerabilities in Mueval (or even potentially dangerous aspects). More interestingly is how close Mueval is now to Lambdabot's capabilities: we can evaluate all the same expressions, and Mueval now even can do QuickCheck tests. The only missing pieces is that currently there's no interface to Smallcheck yet (but hardly anyone in #haskell seems to use that, so I'm considering not even bothering), and currently there's no way to do @let or qualified imports. I hope to add these two missing pieces for the next release - and then I can finally remove hs-plugins from Lambdabot. CHANGES 0.5 Adds support for Unicode in expressions! bash$ mueval -e 'let (ñ) = (+) in ñ 5 5' 10" Adds support for QuickCheck tests! bash-3.2$ mueval -E -e 'myquickcheck (\x -> x == x)' "\"OK, passed 500 tests.\\n\"" bash-3.2$ mueval -E -e 'myquickcheck (\x -> x+1 == x+2)' "\"Falsifiable, after 0 tests:\\n0\\n\"" --inferred-type prints out even more information! bash-3.2$ mueval --inferred-type --expression 'foldr (\x y -> concat ["(f ",x," ",y,")"]) "z" (map show [1..5])' foldr (\x y -> concat ["(f ",x," ",y,")"]) "z" (map show [1..5]) [Char] "\"(f 1 (f 2 (f 3 (f 4 (f 5 z)))))\"" 0.4.6 A README! --print-type was renamed to --inferred-type so as to avoid short-option ambiguity! Refactoring! Haddocks look somewhat nicer! 0.4.5 Even more refactoring and modularization! Performance improvements! (ie. 'call error sooner so mueval doesn't hang around waiting for the timeout to expire') 0.4 Close major security hole! No more: mueval --expression "Foreign.unsafePerformIO $ readFile \"foo\"" (Since now one must declare module imports, and cannot circumvent them due to a certain bad behaviour of the GHC API.) 0.3.1 Blacklist unsafe keywords! mueval --expression "Foreign.unsafePerformIO $ readFile \"/etc/passwd\"" mueval: Unsafe functions to use mentioned. Looser resource limits (so hopefully more people can run it)! A Darcs repository! darcs get http://code.haskell.org/mubot/ (And of course mixed in throughout these changes are various test additions; it's a respectable little suite now.) -- gwern Propaganda Security rip CIO VIP Pershing O/S MPRI JASON AMEMB

1 0