- Haskell-Cafe - Haskell.org

Haskell and arrays
by Morten Olsen lysgaard 05 May '12

05 May '12

I'm a bit confused on what the standard is when it comes to arrays and performance in Haskell. Say I have a function that takes an n dimensional array. The array only contains some primitive type. The function mutates the array in an deterministic but chaotic pattern. The new value of an element might depend on other values in the array, and the order of update might depend on what values are found. How would I implement this in an efficient fashion in Haskell. Is the array package the way to go? Would vector with some custom indexing scheme work well? The array package provides n dimensional indexing, vector doesn't. Does vector perform better than array? Is it realistic to approach the performance of a naive c implementation? What steps would need to be taken in order to achieve that? I know there are many question here, but I don't expect perfect answers to them. I just need to bootstrap my map of numerics in Haskell. PS: An example of the above function could be Gaussian elimination of a matrix. -- Morten Olsen Lysgaard NTNU-IT - Orakeltjenesten NTNU - Physics and Mathematics

3 2

haskellonline.org, a web frontend for ghc
by Gintautas Miliauskas 05 May '12

05 May '12

Hello, check out http://haskellonline.org, an online Haskell typechecker. It is essentially a thin web wrapper over ghc, but it takes some friction out of learning Haskell, if only because no local setup is required. It is a weekend's work, so do not be too harsh, but feedback is welcome! I have also written up an accompanying blog post [1]. [1] http://blog.miliauskas.lt/2012/05/online-haskell-typechecker.html Best regards, -- Gintautas Miliauskas

2 1

Generalizing (++) for monoids instead of using (<>)
by aditya bhargava 04 May '12

04 May '12

After asking this question: http://stackoverflow.com/questions/9963050/standard-way-of-joining-two-data… I found out that the new infix operator for `mappend` is (<>). I'm wondering why ghc 7.4 didn't generalize (++) to work on monoids instead. To me, (++) is much more clear. (<>) means "not equal to" for me. Can anyone shed light on this decision? Adit -- adit.io

9 9

ANN: meta-par, meta-par-accelerate and friends -- heterogeneous parallel programming
by Ryan Newton 04 May '12

04 May '12

Hi all, I'm happy to announce a release of meta-par, a parallel programming library that is a successor to monad-par. - http://hackage.haskell.org/package/abstract-par - http://hackage.haskell.org/package/monad-par - http://hackage.haskell.org/package/meta-par - http://hackage.haskell.org/package/abstract-par-accelerate - http://hackage.haskell.org/package/meta-par-accelerate The distributed programming components aren't released yet, but this blog posts describes how to use the above packages to do hybrid CPU/GPU programming: http://parfunk.blogspot.com/2012/05/how-to-write-hybrid-cpugpu-programs.html Cheers, -Ryan

1 0

conduit: Finalize field in PipeM
by Paolo Capriotti 04 May '12

04 May '12

Hi, I'm trying to write a function to convert a conduit to a Pipe (from pipes-core), and I'm having trouble understanding why the `Finalize` field in a `PipeM` constructor returns `r`. This makes it impossible to create it from the corresponding `M` constructor in pipes-core, since `M` includes an exception handler which is not guaranteed to run fully, hence may not provide a return value. `HaveOutput` presents a similar problem. >From a cursory look at conduit's code, it doesn't look like the return value of `Finalize` is ever used, and it seems that the conduits that actually manage to define it either have `()` as return type or just throw an exception. The definition of `lift` for the `MonadTrans` instance duplicates the base monad action there, which doesn't look quite right to me (isn't that supposed to contain a _cleanup_ action?). Shouldn't the constructor be changed to something like `PipeM (m (Pipe i o m r)) (Finalize m ())` or am I completely off base here? Thanks. BR, Paolo

2 2

Fwd: Problem with forall type in type declaration
by Chris Smith 04 May '12

04 May '12

Oops, forgot to reply-all again... ---------- Forwarded message ---------- From: Chris Smith <cdsmith(a)gmail.com> Date: Fri, May 4, 2012 at 8:46 AM Subject: Re: [Haskell-cafe] Problem with forall type in type declaration To: Magicloud Magiclouds <magicloud.magiclouds(a)gmail.com> On Fri, May 4, 2012 at 2:34 AM, Magicloud Magiclouds <magicloud.magiclouds(a)gmail.com> wrote: > Sorry, it was just a persudo code. This might be more clear: > > run :: (Monad m) => m IO a -> IO a Unfortunately, that's not more clear. For the constraint (Monad m) to hold, m must have the kind (* -> *), so then (m IO a) is meaningless. I assume you meant one of the following: run :: MonadTrans m => m IO a -> IO a or run :: MonadIO m => m a -> IO a (Note that MonadIO is the class from the mtl package; there is no space there). Can you clarify which was meant? Or perhaps you meant something else entirely? -- Chris Smith

1 0

Annoyed at System.Random
by Thomas DuBuisson 04 May '12

04 May '12

Ryan, I've grown annoyed at System.Random enough (specifically, StdGen). How much, if any, pushback would there be if I put together a FFI binding to a C AES-CTR based RNG. There are many advantages: 0) The API wouldn't have to change (though some parts should, and some change is already planned) 1) It can be fast (maybe not MT fast, but really quite good) 2) It's secure and the splitting properties can be related to cryptographic problems. 3) It could use Intel NI instructions when available. 4) It's NIST standardized, so there exist known answer tests. 5) We could expose a method to fill an arbitrary buffer :: Storable a => Ptr a -> Int -> IO () 6) The rest of the community doesn't have to do any of the work. I'd be tempted to pull in the 'entropy' package for seeding, but will make that a separate proposal. Cheers, Thomas

5 15

Indent aware parser
by Lyndon Maydwell 04 May '12

04 May '12

Hi all. What's the best indentation-aware parser at the moment? I see three when I look in cabal: lyndon@pugno:~ » cabal list indent * IndentParser Synopsis: Combinators for parsing indentation based syntatic structures Default available version: 0.2.1 Installed versions: [ Not installed ] Homepage: http://www.cse.iitk.ac.in/~ppk License: GPL * indentparser Synopsis: A parser for indentation based structures Default available version: 0.1 Installed versions: [ Not installed ] Homepage: http://www.cse.iitk.ac.in/users/ppk/code/HASKELL/indentparser License: PublicDomain * indents Synopsis: indentation sensitive parser-combinators for parsec Default available version: 0.3.2 Installed versions: [ Not installed ] Homepage: http://patch-tag.com/r/salazar/indents License: BSD3

2 2

unable to run ghci (7.4.1)?
by Baojun Wang 03 May '12

03 May '12

Dear list, I build haskell-platform 2011.04 by gentoo ~amd64 haskell overlay. But I can't run ghci even I rebuilt ghc or run haskell-updater, error message: ghci GHCi, version 7.4.1: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... ghc: /usr/lib64/ghc-7.4.1/base-4.5.0.0/HSbase-4.5.0.0.o: unknown symbol `stat' ghc: unable to load package `base' Do you have some ideas how to avoid this error? Many thanks. Best Regards Baojun

3 5

ANN: containers-0.5.0.0
by Johan Tibell 03 May '12

03 May '12

Hi all, I'm proud to announce a new major release of the containers package : http://hackage.haskell.org/package/containers-0.5.0.0 There has been a whopping 160 commits since the last release. The major improvements are: * a clearer distinction between value-lazy and value-strict containers, * performance improvements across the board, * a big internal clean-up, and * new functions for e.g. merging, updating, and searching containers. The first item is perhaps the most visible. While the old Data.Map and Data.IntMap modules will continue to exist for the foreseeable future, we've abandoned the practice of having the strict and lazy versions of each function distinguished by an apostrophe. The distinction is instead made at the module level, by introducing four new modules: * Data.Map.Strict * Data.Map.Lazy * Data.IntMap.Strict * Data.IntMap.Lazy This split has three benefits: * It makes the choice between value-strict and value-lazy containers more declarative; you pick once at import time, instead of having to remember to use the strict or lazy versions of a function every time you modify the container. * It alleviates a common source of performance issues, by forcing the user to think about the strictness properties upfront. For example, using insertWith instead of insertWith' is a common source of containers-related performance bugs. * There are fewer functions per module, making it easier to get an overview of each module. Note that the types used in the strict and lazy APIs are the same, so you can still use the same container in a "mixed" manner, if needed. Contributors to this release: Milan Straka Johan Tibell Joachim Breitner Edward Z. Yang Herbert Valerio Riedel Matt West Max Bolingbroke Cheers, The containers maintainers

1 0