January 2011 - Haskell-Cafe

ANN: hmatrix-0.11
by Alberto Ruiz 26 Jan '11

26 Jan '11

Hi all, I am happy to announce hmatrix-0.11, a package for numerical computation based on LAPACK and GSL. Home page: http://code.haskell.org/hmatrix/ Hackage: http://hackage.haskell.org/package/hmatrix Recent changes: http://code.haskell.org/hmatrix/CHANGES By default this version uses Data.Vector.Storable from Roman Leshchinskiy’s vector package. This is useful for working with 1D arrays without explicit conversion (but note that Matrix t /= Vector (Vector t)). As a result, there is a different Show/Read instance for Vector, but no other changes should be observed using only hmatrix functions. (A possible issue is that empty vectors are typically not allowed in hmatrix, but can now be easily created using tools in the vector package.) I hope this is useful. In any case, this can be disabled with flag -f-vector and we can still have O(1) conversion functions. Thanks to everyone who contributed with patches, bug reports, suggestions, and installation problems. Special thanks go to Vivian McPhail for many excellent contributions, including single precision support, API reorganization, and many useful functions. Alberto PS: In relation with the license thread, I am open to change to BSD3 the source code not depending on GSL.

2 1

Are constructors strict?
by Daryoush Mehrtash 26 Jan '11

26 Jan '11

I am having hard time understanding the following paragraph in "Purely functional Lazy non-deterministic programing" paper http://www.cs.rutgers.edu/~ccshan/rational/lazy-nondet.pdf The problem with the naive monadic encoding of non-determinism is that the arguments to a constructor must be deterministic. If these arguments are themselves results of non-deterministic computations, these computations must be performed completely before we can apply the constructor to build a non-deterministic result. Why does the argument to constructors must be deterministic? WHy is it that thunks are not used in this case? Thanks -- Daryoush Weblog: http://onfp.blogspot.com/

4 7

Takusen and QuickCheck
by Dmitry Olshansky 26 Jan '11

26 Jan '11

Hello, I have a problem with using Takusen in my current project because it uses old QuickCheck 1.*. This leads to conflict with other packages which use QuickCheck 2.*. Could someone help me to decide this problem? Best wishes, Dmitry

2 1

Re: [Haskell-cafe] GPL License of H-Matrix and prelude numeric
by Vivian McPhail 25 Jan '11

25 Jan '11

Hi, > On Wed, 2011-01-26 at 08:18 +1000, Ivan Lazar Miljenovic wrote: > > Or else because the developers _wanted_ to license it under the GPL. > > Some people do, you know. > > Sure, but I agree it would be nice to know whether the authors chose the > GPL, or applied it because of linking with GPLed native libraries and > might be open to relicensing the native stuff. Seems like a good > question to me. > > -- > Chris Smith hmatrix contains a binding to (i) the GNU GSL library (for fast vector optimisations) and (ii) BLAS/LAPACK (for linear algebra). The GNU GSL library _requires_ a LGPL license. In the short term I do not think this can change. In the longer term, once dph and friends are working well and performance is optimised (dph can do fast linear algebra), the reliance on GSL and LAPACK might be removed. The hmatrix package can be configured to use Data.Vector.Storable of the vector package, and there is work being done on improving fast vector operations in that package. As far as I know, the vector package is _not_ intending to supply mathematical operations. Alberto? Cheers, Vivian

1 0

HMatrix Vector/Matrix interpolation ala Matlab interp/interp2 ??
by gutti 25 Jan '11

25 Jan '11

Dear Haskellers, I'm looking for Vector and especially Matric interpolation ala: z = interp2 (xMatrix, yMatrix, zMatrix, x, y) - x and y can be single values or also vectors or matrices - indeally with the options nearest, linear, quadratic, qubic.. Any hope that there is something similar especially using the HMatrix matrices (Matrix Double). - If I have to code it any suggestions ? Cheers Phil -- View this message in context: http://haskell.1045720.n5.nabble.com/HMatrix-Vector-Matrix-interpolation-al… Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.

6 15

Monomorphic let bindings and darcs
by Brandon Moore 25 Jan '11

25 Jan '11

The darcs 2.5 package uses the flat -fglasgow-exts, so it picks up MonoLocalBinds. This causes quite a few errors. The error messages are particularly bad in src/Darcs/Commands/Convert.lhs there is code of the form explicitlyPolyMorphicFun arg $- \local -> ... very long body ... let localdef = _ in ... more long body ... where failure to generalize the inner let is reported as insufficient polymorphism at the outer annotation. Glancing over the paper, it seems like the specific problem is just deferring equality constraints between ground types. Can't that be handled with a notion of constrait normalization?

2 1

Re: [Haskell-cafe] Proposal: Applicative => Monad: Call for consensus
by Conor McBride 25 Jan '11

25 Jan '11

Hi Tyson (So OT, I'm switching to cafe.) On 19 Jan 2011, at 18:24, Tyson Whitehead wrote: > On January 17, 2011 16:20:22 Conor McBride wrote: >> Ahem > > : ) > >>> The unfortunate pain you pay for this additional power is manually >>> having to >>> specify the application (<$> and <*>) and merging (join). If the >>> compiler >>> could figure this all out for you based on the underlying types, >>> wow! >> >> To achieve such a thing, one would need to ensure a slightly more >> deliberate separation of "value" and "computation" in the >> presentation >> of types. In Haskell, we use, e.g., [Int], for >> >> * pure computations of lists of integers >> * nondeterministic computations of integers [..] > I'm pretty sure I know what "pure computations of lists of integers" > is, but > I'm not so sure about "nondeterministic computations of integers". > > If it is not too much of an effort, could you clarify with a quick > example? Viewing [] as a monad, you can view [1,2] as a nondeterministic integer with possible values 1 and 2. Lifting operations to this monad gives you "all possible combinations" computation, so, as SHE would have it, (| [1,2] + [3,4] |) = pure (+) <*> [1,2] <*> [3,4] = [4,5,5,6] It's not hard to come up with examples where lifted and unlifted both make sense. With a bit of help from Twitter, we have (courtesy of Andy Gimblett) (["under","over"] ++ ["state","wear"]) = ["under","over","state","wear"] but (|["under","over"] ++ ["state","wear"]|) = ["understate","underwear","overstate","overwear"] and (courtesy of Dan Piponi) (["plan","tan"] ++ ["gent","king"]) = ["plan","tan","gent","king"] but (|["plan","tan"] ++ ["gent","king"]|) = ["plangent","planking","tangent","tanking"] In each case, the former has (++) acting on lists of strings as pure values, while the latter has (++) acting on strings as values given in []-computations. The type [String] determines a domain, it does not decompose uniquely to a notion of computation and a notion of value. We currently resolve this ambiguity by using one syntax for pure computations with [String] values and a different syntax for [] computations with String values. Just as we use newtypes to put a different spin on types which are denotationally the same, it might be worth considering a clearer (but renegotiable) separation of the computation and value aspects of types, in order to allow a syntax in which functions are typed as if they act on *values*, but lifted to whatever notion of computation is ambient. After types for representation, let us have types for explanation. All the best Conor

9 11

getDirectoryContents and sort order
by John Lask 25 Jan '11

25 Jan '11

I have noticed that on my windows box and ghc 6.12.3 I get the return list for System.Direcotry.getDirectoryContents in reverse sorted order. This is a change from previous observed behavior and I would consider it a bug. I would like to verify that it is not just me.

2 1

Data.Ranges show error
by Aaron Gray 25 Jan '11

25 Jan '11

I am getting the following error when trying to do a show on a Ranges object :- C:\Languages\Haskell>ghci rangeTest.hs GHCi, version 6.12.3: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. Loading package ffi-1.0 ... linking ... done. [1 of 1] Compiling Main ( rangeTest.hs, interpreted ) Ok, modules loaded: Main. *Main> main Loading package array-0.3.0.1 ... linking ... done. Loading package containers-0.3.0.0 ... linking ... done. Loading package ranges-0.2.3 ... linking ... done. Ranges [(37*** Exception: <stdout>: hPutChar: invalid argument (character is not in the code page) *Main> I have attached the test source also :- import Data.Ranges test = (ranges [ range 32 33, range 34 35, range 37 39 ]) main :: IO () main = do { putStrLn (show test) } I am probably doing something fundermentaly wrong. Many thanks in advance, Aaron

5 10

Tracing applied functions
by Aaron Gray 25 Jan '11

25 Jan '11

Is there anyway to get a list of applied functions in the running of a Haskell program ? Many thanks in advance, Aaron

3 5