- Haskell-Cafe - Haskell.org

local type denotation
by Serge D. Mechveliani 14 Nov '12

14 Nov '12

Please, how to correctly set an explicit type for a local value in the body of a polymorphic function? Example (tested under ghc-7.6.1): data D a = D1 a | D2 a (a -> a) f :: Eq a => D a -> a f (D1 x) = x f (D2 x g) = let -- y :: Eq a => a y = g x in if x == y then x else g y main = putStr $ shows (f (D2 (1 :: Int) succ)) "\n" This is compiled by ghc --make Main Now I need, for a certain reason, to explicitly set the type for y in `let', with the meaning: "this very `a' which is in the signature for f" (and I think that this type Haskell assignes to y in "y = g x"). I need to declare this type in a separate line: y :: <what ever it is>. Both `y :: a' and `y :: Eq a => a' are not compiled. Please, copy the answer to mechvel(a)botik.ru Thanks, ------ Sergei

5 4

mtl-2.1 severly broken, cabal needs blacklisting
by Andreas Abel 14 Nov '12

14 Nov '12

After 2 days of shrinking 251 modules of source code to a few lines I realized that modify in MonadState causes <<loop>> in mtl-2.1. http://hackage.haskell.org/packages/archive/mtl/2.1/doc/html/src/Control-Mo… The bug has been fixed, apparently seven month ago. https://github.com/ekmett/mtl/pull/1 However, the "malicious" mtl-2.1 still lingers on: it is available from hackage and installed in many systems. This calls for a means of blacklisting broken or malicious packages. cabal update should also pull a blacklist of packages that will never be selected by cabal install (except maybe by explicit user safety overriding). I think such a mechanism is not only necessary for security purposes, but also to safe the valuable resources of our community. Cheers, Andreas -- Andreas Abel <>< Du bist der geliebte Mensch. Theoretical Computer Science, University of Munich Oettingenstr. 67, D-80538 Munich, GERMANY andreas.abel(a)ifi.lmu.de http://www2.tcs.ifi.lmu.de/~abel/

6 9

Does anyone know where George Pollard is?
by Myles C. Maxfield 14 Nov '12

14 Nov '12

Hello, I sent a message to George Pollard (porges(a)porg.es) about his 'idna' package [1] a couple days ago, but he hasn't responded. I'd like to depend on his package for something that I'm working on, but his package fails to build (The email I sent him includes a patch that should fix up the build problems). The package hasn't been updated for 3 years and he hasn't listed a source control repository. Does anyone know where he is? If not, is there an accepted practice to resolve this situation? Should I upload my own 'idna2' package? Thanks, Myles [1] http://hackage.haskell.org/package/idna

3 3

Re: [Haskell-cafe] Strange behavior with listArray
by oleg＠okmij.org 14 Nov '12

14 Nov '12

Alex Stangl wrote: > To make this concrete, here is the real solve function, which computes > a border array (Knuth-Morris-Pratt failure function) for a specified > string, before the broken memoization modification is made: > solve :: String -> String > solve w = let h = length w - 1 > wa = listArray (0, h) w > pI = 0 : solveR (tail w) 0 > solveR :: String -> Int -> [Int] > solveR [] _ = [] > solveR cl@(c:cs) k = if k > 0 && wa!k /= c > then solveR cl (pI!!(k-1)) > else let k' = if wa!k == c > then k + 1 > else k > in k' : solveR cs k' > in (intercalate " " . map show) pI > > Here solveR corresponds to f in the toy program and pI is the list > I would like to memoize since references to earlier elements could > get expensive for high values of k. Ok, let's apply a few program transformations. First we notice that the list pI must have the same length as the string w. Since we have already converted the string w to an array, wa, we could index into that array. We obtain the following version. > solve1 :: String -> String > solve1 w = (intercalate " " . map show) pI > where > h = length w - 1 > wa = listArray (0, h) w > pI = 0 : solveR 1 0 > solveR :: Int -> Int -> [Int] > solveR i k | i > h = [] > solveR i k = > let c = wa!i in > if k > 0 && wa!k /= c > then solveR i (pI!!(k-1)) > else let k' = if wa!k == c > then k + 1 > else k > in k' : solveR (i+1) k' > > t1s1 = solve1 "the rain in spain" > t1s2 = solve1 "aaaaaaaaaaaa" > t1s3 = solve1 "abbaabba" We don't need to invent an index: it is already in the problem. The unit tests confirm the semantics is preserved. The _general_ next step is to use the pair of indices (i,k) as the key to the two dimensional memo table. Luckily, our case is much less general. We do have a very nice dynamic programming problem. The key is the observation k' : solveR (i+1) k' After a new element, k', is produced, it is used as an argument to the solveR to produce the next element. This leads to a significant simplification: > solve2 :: String -> Array Int Int > solve2 w = pI > where > h = length w - 1 > wa = listArray (0, h) w > pI = listArray (0,h) $ 0 : [ solveR i (pI!(i-1)) | i <- [1..] ] > solveR :: Int -> Int -> Int > solveR i k = > let c = wa!i in > if k > 0 && wa!k /= c > then solveR i (pI!(k-1)) > else let k' = if wa!k == c > then k + 1 > else k > in k' > > t2s1 = solve2 "the rain in spain" > t2s2 = solve2 "aaaaaaaaaaaa" > t2s3 = solve2 "abbaabba" The unit tests pass.

2 1

Re: [Haskell-cafe] Strange behavior with listArray
by oleg＠okmij.org 14 Nov '12

14 Nov '12

Alex Stangl posed a problem of trying to efficiently memoize a function without causing divergence: > solve = let a :: Array Int Int > a = listArray (0, 3) (0 : f 0) > f k = if k > 0 > then f (a!0) > else 0 : f 1 > in (intercalate " " . map show . elems) a Daniel Fischer explained the reason for divergence: > The problem, Alex, is that > > f k = if k > 0 > then f (a!0) > else 0 : f 1 > > is strict, it needs to know the value of (a!0) to decide which branch to take. > But the construction of the array a needs to know how long the list (0 : f 0) > is (well, if it's at least four elements long) before it can return the array. > So there the cat eats its own tail, f needs to know (a part of) a before it > can proceed, but a needs to know more of f to return than it does. But the problem can be fixed: after all, f k is a list of integers. A list is an indexable collection. Let us introduce the explicit index: > import Data.Array((!),Array,elems,listArray) > import Data.List(intercalate) > > solve = (intercalate " " . map show . elems) a > where > a :: Array Int Int > a = listArray (0, 3) (0 : [f 0 i | i <- [0..]]) > > f 0 0 = 0 > f 0 i = f 1 (i-1) > f k i = f (a!0) i This converges. Here is a bit related problem: http://okmij.org/ftp/Haskell/AlgorithmsH.html#controlled-memoization

2 1

Quickcheck
by graham＠fatlazycat.com 13 Nov '12

13 Nov '12

Hi, Trying to find some good docs on QuickCheck, if anyone has one ? Been scanning what I can find, but a question. What would be the best way to generate two different/distinct integers ? Use arbitrary ( if so do you have an example ) or a conditional on the property. Though I read the later can reduce passing test count. Thanks Graham

5 6

hackageDB haddock errors
by timothyhobbs＠seznam.cz 13 Nov '12

13 Nov '12

Hello, There are several packages without haddock documentation on hackage. The one that bugs me right now, is the latest version of gtk2hs: http://hackage. haskell.org/package/gtk I've been told that this may be due to hackage updating docs with an infrequent cron job, so that new packages don't get their docs for a long time. Whatever is causing it, the fact that hayoo http://holumbus.fh-wedel. de/hayoo/hayoo.html always uses the most frequent docs, means that I cannot currently search for anything gtk2hs related :( Can someone please look into prodding this cron job so it doesn't take days to run, or look into resolving any errors that might be causing this unfortunate behaviour? Thank you for your time, and happy hacking, Timothy

3 2

Wrong install-name path for shared/dynamic C-library on OS X
by Christiaan Baaij 13 Nov '12

13 Nov '12

Dear List, The context of my problem is the following: - ghc-HEAD uses shared-library _only_ on OS X - A cabalized package that builds a shared/dynamic C-library (GLFW-b) And my problem is: A dynamically linked Haskell library gets the wrong install-name for a C-library: otool -L ~/.cabal/lib/GLFW-b-0.1.0.5/ghc-7.7.20121111/libHSGLFW-b-0.1.0.5-ghc7.7.20121111.dylib: /Users/baaijcpr/.cabal/lib/GLFW-b-0.1.0.5/ghc-7.7.20121111/libHSGLFW-b-0.1.0.5-ghc7.7.20121111.dylib: .... build/libglfw.dylib (compatibility version 1.0.0, current version 1.0.0) .... All the other paths reported by 'otool -L' are absolute, while the C-library (libglfw.dylib) that was part of the Cabalized package is relative. The problem presents itself when I now want to use the GLFW-b package: Loading package GLFW-b-0.1.0.5 ... can't load .so/.DLL for: /Users/baaijcpr/.cabal/lib/GLFW-b-0.1.0.5/ghc-7.7.20121111/libHSGLFW-b-0.1.0.5-ghc7.7.20121111.dylib (dlopen(/Users/baaijcpr/.cabal/lib/GLFW-b-0.1.0.5/ghc-7.7.20121111/libHSGLFW-b-0.1.0.5-ghc7.7.20121111.dylib, 9): Library not loaded: build/libglfw.dylib Referenced from: /Users/baaijcpr/.cabal/lib/GLFW-b-0.1.0.5/ghc-7.7.20121111/libHSGLFW-b-0.1.0.5-ghc7.7.20121111.dylib Reason: image not found) I can update the install-name to the correct path using: install_name_tool -change build/libglfw.dylib @loader_path/libglfw.dylib libHSGLFW-b-0.1.0.5-ghc7.7.20121111.dylib So my question are: - Where does the original install_name path (build/libglfw.dyli) come from? - What should be specified in the .cabal and/or Setup.hs file so that the haskell library is installed using the _correct_ relative path? Cheers, Christiaan

1 0

JavaScript (SpiderMonkey, V8, etc) embedded in GHC?
by Bob Hutchison 13 Nov '12

13 Nov '12

Hi, I've looked around with no success… this surprises me actually. Has anyone embedded SpiderMonkey, V8, or any other relatively decent JavaScript interpreters in GHC (using the FFI)? I did find http://justinethier.github.com/husk-scheme/ which is a scheme R5RS implementation (I could make this work). There's also some work done embedding Lua. I also found a number of packages that compile javascript to Haskell, or the other way around, but I don't need that kind of thing. All I really need is to allow users to write some JavaScript that accepts a single JSON 'file/string' from my Haskell program and produces another JSON 'file/string' that my Haskell program will accept. Thanks, Bob

4 7

Re: [Haskell-cafe] curl package broken in Windows
by Kevin Cantu 13 Nov '12

13 Nov '12

Heh, I was trying to build this yesterday and it was making by eyes bleed. With the curl package on Hackage, I merely need an SSL enabled version of libcurl, and every Linux distro I've tried this on has several variations of such a package. (You have a choice of OpenSSL or GNUTLS, for example.) Apparently neither OpenSSL nor cURL wants to provide such a Windows binary package, though, so I attempted to build a version of libcurl with SSL support... Sigbjorn's instructions are definitely out of date. (I couldn't use his ZIP package, either.) Using MSYS, though, I was able to compile a version of libcurl with winssl, but then every variation of cabal install command wasn't cooperating. What should I be using instead of the following, neither of which seems to tell cabal where to find libcurl? ``` $ CPPFLAGS=-Ic:/code/curl_install/include cabal install curl --ld-options=-Lc:/code/curl_install/lib $ cabal install curl --extra-lib-dirs=c:/code/curl_install/lib --extra-include-dirs=c:/code/curl_install/include ``` Detail, showing some verbose error messages: https://gist.github.com/4056331 Kevin

5 8