November 2010 - Haskell-Cafe

ANNOUNCE: iteratee-compress 0.1.2
by Maciej Piechotka 28 Nov '10

28 Nov '10

Iteratee-compress provides compressing and decompressing enumerators including flushing. Currently only gzip is provided but at bzip and LZMA are planned. This is bug-fixing release Changes from previous version: - Fix infinite loop/segfault bug - Fix bug in which part of the output was lost Next goals: - BZip support - LZMA support - Generic interface for flushing To think about: - Should inner iteratee be able to request flushing? Regards

1 0

Return stack buffer vs Haskell compilers
by Florian Weimer 28 Nov '10

28 Nov '10

Are there any Haskell compilers which use a calling convention which is friendly to return stack buffers found in many modern CPUs? Such a calling convention uses CALL/RET pairs for function calls. Tail calls are implemented with JMP and may require shuffling the stack, including the return address, if the number of arguments passed on the stack differs between tail caller and callee. Such an approach means that the CPU can use the return stack buffer, which allows exact prediction of the jump target of the RET instruction (as long as the return stack buffer does not overflow). In contrast, if an indirect jump is used to implement RET, it enters the same prediction logic as any other indirect call, which is less precise and affects predicition of other indirect calls. This is especially important on the Intel Atom and VIA Nano CPUs, which are reported (Agner Fog, "The microarchitecture of Intel, AMD and VIA CPUs", last updated 2010-09-25) to have a return stack buffer, but a very simple indirect branch predictor which uses the last target as the next predicted target.

1 0

Why does cabal sdist still make tarballs with incorrect file modes?
by John D. Ramsdell 28 Nov '10

28 Nov '10

As I played with GHC 7.0.1, I noticed that Cabal sdist is still hosed. Since ten months ago, cabal sdist fails to preserve the file mode bits of the source files put into the tarball being generated (Ticket #627 reported by draconx). An executable shell script and a source file world readable both have a mode of 600 in the tarball. Does anyone know why this bug is difficult to fix? Why can't one revert to the code used in Cabal 1.6.0.1, which is know to work? John

1 0

Bulletproof resource management
by Florian Weimer 28 Nov '10

28 Nov '10

At least in my experience, in order to get proper resource management for things like file or database handles, you need both a close operation and a finalizer registered with the garbage collector. The former is needed so that you can create resources faster than the garbage collector freeing them. The latter is a safety net in cases where proper resource management is not follwoed (perhaps in GHCi). When the explicit close operation or the finalizer has been invoked, the object must somehow be disabled, so that further operations on it fail (otherwise, you might dereference a dangling pointer). What's the idom for implementing this in Haskell (or GHC)? It seems that a ForeignPtr cannot be written to (otherwise I'd change it to a null pointer when the resource is freed). It's also not possible to check if the finalizers have been run. Can the finalizers hold a reference to the object which in turn holds the ForeignPtr to be finalized? Or would that prevent the finalizers from running at all? Is there a way to avoid the extra level of indirection (or is GHC sufficiently smart to optimize it away)?

6 9

GHC Out of memory
by John Smith 28 Nov '10

28 Nov '10

I get an out-of-memory error about two thirds of the way through building qtHaskell. If I start again, the remaining modules are compiled successfully, although GHC's memory usage creeps up to about 1.5Gb by the time it's finished. Is there a memory leak? (Compiling on Win64 with 4Gb RAM.)

3 4

qtHaskell signal and slots
by c8h10n4o2 27 Nov '10

27 Nov '10

I have the following code that tries to resize window after a click. But nothing happens. If I use a builtin slot that does not need a function, such as "close", "quit" it works, however for any other slot it seems my function is ignored. Which slot/signal should I use and how? module Main where import Qtc.Classes.Qccs import Qtc.Classes.Gui import Qtc.Gui.Base import Qtc.Gui.QApplication import Qtc.Gui.QPushButton import Qtc.Core.QFile import Qtc.Tools.QUiLoader import Qtc.Enums.Core.QIODevice import Qtc.Core.Base import Qtc.ClassTypes.Gui import Qtc.Gui.QAbstractButton main = do app <- qApplication () loader <- qUiLoader () uiFile <- qFile "slide.ui" open uiFile fReadOnly ui <- load loader uiFile close uiFile () add <- findChild ui ("<QPushButton*>", "pushButton") connectSlot add "clicked()" add "click()" $ t ui qshow ui () qApplicationExec () t :: QWidget () -> QWidget () -> IO () t ui _= resize ui (500::Int ,500::Int) -- View this message in context: http://haskell.1045720.n5.nabble.com/qtHaskell-signal-and-slots-tp3282748p3… Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.

1 0

Is EclipseFP alive?
by Dušan Kolář 27 Nov '10

27 Nov '10

Hello, I tried to install EclipseFP using instructions on http://eclipsefp.sourceforge.net/, nevertheless, without any success - after downloading scion I cannot build it > cabal install Resolving dependencies... cabal: dependencies conflict: ghc-6.12.3 requires Cabal ==1.8.0.6 however Cabal-1.8.0.6 was excluded because scion-0.1.0.6 requires Cabal ==1.6.* Running Setup.hs through runhaskell seems not to work as well: > runhaskell Setup.hs configure Configuring scion-0.1.0.6... Setup.hs: At least the following dependencies are missing: AttoJson >=0.5.2, binary ==0.5.*, derive -any, ghc-paths ==0.1.*, ghc-syb -any, ghc-syb-utils -any, hslogger ==1.0.*, list-tries -any, multiset >=0.1 && <0.3, uniplate -any Even if, for instance: > cabal info binary * binary (library) Synopsis: Binary serialisation for Haskell values using lazy ByteStrings Latest version available: 0.5.0.2 Latest version installed: 0.5.0.2 Is there any chance to make it install or should I give up? Dusan

2 1

Language.C++ ?
by Jacques Carette 27 Nov '10

27 Nov '10

Is anyone working on a Language.C++ module, similar to the wonderful Language.C ? Right now just the syntax and a pretty-printer would be really nice to have... Jacques

1 0

Re: Template Haskell: exchanging information between 'splicers'
by oleg＠okmij.org 27 Nov '10

27 Nov '10

Jean-Christophe Mincke wrote: > Does anyone know a clean solution to pass information between 2 executions > of splicers. > > $(splicer ....) -- first invocation gather and store some data > .... > $(splicer ...) -- second one use the data gathered above. Just the other day I wrote such code; I'm not sure it qualifies as a `clean solution' as it was meant for a nefarious purpose. > e2 = runIO $ do > varref <- newIORef undefined > let put v = runIO $ do > putStrLn "put" > writeIORef varref v > runQ [|$v + 1|] > let get v = runIO $ do > putStrLn "get" > vold <- readIORef varref > runQ [|$v + $vold|] > > c1 <- runQ [|\x -> $(put [|x|]) + 1::Int|] > c2 <- runQ [|\x -> $(get [|x|]) + 1::Int|] > runQ $ [| ($(return c1),$(return c2)) |] > > show_code cde = runQ cde >>= putStrLn . pprint > te2 = show_code e2 One may even try to splice e2 at the top level (in a different module). The code does exchange data between two splicers -- one may argue, it exchanges too much data.

1 0

version of findIndex that works with a monadic predicate
by José Romildo Malaquias 26 Nov '10

26 Nov '10

Hello. I need a function findIndexM, similar to findIndex from the standard module Data.List, but which works with a monadic predicate to test list elements. findIndex :: (a -> Bool) -> [a] -> Maybe Int findIndexM :: (Monad m, Num a) => (t -> m Bool) -> [t] -> m (Maybe a) findIndexM p xs = go 0 xs where go _ [] = return Nothing go n (x:xs) = do res <- p x if res then return (Just n) else go (n+1) xs How can this function be rewritten using combinators? Romildo

3 2