- Haskell-Cafe - Haskell.org

Using _ on the RHS of an equation?
by Jason Dagit 06 Apr '11

06 Apr '11

I think I got this idea from JHC, but I'm unable to find it documented in the JHC manual at the moment. The idea is that when _ appears on the RHS of an equation it is another name for undefined. Here is a hypothetical example, inspired by code I've seen in FFI definitions: \begin{code} {-# LANGUAGE EmptyDataDecls #-} data Bar instance Storable Bar where sizeOf _ = sizeOf #{size struct Bar} alignment _ = alignment (_ :: CDouble) -- here we could have just as succinctly said alignment (1 :: CDouble) data Foo -- the struct Foo on the C side contains a Bar instance Storable Foo where sizeOf _ = sizeOf #{size struct Foo} alignment _ = alignment (_ :: Bar) -- we can't instantiate a Bar \end{code} Is this something people would be interested in having as an extension in GHC? Or is it just too fluffy for anyone to really care? Thanks, Jason

4 5

Re: [Haskell-cafe] ANN: stm-chans: Additional types of channels for STM.
by wren ng thornton 06 Apr '11

06 Apr '11

On 4/3/11 11:58 PM, Thomas DuBuisson wrote: > Wren, > Glad to see someone is doing a more complete packaging of STM helpers > and derivatives! > > I've done a little work on bounded TChans[1] (hackage "bounded-tchan" > package) and I think you should consider a few things: Ah, somehow I missed that in my searching. I'll give it a look and try to integrate things. > 1) Split the reader counter and writer counters as I've done in > bounded-tchan. This gives 2-5 times better performance (I benchmarked > using Criterion, a single reader, a single writer, and tested channels > with bounds of 10, 100, and 1000 elements). That's crazy. I wonder why... must be a contention issue... > 2) Implement a "tryWriteTBChan" for obvious reasons. Good idea. > 4) Help me figure out a good way to build dupTBChan [2]. It's not easy. I've yet to figure out a way of doing this that has the right semantics. Perhaps splitting the reader and writer counters will help, though it sounds like you haven't had much luck either. > [2] http://stackoverflow.com/questions/5446484/how-to-add-a-finalizer-on-a-tvar Have you taken a look at TwilightSTM? http://proglang.informatik.uni-freiburg.de/projects/syncstm/ http://hackage.haskell.org/package/twilight-stm Since I'm aiming to have stm-chans eventually incorporated into stm proper, that wouldn't work for me, though it sounds like it might solve the ideological question about adding finalizers to TVars -- Live well, ~wren

1 0

Problems porting ghc6-6.10.1 to GNU/Hurd using 6.8.2
by Svante Signell 05 Apr '11

05 Apr '11

Hi, I'm currently trying to upgrade Debian GNU/Hurd ghc6.8.2 to 6.10.1 (and further from there, due to requirements of ghc and haddock versions), and get stuck on configuring the random-1.0.0.1 library in stage1. I have tried versions of ghc6 are 6.10.1+dfsg1-1 to 6.10.1+dfsg1-5 (the latest not requiring ghc6.10 due to haddock 2.4.2). The commands are shown below: (added the -verbose=3 parameter) cd ghc6-6.10.1+dfsg1/libraries (sorry for the missing \ due to cut-and-paste to the mailer) ( cd random && /home/srs/DEBs/ghc/ghc6-6.10.1+dfsg1/libraries/cabal-bin --verbose=3 /usr/bin/ghc6 /home/srs/DEBs/ghc/ghc6-6.10.1 +dfsg1/libraries/bootstrapping.conf configure --verbose=3 --prefix=/NONEXISTENT --bindir=/NONEXISTENT --libdir=/NONEXISTENT --libexecdir=/NONEXISTENT --datadir=/NONEXISTENT --docdir=/NONEXISTENT --haddockdir=/NONEXISTENT --htmldir=/NONEXISTENT --with-compiler=/home/srs/DEBs/ghc/ghc6-6.10.1 +dfsg1/ghc/stage1-inplace/ghc --with-hc-pkg=/home/srs/DEBs/ghc/ghc6-6.10.1 +dfsg1/utils/ghc-pkg/install-inplace/bin/ghc-pkg --libsubdir='$pkgid' --with-gcc=gcc --with-ld=/usr/bin/ld --with-hscolour=/usr/bin/HsColour --with-alex=/usr/bin/alex --with-happy=/usr/bin/happy --configure-option='--build' --configure-option='i486-gnu' --configure-option='--prefix=/usr' --configure-option='--with-ghc=/usr/bin/ghc6' --configure-option='build_alias=i486-gnu' --configure-option='CFLAGS=-g -O2' --configure-option='LDFLAGS=' --configure-option='CPPFLAGS=' --configure-option=--with-cc=gcc --with-hsc2hs=/home/srs/DEBs/ghc/ghc6-6.10.1 +dfsg1/utils/hsc2hs/install-inplace/bin/hsc2hs --haddock-options="--use-contents=../index.html --use-index=../doc-index.html" --enable-library-profiling --enable-split-objs ) && touch stamp/configure.library.build-profiling-splitting.random || touch random/unbuildable Output: ======= Creating dist (and its parents) ("/home/srs/DEBs/ghc/ghc6-6.10.1\ +dfsg1/ghc/stage1-inplace/ghc",["--numeric-version"]) /home/srs/DEBs/ghc/ghc6-6.10.1+dfsg1/ghc/stage1-inplace/ghc is version 6.10.1 ("/home/srs/DEBs/ghc/ghc6-6.10.1\ +dfsg1/utils/ghc-pkg/install-inplace/bin/ghc-pkg",["--version"]) /home/srs/DEBs/ghc/ghc6-6.10.1\ +dfsg1/utils/ghc-pkg/install-inplace/bin/ghc-pkg is version 6.10.1 ("/home/srs/DEBs/ghc/ghc6-6.10.1\ +dfsg1/ghc/stage1-inplace/ghc",["--supported-languages"]) Reading installed packages... ("/home/srs/DEBs/ghc/ghc6-6.10.1\ +dfsg1/utils/ghc-pkg/install-inplace/bin/ghc-pkg",["dump","--global"]) Hangs here! Running manually works OK!! /home/srs/DEBs/ghc/ghc6-6.10.1\ +dfsg1/utils/ghc-pkg/install-inplace/bin/ghc-pkg "dump" "--global" Gives a lot of output here... Below is the ghc-pkg script: cat /home/srs/DEBs/ghc/ghc6-6.10.1\ +dfsg1/utils/ghc-pkg/install-inplace/bin/ghc-pkg PKGCONF=/home/srs/DEBs/ghc/ghc6-6.10.1\ +dfsg1/inplace-datadir/./package.conf exec /home/srs/DEBs/ghc/ghc6-6.10.1\ +dfsg1/utils/ghc-pkg/install-inplace/libexec/ghc-pkg --global-conf\ $PKGCONF ${1+"$@"} I've tried changing the shell from dash to bash but no changes. I also tried to escape some of the variables in the gkc-pkg shell script but without any luck. Anybody can help me here??

2 2

Encoding of Haskell source files
by Roel van Dijk 05 Apr '11

05 Apr '11

Hello, The Haskell 2010 language specification states that "Haskell uses the Unicode character set" [1]. I interpret this as saying that, at the lowest level, a Haskell program is a sequence of Unicode code points. The standard doesn't say how such a sequence should be encoded. You can argue that the encoding of source files is not part of the language. But I think it would be highly practical to standardise on an encoding scheme. Strictly speaking it is not possible to reliably exchange Haskell source files on the byte level. If I download some package from hackage I can't tell how the source files are encoded from just looking at the files. I propose a few solutions: A - Choose a single encoding for all source files. This is wat GHC does: "GHC assumes that source files are ASCII or UTF-8 only, other encodings are not recognised" [2]. UTF-8 seems like a good candidate for such an encoding. B - Specify encoding in the source files. Start each source file with a special comment specifying the encoding used in that file. See Python for an example of this mechanism in practice [3]. It would be nice to use already existing facilities to specify the encoding, for example: {-# ENCODING <encoding name> #-} An interesting idea in the Python PEP is to also allow a form recognised by most text editors: # -*- coding: <encoding name> -*- C - Option B + Default encoding Like B, but also choose a default encoding in case no specific encoding is specified. I would further like to propose to specify the encoding of haskell source files in the language standard. Encoding of source files belongs somewhere between a language specification and specific implementations. But the language standard seems to be the most practical place. This is not an official proposal. I am just interested in what the Haskell community has to say about this. Regards, Roel [1] - http://www.haskell.org/onlinereport/haskell2010/haskellch2.html#x7-150002.1 [2] - http://www.haskell.org/ghc/docs/7.0-latest/html/users_guide/separate-compil… [3] - http://www.python.org/dev/peps/pep-0263/

17 28

Re: [Haskell-cafe] Using _ on the RHS of an equation?
by Paul Keir 05 Apr '11

05 Apr '11

Hi Jason, I like the idea. I've seen some code from Oleg Kiselyov which uses __ (two underscores) in this way. The thing that stops me though, is when I get it wrong, and undefined fires somewhere, but I don't know where. Something like you propose, but with a line number, would be sweet. Paul The University of Glasgow, charity number SC004401

3 2

broken on build encoding
by Gang Yu 05 Apr '11

05 Apr '11

Hi, cafe, I always get the following build failure when I try to install encoding: Configuring encoding-0.6.4... Warning: This package indirectly depends on multiple versions of the same package. This is highly likely to cause a compile failure. package haskell98-1.0.1.1 requires random-1.0.0.2 package HaXml-1.20.2 requires random-1.0.0.2 package random-1.0.0.2 requires time-1.1.4 package random-1.0.0.2 requires time-1.2.0.4 Preprocessing library encoding-0.6.4... Building encoding-0.6.4... Data/Encoding/ByteSource.hs:124:0: warning: extra tokens at end of #ifndef directive Data/Encoding/ByteSink.hs:104:0: warning: extra tokens at end of #ifndef directive [ 1 of 65] Compiling Data.Static ( Data/Static.hs, dist/build/Data/Static.o ) [ 2 of 65] Compiling Data.Array.Static ( Data/Array/Static.hs, dist/build/Data/Array/Static.o ) Data/Array/Static.hs:14:35: Warning: Bindings containing unlifted types must use an outermost bang pattern: { (I# ri) = index (s, e) i } *** This will be an error in GHC 6.14! Fix your code now! In the expression: let (I# ri) = index (s, e) i in extract addr ri In the definition of `!': ! (StaticArray s e addr) i = let (I# ri) = index ... i in extract addr ri [ 3 of 65] Compiling Data.Map.Static ( Data/Map/Static.hs, dist/build/Data/Map/Static.o ) [ 4 of 65] Compiling Control.Throws ( Control/Throws.hs, dist/build/Control/Throws.o ) [ 5 of 65] Compiling Data.Encoding.Exception ( Data/Encoding/Exception.hs, dist/build/Data/Encoding/Exception.o ) [ 6 of 65] Compiling Data.Encoding.ByteSource ( Data/Encoding/ByteSource.hs, dist/build/Data/Encoding/ByteSource.o ) Data/Encoding/ByteSource.hs:151:17: Not in scope: data constructor `State' cabal: Error: some packages failed to install: encoding-0.6.4 failed during the building phase. The exception was: ExitFailure 1https://mail.google.com/mail/?shva=1#compose I manually installed a ghc-6.12.3, a ubuntu ghc-6.12.1, all the same? anybody can provide helps? thanks Gang

3 7

ANN: Google Summer of Code student application period opens today
by Johan Tibell 05 Apr '11

05 Apr '11

Hi, The Google Summer of Code student application period starts today at 19:00 UTC. If you're a student and like to get paid to work on a Haskell project this summer I recommend you go find an interesting project [1] and start working on your application. You can find more information on the wiki [2]. Cheers, Johan 1. http://hackage.haskell.org/trac/summer-of-code/report/1 2. http://hackage.haskell.org/trac/summer-of-code/wiki/Soc2011

4 5

lhs2tex and comment
by Mitar 05 Apr '11

05 Apr '11

Hi! I have problems with lhs2tex and nested (long) comments. It formats it all in one long line and does not wrap it. Is there a way to define those comments so that it would maybe automatically wrap them at the end of the page? (And still take into the account possible indentation of comments.) Or at least to not ignore newlines in comments? I am using poly mode. Mitar

1 0

why is ghci trying to load hsc file ??
by briand＠aracnet.com 04 Apr '11

04 Apr '11

Howdy, I worked out a small hdf5 binding using cabal and bindings-DSL and sqlite3 as my example. Time to try it ! ghci -idist/build/ dist/build/Bindings/HDF5.o -lhdf5 -lhdf5_hl hdf5_pkg_test.hs GHCi, version 6.12.1: 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 object (static) dist/build/Bindings/HDF5.o ... done Loading object (dynamic) hdf5 ... done Loading object (dynamic) hdf5_hl ... done final link ... done [1 of 2] Compiling Bindings.HDF5 ( dist/build/Bindings/HDF5.hs, interpreted ) src/Bindings/HDF5.hsc:49:8: parse error on input `import' Failed, modules loaded: none. Huh ? Why is it trying to read HDF5.hsc ?? What's even more interesting is that line 49 of that file doesn't have an import on it, so something is fubar. No idea how this could be happening. I've included a copy of my cabal file. BTW. I have to specify the hdf5 libraries, i.e. libhdf5 and libhdf5_hl on the command line. It seems like the build process should have taken care of that in some way, maybe... ? Certainly when I use something like sqlite3, I'm not specifying libsqlite3 on the command line. Thanks, Brian

7 11

Fast conversion between Vector Double and Vector CDouble
by Bas van Dijk 04 Apr '11

04 Apr '11

Hello, I just rewrote my levmar library[1] to use Vectors (from the vector package[2]) instead of lists. I was expecting to see a significant performance improvement. Unfortunately I only saw a 10% improvement. However, I noticed I had a lot of conversions from Vector Double to Vector CDouble and visa versa in my code: import Data.Vector.Storable ( Vector ) mapRealToFrac ∷ (Storable α, Storable β, Real α, Fractional β) ⇒ Vector α → Vector β mapRealToFrac = VS.map realToFrac When I replace this with: mapRealToFrac = unsafeCoerce My application computes the same result but does it 28 times faster! My question are: 1) Is this always safe? In other words: are the runtime representations of Double and CDouble always equivalent or do they vary between platforms? 2) Can the same improvement be accomplished using RULE pragma's? 3) Are there disadvantages of using CDouble instead of Double in the levmar API? For some reason it feels wrong to use CDouble in the API but I don't have a good argument against it yet. Thanks, Bas [1] http://hackage.haskell.org/package/levmar [2] http://hackage.haskell.org/package/vector

4 4