- Haskell-Cafe - Haskell.org

"doctest" for haskell -- a good project?
by Shaun Cutts 23 Mar '08

23 Mar '08

Hello, I am an experienced programmer, currently learning Haskell. Currently I write many things in python. I use both the "doctest" and "unittest" modules extensively. As I write code, I simultaneously write doctest code in the doc strings to explain/set out the "typical narrative" of how the code is used. Then finishing off a module I write unittests for boundary conditions, more complex test cases, and generally code that would be annoying to write & read in doctests. I note that there is a unit testing framework for Haskell, but I don't see any doctest module. Might this be a good project? If so, suggestions as to resources would be greatly appreciated. I believe I can't just "introspect" Haskell modules to get at documentation/comments, like I can in python? (Why not? :)) I notice that there are a few "documentation generators". Should I try to write an extension of one of these? Haddock, for instance? Are there any Haddock developers hanging out on this list, to encourage or dissuade me? :) (And where is the Haddock doc for Haddock?) In any case, thanks in advance for any comments & advice. - Shaun Cutts

9 15

Re: Haskell-Cafe Digest, Vol 55, Issue 30
by Deng Chao 23 Mar '08

23 Mar '08

Hi Sebastian Sylvan, Thank you very much! I thick the last column return a string because of the Haskell "Show" function, for when I made test, I got such an error message: Show ([SqlValue] -> IO [[SqlValue]]). Any way, thank you very much! Deng Chao 在 2008-03-22六的 13:55 -0400，haskell-cafe-request@haskell.org写道： > Send Haskell-Cafe mailing list submissions to > haskell-cafe(a)haskell.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://www.haskell.org/mailman/listinfo/haskell-cafe > or, via email, send a message with subject or body 'help' to > haskell-cafe-request(a)haskell.org > > You can reach the person managing the list at > haskell-cafe-owner(a)haskell.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Haskell-Cafe digest..." > > > Today's Topics: > > 1. compile error with FFI code (Galchin Vasili) > 2. "doctest" for haskell -- a good project? (Shaun Cutts) > 3. Re: [GSoC] Plugins for GHC (Bulat Ziganshin) > 4. Re: "doctest" for haskell -- a good project? (Roman Cheplyaka) > 5. Re: "doctest" for haskell -- a good project? (Don Stewart) > 6. Re: [GSoC] Plugins for GHC (Neil Mitchell) > 7. Re: "doctest" for haskell -- a good project? (Neil Mitchell) > 8. Re: An ugly zip3 problem.. (Michael Feathers) > 9. Re: An ugly zip3 problem.. (Michael Feathers) > 10. How to program with sqlite? (Deng Chao) > 11. Re: "doctest" for haskell -- a good project? (Thomas Schilling) > 12. Re: How to program with sqlite? (Sebastian Sylvan) > 13. Re: "doctest" for haskell -- a good project? (Sebastian Sylvan) > 14. HXT and types in Control.Arrow.ArrowTree (Robert Vollmert) > 15. Opening Windows .lnk Files (Dominic Steinitz) > 16. RE: "doctest" for haskell -- a good project? (Shaun Cutts) > 17. Re: An ugly zip3 problem.. (Bryan O'Sullivan) > 18. Re: "doctest" for haskell -- a good project? (Paul Johnson) > 19. Re: AMQP framing layer: design help requested. (Paul Johnson) > 20. Re: Problem with OpenAL (Antoine Latter) > 21. Re: AMQP framing layer: design help requested. (Derek Elkins) > 22. Re: Opening Windows .lnk Files (Neil Mitchell) > 23. Re: "doctest" for haskell -- a good project? (Neil Mitchell) > 24. Re: Opening Windows .lnk Files (Dominic Steinitz) > 25. RE: "doctest" for haskell -- a good project? (Shaun Cutts) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Fri, 21 Mar 2008 23:21:26 -0500 > From: "Galchin Vasili" <vigalchin(a)gmail.com> > Subject: [Haskell-cafe] compile error with FFI code > To: Haskell-cafe <haskell-cafe(a)haskell.org> > Cc: Galchin Vasili <vigalchin(a)gmail.com> > Message-ID: > <5ae4f2ba0803212121w17fe482ek105c13cf6999e603(a)mail.gmail.com> > Content-Type: text/plain; charset="iso-8859-1" > > Hello, > > In my blah.hsc, I have "allocbytes (#const (struct bozo)) .. where > > struct bozo is a bunch of "long int" ,,, In the "runhaskell Setup.hs build" > step I get a nasty error message about an "incomplete type". I have look at > many times but this error doesn't make sense to me. ?? > > Kind regards, Vasili >

1 0

ANNOUNCE: wxHaskell 0.10.3 rc1
by Jeremy O'Donoghue 23 Mar '08

23 Mar '08

We are pleased to announce the first release candidate of wxHaskell 0.10.3. This is the first update with binary packages available since June 2005, and is the result of a great deal of work by a new team of contributors. We are hoping to make a full release shortly, and issues and bug reports either through the wxHaskell user mailing list (wxhaskell-users(a)sourceforge.net) or via the Sourceforge bug tracker (http://sourceforge.net/tracker/?group_id=73133&atid=536845) Highlights of 0.10.3 rc1 include: - Support for Unicode builds of wxWidgets - Support for additional widgets including calendar, toolbar divider, styled text control (wxScintilla), media control - Support for clipboard, drag and drop - Support for 64bit (Linux) targets - Support for wxWidgets 2.6.x (support for wxWidgets 2.4.2 retained if you compile from source) - Support for building with GHC 6.6.x and 6.8.x - Parts of wxHaskell are now built with Cabal - New test cases - Removed support GHC version < 6.4 - Profiling support - Smaller generated binary sizes (using --split-objs) Binary packages are available from the wxHaskell download site at http://sourceforge.net/project/showfiles.php?group_id=73133, for the following platforms: - Debian - Windows - OS X (Intel and PPC platforms) - Source code .tar.gz and .zip - Documentation (cross-platform) About wxHaskell wxHaskell is a Haskell binding to the wxWidgets GUI library. It provides a native look and feel on Windows, OS X and Linux, and a medium level programming interface. The main project page for wxHaskell is at http://wxhaskell.sourceforge.net. The latest source code for wxHaskell can always be obtained from http://darcs.haskell.org/wxhaskell. There are developer (wxhaskell-devel(a)lists.sourceforge.net and user (wxhaskell-users(a)lists.sourceforge.net) mailing lists, and a wiki page at http://haskell.org/haskellwiki/WxHaskell which can provide more information to those interested. wxHaskell was originally created by Daan Leijen. The contributors to this new release include: - Eric Kow - shelarcy - Arie Middelkoop - Mads Lindstroem - Jeremy O'Donoghue - Lennart Augustson -- Jeremy O'Donoghue jeremy.odonoghue(a)gmail.com

7 6

compile error with FFI code
by Galchin Vasili 23 Mar '08

23 Mar '08

Hello, In my blah.hsc, I have "allocbytes (#const (struct bozo)) .. where struct bozo is a bunch of "long int" ,,, In the "runhaskell Setup.hs build" step I get a nasty error message about an "incomplete type". I have look at many times but this error doesn't make sense to me. ?? Kind regards, Vasili

2 1

Using HaskellDb to connect to PostgresSql
by Marc Mertens 22 Mar '08

22 Mar '08

Hello, I'm trying to learn to use HaskellDb. I have managed to finally compile and install it on my linux box (I have ghc 6.8.2). But when I try to create a database description (as described in http://haskelldb.sourceforge.net/getting-started.html) (using DBDirect-dynamic instead of DBDirect) I'm stuck. I have tried different names for the driver but I always get the message 'DBDirect-dynamic: user error (Missing field driver)'. Has someone get this working for PostgresSQL and if so, can he/she help me out with some instructions. Also if someone has some pointers to more recent documentation of HaskellDB, I would really be pleased if I could get these. Marc Mertens

3 2

Opening Windows .lnk Files
by Dominic Steinitz 22 Mar '08

22 Mar '08

Does anyone know how to do this? If I open a file on Windows e.g. "asn_application.h.lnk" then I get the link data rather than the data in the linked file. Thanks, Dominic.

2 2

Re: [Haskell-cafe] AMQP framing layer: design help requested.
by Paul Johnson 22 Mar '08

22 Mar '08

Dean Herington wrote: > At 6:41 PM -0700 3/21/08, Adam Langley wrote: >> >> Also >>> getter <- fmap (amqpGetTable !) getWord8 >>> getter >> >> is just >> >> > fmap (amqpGetTable !) getWord8 > > I don't think so. Aren't there two "gettings": the first to get the > "type" byte and the second to get the item? > Yes. I didn't use it because it seemed obfuscated, but in fact the point-free version is > fmap (amqpGetTable !) getWord8 >>= id And I've also got an AmqpWire class similar to Dean's AmqpValue class. But both of these miss the point (which is why I didn't clutter up my simplified explanation with them). I'm looking for an alternative to the honking big AmqpVariant and AmqpArray types. I think I want something along the lines of: > class (Binary v) => VariantClass v where > typeCode :: v -> Word8 > fromVariant :: AmqpVariant -> Maybe v > > newtype AmqpVariant = forall a . (VariantClass a) => Variant a > > newtype AmqpArray = forall a . (VariantClass a) => AmqpArray [a] But I can't see how to write "fromVariant". I'm also unsure what the "amqpGet" and "amqpPut" methods might look like. Or maybe these two types are actually the best design. They do let you pattern-match on the contents. Extracting the contents of a variant from the design above would be something like: > processVariant (Variant v) = > case typeCode v of > 0x01 -> processWord8 $ fromJust $ fromVariant v > 0x02 -> ... and so on. Compare this with: > processVariant (VariantWord8 v) = processWord8 v > processVariant (VariantWord16 v) = processWord16 v > ... and so on. What do you think? Paul.

2 1

[GSoC] Plugins for GHC
by Max Bolingbroke 22 Mar '08

22 Mar '08

Hi, I'm interested in working on a plugins system for the Glasgow Haskell Compiler. The idea here is to allow third parties to add more capabilities to the compiler in a way that doesn't involve messing around with modifying GHC itself. Potential use cases are: * Allow Haskell code to be selectively run on GPUs * Idiosyncratic application-specific optimization passes * And whatever clever ideas the community comes up with! This project has previously been proposed as an Microsoft Research internship project (http://hackage.haskell.org/trac/ghc/wiki/Internships).I have a preliminary wiki page with details at http://hackage.haskell.org/trac/ghc/wiki/Plugins I'd particularly like to get feedback from the list on: * Other design goals people would like to see * Improvements to the proposed (very half-baked!) user interface * Other hooks people think they might like to have into the compiler * Interested mentors. Simon PJ has indicated he would be interested in mentoring this, but there might be other projects that need him more? This project would satisfy a dependency that Manuel mentioned in http://groups.google.com/group/fa.haskell/msg/3ca260fb075a901e to implement an "array DSL" for GPUs in Haskell. It would mean GHC taking a dependency on hs-plugins (http://www.cse.unsw.edu.au/~dons/hs-plugins/) as that's our current strategy to allow GHC to dynamically link with the plugins. A little about my experience with GHC: this year I worked on implementing comprehensive comprehensions (http://research.microsoft.com/~simonpj/papers/list-comp/index.htm) as my final year undergraduate project, supervised by Simon PJ. I have subsequently implemented a few other patches, adding some core-level optimizations / bug fixes (some of which are pending review). Thanks in advance for your comments! Max

3 2

AMQP framing layer: design help requested.
by Paul Johnson 22 Mar '08

22 Mar '08

I'm trying to implement the AMQP framing layer. This is the bottom layer of the protocol that packs and unpacks data in lazy ByteStrings. I'm using the Binary package with its "Get" and "Put" monads to do the unpacking, and that works well. But I've run into a problem and I can't find an elegant solution. I'm wondering if someone can help. The following contains a simplified version of my current design. I've mapped the AMQP basic types to Haskell types. Sometimes this is a basic mapping, sometimes to something more complex. Strings have several AMQP representations, so I've created corresponding newtype declarations and instances. Some AMQP types can contain tagged types. For instance a "map" contains (key, tag, value) triples, where the tag is a single byte indicating the type of the value. I've defined the following type: data AmqpVariant = AmqpVarWord8 Word8 | AmqpVarWord16 Word16 | AmqpVarStr8 Str8 -- And so on for about 20 more types instance Binary AmqpVariant where put (AmqpVarWord8 v)= putWord8 0x01 >> putWord8 v put (AmqpVarWord16 v) = putWord8 0x02 >> putWord16be v put (AmqpVarStr8 v) = putWord8 0x03 >> put v -- And so on for about 20 more types get = do getter <- fmap (amqpGetTable !) getWord8 getter amqpGetTable :: Array Word8 (Get AmqpVariant) amqpGetTable = array (0,255) [ (0x01, fmap AmqpVarWord8 getWord8), (0x02, fmap AmqpVarWord16 getWord16be), (0x03, fmap AmqpVarStr8 get), -- And so on for about 20 more types ] This is a Brute Force and Ignorance solution, but it will probably work. But now I've come up against another type, which AMQP calls an "array". This has a single type tag at the start, followed by N items of that type. I'm wondering how to do this. One option would be to declare data AmqpArray = AmqpArrWord8 [Word8] | AmqpArrWord16 [Word16] -- and so on This effectively duplicates the AmqpVariant solution. It would work, but it feels horrible. I get the feeling that there ought to be a better way, possibly using classes or existential types. Any suggestions? Thanks, Paul.

2 2

ANN: Takusen 0.8.1
by Alistair Bayley 21 Mar '08

21 Mar '08

http://hackage.haskell.org/cgi-bin/hackage-scripts/package/Takusen-0.8.1 This is a fairly minor bugfix release. although there are some API changes. Mainly, we've re-exported a lot of the types from Database.InternalEnumerator in Database.Enumerator. This will hopefully address a common complaint that it's impossible to always write type sigs for functions doing database stuff. We've also tidied up some of the exports; there was stuff exposed that shouldn't have been. There's an outstanding bug involving datetime parameter marshaling in ODBC, which I have not had time to resolve (parameters after the datetime parameter get mangled somehow). I will be on holiday for a few weeks, so this will not be addressed by me for some time. Caveat user. The test suite highlights it, should someone else feel sufficiently motivated; I've been testing against PostgreSQL. Alistair

2 1