- Haskell-Cafe - Haskell.org

ANN: bindings-DSL 1.0.4 (Category: FFI)
by Maurício CA 29 Jan '10

29 Jan '10

Hi, bindings-DSL is a mature and well documented preprocessor domain specific language you can use to generate bindings to a C API. It's based on functionality provided by hsc2hs. These are links to Hackage page and documentation: http://hackage.haskell.org/package/bindings-DSL http://bitbucket.org/mauricio/bindings-dsl New in this version: * Support for C unions. * Use of <inttypes.h> macros for wider portability. An example on the use of unions follow. Suppose you have: union example { int k; char c[4]; }; Then, using bindings-DSL, you'll write: #starttype union example #union_field k , CInt #union_array_field c , CChar #stoptype Here is a ghci session showing the resulting code: > let value = C'example 1 [] > u'example'k value (2^10) C'example {c'example'k = 1024, c'example'c = [0,4,0,0]} > u'example'c value [1,0,0,0] C'example {c'example'k = 1, c'example'c = [1,0,0,0]} > u'example'c value [0,0,0,1] C'example {c'example'k = 16777216, c'example'c = [0,0,0,1]} Peeking an uninitialized memory address: > v <- alloca $ \p -> peek p :: IO C'example > v C'example {c'example'k = 81842189, c'example'c = [13,-48,-32,4]} Hope it's useful to you. Best, Maurício

2 4

ANNOUNCE: ThreadScope 0.1
by Satnam Singh 28 Jan '10

28 Jan '10

I've just released ThreadScope version 0.1 on Hackage. Threadscope is a graphical utility for viewing profiling information about Haskell threads. It was written jointly with Simon Marlow and Donnie Jones. It uses Gtk2HS so it works under Windows and the L-word operating system although there seem to be problems making it work with OS-X (due to Gtk2Hs issues I think). $ cabal install threadscope You may need to do cabal update to update your packages list. On Windows operating systems you may need to issue the command from a shell that is running with administrative privileges. Resources: * Web page: http://research.microsoft.com/threadscope * Parallel Performance Tuning for Haskell<http://research.microsoft.com/apps/pubs/default.aspx?id=80976> (paper at Haskell Symposium 2009) Please let me know if you have any problems. Also, please do let us know about your experience with ThreadScope and in particular about accounts of how you used ThreadScope to shed light on parallel performance bugs. If I get enough feedback I may collate the responses into an experience report or paper for ICFP or the Haskell Symposium. Thank you kindly. Enjoy! Cheers, Satnam ________________________________ Satnam Singh Microsoft 7 JJ Thomson Avenue Cambridge CB3 0FB United Kingdom Email: satnams(a)microsoft.com<mailto:satnams@microsoft.com> UK tel: +44 1223 479905 Fax: +44 1223 479 999 UK mobile: +44 7979 648412 USA cell: 206 330 1580 USA tel: 206 219 9024 URL: http://research.microsoft.com/~satnams Live Messenger: satnam(a)raintown.org<mailto:satnam@raintown.org>

3 3

Re: [Haskell-cafe] scheduling an alarm
by Thomas DuBuisson 28 Jan '10

28 Jan '10

Brian Denheyer <briand(a)aracnet.com> wrote: > On Mon, 25 Jan 2010 23:19:53 -0800 > Thomas DuBuisson <thomas.dubuisson(a)gmail.com> wrote: > > > 1) Don't use System.Posix.Signals > > It isn't necessary and makes your code less portable > > > > 2) The POSIX SIGALRM is used/caught by the RTS and that is why you are > > seeing strange behavior. > > > > 3) Consider using Haskell exceptions from Control.Concurrent > > (throwTo). Not sure what you want to do but you can always > > "myThreadId >>= \tid -> forkIO $ threadDelay someDelayTime >> > > (throwTo tid someExceptionVal)" > > > > I just want a routine to run every 15 min. > You can still use Control.Concurrent: > import Control.Concurrent > > doEvent f usDelay = forkIO $ > threadDelay usDelay > doEvent f usDelay > f Or a wrapper around Control.Concurrent - Ex: The Control-Event package. > I'm not clear on why the throwTo is in you example. > It will give you an exception just like you were expecting from SIGALRM. Thomas

4 8

ghc-core 0.5 build fails
by Brian Denheyer 28 Jan '10

28 Jan '10

Hi all, ghc-core.hs:263:13: Not in scope: data constructor `C.ExitException' Looks like this comes from the base control.exception ? When I go to the web page for control.exception, there is no exitexception. My reading of the hackage entry for ghc-core makes me think that I've got the correct version of everything. ghc is 6.10.4 Brian Creating dist/build/ghc-core (and its parents) Creating dist/build/ghc-core/ghc-core-tmp (and its parents) /usr/bin/ghc -o dist/build/ghc-core/ghc-core --make -hide-all-packages -i -idist/build/ghc-core/ghc-core-tmp -i. -idist/build/autogen -Idist/build/autogen -Idist/build/ghc-core/ghc-core-tmp -optP-include -optPdist/build/autogen/cabal_macros.h -odir dist/build/ghc-core/ghc-core-tmp -hidir dist/build/ghc-core/ghc-core-tmp -stubdir dist/build/ghc-core/ghc-core-tmp -package base-4.1.0.0 -package colorize-haskell-1.0.0 -package directory-1.0.0.3 -package filepath-1.1.0.2 -package pcre-light-0.3.1 -package process-1.0.1.1 -O -Wall -XPatternGuards ./ghc-core.hs

3 3

UTF-16 to UTF-8
by Günther Schmidt 28 Jan '10

28 Jan '10

Hi, is there a library which converts from utf-16 to utf-8? Günther

5 7

Map unionWith generalization
by Hans Aberg 28 Jan '10

28 Jan '10

I need ideally some generalizations of unionWith and unionWithKey, for efficiency matters (i.e. avoiding conversions and traversing the maps more than once). I could use a modification of the code in Map.hs, but then the problem is that the module Map interface does not export the constructors of data Map. So suggestions are welcome. For example, in Map String Integer (sparse representation of monomials) compute the minimum value of all associative pairs with the same key (the gcd); if only one key is present, the absent should be treated as having value 0. So unionWith min xs ys will not work, because unionWith will always apply the identity to the remaining value when one key is missing, whereas it should be sent to 0. So here, one would want: (a -> c) -> (b -> c) -> (a -> b -> c) -> Map k a -> Map k b -> Map k c where the two first functions are applied when the first or second key is missing. Hans

3 8

Why no merge and listDiff?
by Will Ness 28 Jan '10

28 Jan '10

Hello cafe, I wonder, if we have List.insert and List.union, why no List.merge (:: Ord a => [a] -> [a] -> [a]) and no List.minus ? These seem to be pretty general operations. Brief look into haskell-prime-report/list.html reveals nothing. Could we please have them? On the wider perspective, is their a way to declare an /ordered/ list on the type level (e.g. [1,2,3] would be one, but not [2,3,1])? Non-decreasing lists? Cyclical, or of certain length? What are such types called? TIA!

5 12

Supporting GHC 6.10 and 6.12 in HDBC-postgresql Setup.hs
by John Goerzen 27 Jan '10

27 Jan '10

Hi folks, I've gotten some reports (see below) of issues with building HDBC-postgresql on GHC 6.12. Its Setup.hs file [1] is the culprit. The problem is that AnyVersion needs to be removed to work with Cabal in GHC 6.12. But I still want to support older Cabal. Following the directions in the Cabal manual, I tried: #if MIN_VERSION_Cabal(1,8,0) pgconfigProgram (withPrograms lbi) #else pgconfigProgram AnyVersion (withPrograms lbi) #endif While of course adding the needed bits to invoke CPP. This didn't resolve it; apparently Cabal doesn't define macros for use in Setup.hs, only for use in the application. There also seems to be no conditional for use in the .cabal file to resolve this. I could check the GHC version, but that doesn't necessarily correspond to Cabal version. Any ideas? -- John [1] http://git.complete.org/hdbc-postgresql?a=blob;f=Setup.hs;h=0656cb41adc814d…

3 3

wxHaskell - using XRC files
by Günther Schmidt 27 Jan '10

27 Jan '10

Hi, I'm looking for documentation on using XRC files with wxHaskell. I finally managed to cabal-install wxHaskell last night alas sans docs. Günther BTW: I'm using wxFormBuilder, any other good tools out there?

3 4

custom widgets
by Andrew U. Frank 27 Jan '10

27 Jan '10

we used wx originally but switched to gtk. the great advantage is glade... which is very flexible and still you can add/change whatever you need to in gtk in your haskell code. we checked porting between linux (ubuntu) and windows (xp) and encountered no problems.

1 0

Re: classes with types which are wrapped in
by Andrew U. Frank 27 Jan '10

27 Jan '10

i thank for the two suggestions (newtype as a wrapper and type families) and i have explored them. haskell-cafe was very helpful! i use often a piece of code to update a part of a record with field names which codes as follows: data X a b c = X {afield :: a, bfield :: b, cfield :: c} dotoBfield :: (b -> b) -> X a b c -> X a b c dotoBfield op x = x { bfield = op (bfield x)} and similar for A and C. is there a better idiom to achieve the same effect? can this be automated (for example, using generics)?

2 1

Maybe, maybe not.
by michael rice 27 Jan '10

27 Jan '10

Just noticed this difference in the definition of fromMaybe in two different places: http://haskell.org/ghc/docs/latest/html/libraries/base-4.2.0.0/src/Data-May… -- | The 'fromMaybe' function takes a default value and and 'Maybe' -- value. If the 'Maybe' is 'Nothing', it returns the default values; -- otherwise, it returns the value contained in the 'Maybe'. fromMaybe :: a -> Maybe a -> a fromMaybe d x = case x of {Nothing -> d;Just v -> v} and http://en.wikibooks.org/wiki/Haskell/Hierarchical_libraries/Maybe fromMaybe :: a -> Maybe a -> a fromMaybe z = maybe z id Michael

5 8

ghc: unrecognised flags: -I
by Lyndon Maydwell 27 Jan '10

27 Jan '10

Hi Cafe! That's a capital "i" for anyone with font issues. I posted this question to beginners a while ago, but received no meaningful response so I'm trying cafe instead :-) Most packages I try to install off Hackage with cabal are giving me the error "ghc: unrecognised flags: -I". For example, when I "cabal install -v storable-complex" I get the following: /usr/bin/ghc -package-name storable-complex-0.2.1 --make -hide-all-packages -i -idist/build -i. -idist/build/autogen -Idist/build/autogen -Idist/build -I -optP-include -optPdist/build/autogen/cabal_macros.h -odir dist/build -hidir dist/build -stubdir dist/build -package base-3.0.3.1 -O Foreign.Storable.Complex ghc: unrecognised flags: -I I've updated everything several times since this error started occurring, but it persists. I've tried digging to find out where the lone -I is being injected but can't seem to track it down. Has anyone encountered this before, or more realistically, can anyone give me some advice on how to narrow this problem down further? I'm running cabal version 1.6.0.1, ghc 6.10.4 on OS X 10.5. Thanks guys!

2 2

foldl in terms of foldr
by Xingzhi Pan 26 Jan '10

26 Jan '10

On Tue, Jan 26, 2010 at 11:24 PM, Eduard Sergeev <Eduard.Sergeev(a)gmail.com> wrote: > > > Xingzhi Pan wrote: >> >> The first argument to foldr is of type (a -> b -> a), which takes 2 >> arguments. But 'step' here is defined as a function taking 3 >> arguments. What am I missing here? > > You can think of step as a function of two arguments which returns a > function with one argument (although in reality, as any curried function, > 'step' is _one_ argument function anyway): > step :: b -> (a -> c) -> (b -> c) > > e.g. 'step' could have been defined as such: > step x g = \a -> g (f a x) > > to save on lambda 'a' was moved to argument list. Right. But then step is of the type "b -> (a -> c) -> (b -> c)". But as the first argument to foldr, does it agree with (a -> b -> a), which was what I saw when I type ":t foldr" in ghci? > -- > View this message in context: http://old.nabble.com/foldl-in-terms-of-foldr-tp27322307p27324376.html > Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com. > > _______________________________________________ > Haskell-Cafe mailing list > Haskell-Cafe(a)haskell.org > http://www.haskell.org/mailman/listinfo/haskell-cafe > -- Pan, Xingzhi http://www.panxingzhi.net

6 10

Darcs fundraising 2010 - send us to ZuriHac! ($2000 by 15 Feb)
by Eric Y. Kow 26 Jan '10

26 Jan '10

Dear Haskellers, Since 2008, the Darcs Team has made a big push to find hacking time. We've been streamlining our patch approval processes, running Google Summer of Code projects and holding bi-annual hacking sprints. These efforts are starting to pay off, but we're not quite there yet and we still need your help! If you love Darcs and you want to see it succeed, if you want to see it get faster and easier to use, the best thing you can do is to send us your patches (how about attending a sprint?). But if you cannot do this, you can still support the cause. The 4th Darcs Hacking Sprint will be taking place in March as part of the Zurich Haskell Hackathon. Zurich presents a opportunity for us pull a lot of the Darcs team together, to recruit new developers to the project, and also to mingle with the wider Haskell community. We would love to subsidise travel costs to make it easier for developers to attend this sprint and the one after. Do you think you can help? How about sending just $20 our way? Donating to Darcs is now easier than ever, as we are now proud members of the Software Freedom Conservancy. You can either send us a cheque or use Paypal or Google Checkout. Furthermore, donations from US tax-payers will now be tax-deductible. More details below, or at <http://darcs.net/donations.html> Many thanks! Eric PS. If you could send me an email when you make a donation, I can provide a daily update on our fundraising status! Last year, we managed to raise a thousand dollars to send Darcs hackers to the Utrecht and Vienna hackathons. This year, I want to see if we can raise $2000 for Darcs hacking by the 14th of February. Stay tuned at http://darcs.net/donations.html for our progress. Only $1710 to go... :-) Donating by check (USA) ---------------------------------------------------------------------- If you have a US bank account, we prefer donations by check as they will go the longest way. Checks should be made payable to "Software Freedom Conservancy, Inc.", with "Directed donation: Darcs" in the memo. They can be mailed to Software Freedom Conservancy 1995 BROADWAY FL 17 NEW YORK NY 10023-5882 USA Donating by Paypal or Google Checkout ---------------------------------------------------------------------- Please visit http://darcs.net/donations.html for more details. -- Eric Kow <http://www.nltg.brighton.ac.uk/home/Eric.Kow> PGP Key ID: 08AC04F9

1 2

foldl in terms of foldr
by Xingzhi Pan 26 Jan '10

26 Jan '10

Hi, I am reading Real World Haskell and have some questions about the piece of code implementing foldl in terms of foldr: -- file: ch04/Fold.hs myFoldl :: (a -> b -> a) -> a -> [b] -> a myFoldl f z xs = foldr step id xs z where step x g a = g (f a x) The first argument to foldr is of type (a -> b -> a), which takes 2 arguments. But 'step' here is defined as a function taking 3 arguments. What am I missing here? (Partial application? The order of execution?) Btw, is there a way I can observe the type signature of 'step' in this code? Thanks in advance! -- Pan, Xingzhi http://www.panxingzhi.net

5 4

darcs 2.4 beta 3 release
by Reinier Lamers 26 Jan '10

26 Jan '10

Hi all, The darcs team would like to announce the immediate availability of darcs 2.4 beta 3. darcs 2.4 will contain many improvements and bugfixes compared to darcs 2.3.1. Highlights are the faster operation of record, revert and related commands, and the experimental interactive hunk editing. This beta is your chance to test-drive these improvements and make darcs even better. Compared to darcs 2.4 beta 2, the interface of interactive hunk editing has become more user-friendly. The easiest way to install darcs is using the Haskell Platform [1]. If you have installed the Haskell Platform or cabal-install, you can install this beta release by doing: $ cabal update $ cabal install --reinstall darcs-beta Alternatively, you can download the tarball from http://darcs.net/releases/darcs-beta-2.3.98.3.tar.gz , and build it by hand as explained in the README file. Interactive hunk editing ------------------------ To try out interactive hunk editing, press 'e' when you are prompted with a hunk patch by 'darcs record'. You will then be shown an editor screen in which you can edit the state you want to record between the last two ruler lines. You can find more information about the hunk editing feature on http://wiki.darcs.net/HunkEditor . Reporting bugs -------------- If you have an issue with darcs 2.4 beta 3, you can report it via the web on http://bugs.darcs.net/ . You can also report bugs by email to bugs(a)darcs.net. What's New ---------- A list of important changes since 2.3.1 is as follows (please let me know if there's something you miss!): * Use fast index-based diffing everywhere (Petr) * Interactive patch splitting (Ganesh) * An 'optimize --upgrade' option to convert to hashed format in-place (Eric) * Hunk matching (Kamil Dworakowski, tat.wright) * Progress reporting is no longer deceptive (Roman) * A 'remove --recursive' option to remove a directory tree from revision control (Roman) * 'show files' accepts arguments to show a subset of tracked files (Luca) * A '--remote-darcs' flag for pushing to a host where darcs isn't called darcs * Many miscellaneous Windows improvements (Salvatore, Petr and others) * 'darcs send' now mentions the repository name in the email body (Joachim) * Handle files with boring names in the repository correctly (Petr) * Fix parsing of .authorspellings file (Tomáš) * Various sane new command-line option names (Florent) * Remove the '--checkpoint' option (Petr) * Use external libraries for all UTF-8 handling (Eric, Reinier) * Use the Haskell zlib package exclusively for compression (Petr) A list of issues resolved since 2.3.1: * 183: do not sort changes --summary output * 223: add --remote-darcs flag to specify name of remote darcs executable * 291: provide (basic) interactive patch splitting * 540: darcs remove --recursive * 835: 'show files' with arguments * 1122: get --complete should not offer to create a lazy repository * 1216: list Match section in ToC * 1224: refuse to convert a repo that's already in darcs-2 format * 1300: logfile deleted on unsucessful record * 1308: push should warn about unpulled patches before patch-selection * 1336: sane error message on --last "" (empty string to numbers parser) * 1362: mention repo name in mail send body * 1377: getProgname for local darcs instances * 1392: use parsec to parse .authorspelling * 1424: darcs get wrongly reports "using lazy repository" if you ctrl-c old-fashioned get * 1447: different online help for send/apply --cc * 1488: fix crash in whatsnew when invoked in non-tracked directory * 1548: show contents requires at least one argument * 1554: allow opt-out of -threaded (fix ARM builds) * 1563: official thank-you page * 1578: don't put newlines in the Haskeline prompts * 1583: on darcs get, suggest upgrading source repo to hashed * 1584: provide optimize --upgrade command * 1588: add --skip-conflicts option * 1594: define PREPROCHTML in makefile * 1620: make amend leave a log file when it should * 1636: hunk matching * 1643: optimize --upgrade should do optimize * 1652: suggest cabal update before cabal install * 1659: make restrictBoring take recorded state into account * 1677: create correct hashes for empty directories in index * 1681: preserve log on amend failure * 1709: fix short version of progress reporting * 1712: correctly report number of patches to pull * 1720: fix cabal haddock problem * 1731: fix performance regression in check and repair Kind Regards, the darcs release manager, Reinier Lamers [1]: You can download the Haskell platform from http://hackage.haskell.org/platform/

1 0

Job opportunities at Citrix Systems (Cambridge, UK)
by Matthias Görgens 26 Jan '10

26 Jan '10

Hi folks, I have recently started working for Citrix in Cambridge. We are working on the open source software in Ocaml. Admittedly Ocaml is only a second best compared to Haskell ;o) and I hope my post does not count as too off-topic here. But Ocaml is still a decent language. My experience was primarily with Haskell and I only started working with Ocaml in earnest at Citrix. (So I hope that showing a way to transplant Haskell skills to the real world --- via the roundabout route of Ocaml --- justifies my posting.) So --- if you are looking for a job in Cambridge that involves functional programming, free software [1] and free beer [2], please feel free to drop me a line. I have included our original advert to the Ocaml mailing list for those who want more information first. Cheers, Matthias. [1] As in free speech. LGPL [2] As in free beer. We also get other perks, like snacks and a pool table. ---------- Forwarded message ---------- From: Dave Scott <Dave.Scott(a)eu.citrix.com> Date: 2010/1/18 Subject: [Caml-list] [jobs] Citrix Systems (Cambridge, UK) To: "caml-list(a)yquem.inria.fr" <caml-list(a)yquem.inria.fr>, "ocaml-jobs(a)inria.fr" <ocaml-jobs(a)inria.fr> Dear fellow OCaml users, Summary: We (Citrix Systems) are looking for OCaml programmers to join our team in Cambridge, UK. We use OCaml extensively in the "xapi tool-stack": the control-plane used in the Xen Cloud Platform[1] on which the widely used XenServer server virtualisation product[2] is based. XCP aims to provide a complete open-source cloud infrastructure platform with a powerful management stack based on standards-based APIs, support for multi-tenancy, SLA guarantees and detailed metrics for consumption based charging. We are looking to recruit top-class engineers to work on the tool-stack; applicants must have a good knowledge of data structures and algorithms, experience of programming in the context of large systems and general aesthetic good taste when it comes to code and architecture. Our code base is significant and varied: over 130,000 lines of OCaml, solving problems ranging from the low-level (Xen hypercalls) to the high-level (resource pool management), to the compiler-driven (generating language bindings for our Xen datamodel). Our ideal candidate will have: * significant experience of applications programming in high-level languages (such as OCaml :-) * an aptitude for implementing (and reasoning about) complex concurrent, distributed systems * the skills required to contribute to both the architectural design and day-to-day development of a large code-base * strong communication skills and problem solving ability * a determination to deliver great products that perform brilliantly and meet our customers' needs So if you want to tackle interesting and challenging programming problems and contribute to an innovative, fast-growing product that is already used by tens of thousands of customers across the world, please don't hesitate to send me your CV. [1] http://www.xen.org/products/cloudxen.html [2] http://www.citrix.com/English/ps2/products/feature.asp?contentID=1686939 Thanks, -- Dave Scott <dave.scott(a)eu.citrix.com>

1 0

Re: Codensity improvement of free monads
by Janis Voigtländer 26 Jan '10

26 Jan '10

> On Mon, Jan 25, 2010 at 7:37 PM, Luke Palmer <lrpalmer(a)gmail.com> wrote: >> Hello haskell-cafe, >> >> I have just read "Asymptotic Improvement of Computations over Free >> Monads" by Janis Voigtlander, since I have been working with free >> monads a lot recently and don't want to get hit by their quadratic >> performance when I start to care about that. >> >> But after reading the paper, I still don't really understand how >> "improve" actually improves anything. Can anyone provide a good >> explanation for where the work is saved? Edward gives the perfect explanation below. Thanks, Janis. > With a free monad, the structure keeps growing via substitution. This > requires you on each bind to re-traverse the common root of the free monad, > which isn't changing in each successive version. > > Lets say the size of this accumulated detritus increases by one each time. > Then you will be traversing over 1, 2, 3, 4, ... items to get to the values > that you are substituting as you keep binding your free monadic > computation. > The area near the root of your structure keeps getting walked over and > over, > but there isn't any work do to there, the only thing bind can do is > substitution, and all the substitution is down by the leaves. > > On the other hand, when you have CPS transformed it, at each layer you only > have to climb over the 1 item you just added to get to where you apply the > next computation. > > This only gets better when you are dealing with free monads that provide > multiple points for extension: i.e. that grow in a treelike fashion like: > > data Bin a = Bin a a > data Free f a = Return a | Free (f (Free f a)) > data Codensity f a = Codensity (forall r. (a -> f r) -> f r) > > If you build the free monad Free Bin, then you will wind up having to walk > all the way down to the leaves on each bind, well, modulo demand, that is. > But since it is a free monad, the 'body' of the tree will never change. > Just > the data you are substituting at the leaves. Codensity (Free Bin) lets you > only generate the body of the tree once, while your substitutions just get > pushed down to the leaves in one pass. > > An interesting exercise is to work out why Codensity can change the > asymptotics of certain operations over Free Bin, but Density doesn't change > the asymptotics of anything non-trivial over Cofree Bin for the better. > > -Edward Kmett > -- Jun.-Prof. Dr. Janis Voigtländer http://www.iai.uni-bonn.de/~jv/ mailto:jv@iai.uni-bonn.de

1 0

OT: How a Common Lisp user views other programming languages
by Scott Michel 26 Jan '10

26 Jan '10

Off topic, but funny: http://kvardek-du.kerno.org/2010/01/how-common-lisp-programmer-views-users.… The Haskell icon is very apt! -scooter

2 1

ANN: The Monad.Reader Issue 15
by Brent Yorgey 26 Jan '10

26 Jan '10

I am very pleased to announce that Issue 15 of The Monad.Reader is now available for your reading pleasure [1]. Issue 15 consists of the following four articles: * "The hp2any project" by Gergely Patai * "Adventures in Three Monads" by Edward Z. Yang * "The Operational Monad Tutorial" by Heinrich Apfelmus * "Implementing STM in pure Haskell" by Andrew Coppin Feel free to browse the source files. You can check out the entire repository using darcs: darcs get http://code.haskell.org/~byorgey/TMR/Issue15 If you'd like to write something for Issue 16, please get in touch. The deadline will likely be sometime in mid-late April; more details will be forthcoming. -Brent [1] http://themonadreader.files.wordpress.com/2010/01/issue15.pdf

2 1

HList darcs repo missing?
by Jake Wheat 26 Jan '10

26 Jan '10

Hello all, I was looking for the HList darcs repo at: http://darcs.haskell.org/HList/ but it seems to be missing. Has it been moved somewhere else? Thanks, Jake Wheat

4 4

list mail
by 田其雨 26 Jan '10

26 Jan '10

Sorry, for some reasons, I had to change my E-mail, so please don't send the list mail anymore, and I will get it with a new mail. Best wishes. _________________________________________________________________ Windows Live: Make it easier for your friends to see what you’re up to on Facebook. http://www.microsoft.com/middleeast/windows/windowslive/see-it-in-action/so…

1 0

scheduling an alarm
by Brian Denheyer 26 Jan '10

26 Jan '10

I cobbled the following together from bits and pieces I found on mailing lists. Seems straightforward, but the code I've included just sits there. The awaitSignal seems to be causing a problem, in that if I remove it (and don't call forever recursively) then I get my ALARM 5 sec later. Otherwise nothing. The problem is that if I don't include awaitSignal, then, of course, forever is called just as fast as the CPU will go :-) Brian import System.Posix.Signals import System.IO import Control.Concurrent alarm = do putStrLn "ALARM" hFlush stdout return () forever = do putStrLn "foo" scheduleAlarm 5 awaitSignal Nothing yield forever main = do installHandler realTimeAlarm (Catch alarm) Nothing forever

2 1

Codensity improvement of free monads
by Luke Palmer 26 Jan '10

26 Jan '10

Hello haskell-cafe, I have just read "Asymptotic Improvement of Computations over Free Monads" by Janis Voigtlander, since I have been working with free monads a lot recently and don't want to get hit by their quadratic performance when I start to care about that. But after reading the paper, I still don't really understand how "improve" actually improves anything. Can anyone provide a good explanation for where the work is saved? Luke

2 1

ANN: ftdi-0.1
by Roel van Dijk 25 Jan '10

25 Jan '10

Hello, I just released a very early version of my 'ftdi' library [1]. It is a small layer on top of the usb package [2] that enables you to communicate with FTDI [3] devices. Its design is based on the C-library libftdi [4], but it implemented completely in Haskell. Most functionality is untested. I only have access to an FT2232C chip, so testing other chips is not possible at the moment. Furthermore, the only thing I do with that chip is perform bulk reads and the occasional small (1 byte) bulk write. The good news is that it is quite good a bulk reads. It outperformed both libftdi-0.16 and the official driver [5] by about 40% for my use case. Although to be fair, both C libraries used the old libusb-0.1 library. This library is ultimately linked with libusb-1.0.* which is a complete rewrite. To play with the library either cabal install ftdi or darcs get http://code.haskell.org/~roelvandijk/code/ftdi Regards, Roel van Dijk 1 - http://hackage.haskell.org/package/ftdi 2 - http://ftdichip.com/ 3 - http://hackage.haskell.org/package/usb 4 - http://www.intra2net.com/en/developer/libftdi/ 5 - http://www.ftdichip.com/Drivers/D2XX.htm

1 0

Adopting hpodder?
by John Goerzen 25 Jan '10

25 Jan '10

Hi folks, I wrote hpodder a little while back. I no longer listen to podcasts at all (blame my Kindle for that). Is anyone here interested in adopting it and giving it some needed care? Thanks, -- John

1 0

Failing to install hxt
by haskell＠kudling.de 25 Jan '10

25 Jan '10

Hi guys, i have some troubles to install hxt. My first naive attempt "cabal install hxt" fails with "cabal: cannot configure hxt-8.5.0. It requires base >=4.2 && <5". I guess that's because i use ghc 6.10.4, since "base-4.2.0.0" depends on "integer-simple" which seems to only come with ghc 6.12. I then tried to install hxt-8.3.2, which seems to be the last version supporting ghc 6.10.4. If fails with: " src/Text/XML/HXT/Parser/TagSoup.hs:292:8: Not in scope: `optLookupEntity' cabal: Error: some packages failed to install: hxt-8.3.2 failed during the building phase. The exception was: exit: ExitFailure 1 " I have no idea, what that means. What am i doing wrong? Thanks, Lenny

5 13

Naive booleans and numbers - type-checking fails
by Dušan Kolář 25 Jan '10

25 Jan '10

Dear cafe, I'm trying to prepare a naive definition of booleans, numbers and some helper functions such a way, so that definition in lambda-calculus can be used in a straightforward way in Haskell. I was caught in trouble quite soon during change of lambda-expressions to Haskell - defining prefn as a helper for prev. When using Haskell-ish if-then-else then there is no problem (the code commented out), while using defined if-then-else (mif), the type-checking fails, but just for this case! Other simple tests are OK for the mif. Do I need some extra option for type-checker, or is it a principal failure (infinite type is reported) - I'm running ghci 6.10.4. > mtrue x y = x > mfalse x y = y > m0 f n = n > m1 f n = f n > m2 f n = f (f n) > msucc x g m = x g (g m) > iszero m = m (\_ -> mfalse) mtrue > mif c t f = c t f > mpair f s = \e -> e f s > mfst p = p mtrue > msnd p = p mfalse > -- mprefn f p = if mex True False then mth else mel > mprefn f p = mif mex mth mel > where > mex = mfst p > mth = mpair mfalse (msnd p) > mel = mpair mfalse (f (msnd p)) Please, change of definitions is not a solution, I'm trying to follow available resources, so using something else is not an option. :-( Thanks for any help Dusan

4 3

Re: [Haskell-cafe] Type level splices and instance deriving
by John Lato 25 Jan '10

25 Jan '10

Hello, > From: Khudyakov Alexey <alexey.skladnoy(a)gmail.com> > Hello > > However I run into problem with them. It's possible to create instance for > type class which doesn't have superclass. If it does have one compiler > complains that it could not deduce context. All my attempts to provide context > fail. What have you tried? > > Is that accidental limitation or because of Some Good Reason? Or just due to > lack of understanding on my side? > > Below is simplest example. >> {-# LANGUAGE FlexibleInstances #-} >> {-# LANGUAGE TemplateHaskell #-} >> import Language.Haskell.TH >> >> -- OK but require FlexibleInstances >> makeEq :: Name -> Q [Dec] >> makeEq name = >> [d| instance Eq $(conT name) where >> (==) = undefined >> |] >> >> -- Could not deduce Eq context >> makeOrd :: Name -> Q [Dec] >> makeOrd name = >> [d| instance Ord $(conT name) where >> compare = undefined >> |] > > And GHC output: > > test.hs:14:17: > Could not deduce (Eq t) from the context () > arising from the superclasses of an instance declaration > at test.hs:14:17-32 > Possible fix: add (Eq t) to the context of the instance declaration > In the instance declaration for `Ord t_aS5' > In the Template Haskell quotation > [d| > instance Ord $(conT name) where > { compare = undefined } |] > In the expression: > [d| > instance Ord $(conT name) where > { compare = undefined } |] I think this will work if you add the Eq constraint in the instance definition: makeOrd :: Name -> Q [Dec] makeOrd name = [d| instance Eq $(conT name) => Ord $(conT name) where compare = undefined |] The problem is that, when the (Ord t) instance is compiled, the compiler needs to know where to find the (Eq t) dictionary because Eq is a superclass of Ord. If 't' is a concrete type, e.g. Int, the compiler will look for the Eq instance and use it directly. If 't' is a type variable, as in your code, then the compiler needs to have the Eq constraint so that it can find the dictionary. In your case you provide an Eq instance for any type 't'. In theory I think the compiler could examine the instance, see that it's valid for any 't', and use that in your definition, but I don't think it's a good idea. This would be an extra branch that would only apply to a class that is instantiated for all types. Such an instance couldn't do anything useful; as an example your Eq instance is undefined. That's why GHC doesn't look for instances like the one you provided when determining type class constraints; they can't provide the necessary functionality. I suspect this feature could be added and then your code would compile, but you couldn't use it for anything so there's no point to doing so. Cheers, John

1 0

Non-termination due to context
by Emil Axelsson 25 Jan '10

25 Jan '10

Hello all! Consider the following program: > {-# LANGUAGE FlexibleInstances, OverlappingInstances, UndecidableInstances #-} > > class B a => A a > > instance A Int > > class Eq a => B a > > instance (A a, Eq a) => B a > > eq :: B a => a -> a -> Bool > eq = (==) > > test = 1 `eq` (2::Int) (This is a condensed version of a much larger program that I've been debugging.) It compiles just fine, but `test` doesn't terminate (GHCi 6.10.4). If I change the context `B a` to `Eq a` for the function `eq`, it terminates. Although I don't know all the details of the class system, it seems unintuitive that I can make a program non-terminating just by changing the context of a function (regardless of UndecidableInstances etc.). Is this a bug or a feature? / Emil

5 6

ANN: afv-0.1.0
by Tom Hawkins 25 Jan '10

25 Jan '10

AFV is an infinite state model checker to verify assertions in embedded C programs. New in this release: - Starts analysis from 'main' entry point. Automatically identifies the main loop (for (;;), while (1)). - Better counter example generation. - Enforces stateless expressions. Unfortunately, the list of unsupported C is still long. No... - arrays - pointers - structs - unions - typedefs - type casts - static top-level declarations - non void functions - switch statements - arbitrary loop statements http://hackage.haskell.org/package/afv

1 0

PROPOSAL: Web application interface
by Michael Snoyman 24 Jan '10

24 Jan '10

Following up on the previous thread, I've started a github project for some ideas of a web application interface. It's borrowing from both Hyena and Hack, with a few of its own ideas. The project is available at http://github.com/snoyberg/wai, and the Network.Wai module is available at http://github.com/snoyberg/wai/blob/master/Network/Wai.hs. The repository also includes a port of hack-handler-simpleserver, and an incredibly simple webapp to demonstrate usage. I intend to make the demonstration slightly more sophisticated. Finally, the repository is now yet cabalized. I consider this currently to be a straw-man proposal, intended to highlight the issues of contention that may arise. It would be wonderful is we could get the major players in the Haskell web space to get behind a single WAI. The entire Network.Wai module right now weighs in at only 74 lines, so I do not consider this to be a heavy-weight proposal. Here as some design notes: - Most important point: RequestBody and ResponseBody. I will explain below. - I've renamed "Env" in Hack and "Environment" in Hyena to "Request." This seems more consistent with other technologies out there. However, I have no feelings on this subject at all, and can easily bend to public demand. - I've stuck with UrlScheme from Hack, while Hyena called it protocol. Similar, the RequestMethod constructors are ALLCAPS like Hack, unlike Hyena's Uppercase. Once again, no strong feelings. - I've sided with Hyena as far as making all representations in ByteString. Current exception is remoteHost, which is a Hack-only variable in any event. - Instead of representing the response as a tuple ala Hyena, created a data type like Hack. - The only dependency for this module is bytestring. It might be tempting to represent RequestBody and ResponseBody with a ReaderT IO monad, but this would introduce a dependency on either mtl or transformers, which I would consider a Very Bad Idea. The main complaint against Hack is its lack of an enumerator interface for the request and response body. However, this simply words the complaint incorrectly; I don't think anyone is married to the need of an enumerator. Rather, we want to be able to efficiently handle arbitrarily lengthed content in constant space, without necesarily resorting to unsafeInterleaveIO (ie, lazy I/O). There are a number of issues with left-fold enumerators IMO. It is basically promoting an inversion of control. This may be often times valuable. However, to make this the *only* interface precludes other use cases. The most basic one I ran into was wanting to interleave with read processes. I do not mean to say that it's impossible to interleave reads in such a manner, but I think it's more natural in the approach advocated by wai. I consider RequestBody and ResponseBody to be mirroring the CGI protocol. Essentially, each handler (CGI, simpleserver, FastCGI, happstack server, etc) will define data types which instanciate RequestBodyClass and ResponseBodyClass. RequestBodyClass provides a single method, receiveByteString, to extract a chunk of data from the request body. ResponseBodyClass provides (currently) three methods, for sending strict bytestrings, lazy bytestrings, and files. While default implementations are provided for the last two based on the first, implementations can provide more efficient versions of them if desired. For example, sendFile might be replaced by a system call to sendfile. Let me know your thoughts. I'm purposely leaving out many of my reasons for the decisions I've made for brevity, since this e-mail is long enough as is. I'm happy to answer any questions as to why I went in a certain direction. It's also possible that I simply overlooked a detail. Michael

4 9

Haskell Weekly News: Issue 147 - January 24, 2010
by jfredett＠gmail.com 24 Jan '10

24 Jan '10

--------------------------------------------------------------------------- Haskell Weekly News http://sequence.complete.org/hwn/20100124 Issue 147 - January 24, 2010 --------------------------------------------------------------------------- Welcome to issue 147 of HWN, a newsletter covering developments in the [1]Haskell community. This week on the HWN, we have a new blogger in town, Bartek, over at 'The Power of Types Compels You', wrote an excellent article which is highlighted in the blogs section. Many new packages this week as well, and some interesting discussions about performance and existential types. So, Haskellers, till next week, your Haskell Weekly News! Announcements ForSyDe DSL v3.1. Seyed Hosein Attarzadeh Niaki [2]announced a new version of the ForSyDe DSL. This version provides more freedom in the declaration of process functions, as well as compatibility with base 4. parameterized-data library v0.1.4. Seyed Hosein Attarzadeh Niaki [3]announced a new version of the parameterized-data library. The parameterized-data library provides fixed-sized vectors, this update is provide minor compatibility fixes. Updates and a new member of the Monadic Regions family. Bas van Dijk [4]announced several new updates and a new member of the Monadic Regions family. bindings-DSL 1.0.4 (Category: FFI). Mauricio CA [5]announced a new version of his package bindings-DSL. Haskell XML Toolbox Version 8.5.0. Uwe Schmidt [6]announced a new version of the Haskell XML Toolbox (HXT). The main change in this version is the separation of the XPath and XSLT modules. Cardinality-0.1. Andrey Sisoyev [7]announced the release of his package Cardinality, a package for transforming between container types safely. afv-0.0.3. Tom Hawkins [8]announced a new release of AFV, an infinite state model checker for simple, iterative C programs. CfP: PPDP 2010. Temur Kutsia [9]announced a call for papers for the PPDP 2010 conference in July Mini-announce: A few package updates. Andrew Coppin [10]announced several new releases and updates to AC-EasyRaster-GTK, AC-Vector, and other packages. amqp-0.1. Holger Reinhardt [11]announced the release of his AMQP library. It currently only works with RabbitMQ and supports most of the 0-8 spec. An introduction to AMQP can be found [12]here. nntp 0.0.3. Maciej Piechotka [13]1f4e gmane.comp.lang.haskell.cafe/69228 announced a bugfix release to nntp. haskell-src-exts 1.7.0. Niklas Broberg [14]announced a new version of haskell-src-exts, featuring many new changes to the API. Discussion Existential Types (I guess). Ozgur Akgun [15]asked about a problem he encountered when trying to use Existential types. Is Haskell capable of matching C in string processing performance? John Millikin [16]asked a question about Haskell Performance... Quick! Someone get Don Stewart! Blog noise [17]Haskell news from the [18]blogosphere. Blog posts from people new to the Haskell community are marked with >>>, be sure to welcome them! * ++Bartek Paczesiowa++: [19]Pure, extensible exceptions and self-returning functions. Bartek, over at 'The Power of Types Compels You', wrote a fantastic introduction to his forthcoming exceptions package. It's funny, and a great read about using type families (particularly, equality constraints) to create a pure exception handling mechanic. Definitely worth a visit. * Neil Mitchell: [20]Optimising HLint. * Kevin Reid (kpreid): [21]Darcs repositories back. * Darcs: [22]darcs weekly news #52. * Bryan O'Sullivan: [23]New GHC I/O manager, first sets of benchmark numbers. * Galois, Inc: [24]POPL 2010, Day 1. * Galois, Inc: [25]Tech Talk: A Scalable I/O Manager for GHC. * Galois, Inc: [26]PADL/PEPM 2010, Day 2. * Neil Brown: [27]The Process Composition Monad. * Galois, Inc: [28]PADL/PEPM 2010, Day 1. * Thomas M. DuBuisson: [29]GHC on ARM. * David Sankel: [30]Semantic Editor Combinators - one of my favorite blog posts. * David Sankel: [31]Best Haskell Papers of 2009. * Gtk2HS: [32]Compiling with ghc 6.12. * Arch Haskell News: [33]wxHaskell packaged for Arch. * Darcs: [34]darcs weekly news #51. * Don Stewart (dons): [35]Playing with the new Haskell epoll event library. * Dan Piponi (sigfpe): [36]Target Enumeration with the Euler Characteristic. Parts 1 & 2. Quotes of the Week * tensorpudding: the Plot monad allow you to keep the story pure by containing all the glaring time travel silliness * sproingie: quickcheck myLanguage.hs --> "Web browser created after 285,731 tests" * RossPaterson: I'm afraid you voided the warranty when you used UndecidableInstances. * aavogt: strong static typing is not a substitute for sleep * tensorpudding: fixity goes up to 11 * temoto: Backwards written in as Forth? * Paczesiowa: oh. can't argue with Cale :) * Axman6: and smilies make code run faster About the Haskell Weekly News New editions are posted to [37]the Haskell mailing list as well as to [38]the Haskell Sequence and [39]Planet Haskell. [40]RSS is also available, and headlines appear on [41]haskell.org. To help create new editions of this newsletter, please see the information on [42]how to contribute. Send stories to jfredett . at . gmail . dot . com. The darcs repository is available at darcs get [43]http://patch-tag.com/r/jfredett/HWN2/pullrepo HWN2 . References 1. http://haskell.org/ 2. http://article.gmane.org/gmane.comp.lang.haskell.cafe/69474 3. http://article.gmane.org/gmane.comp.lang.haskell.cafe/69473 4. http://article.gmane.org/gmane.comp.lang.haskell.cafe/69470 5. http://article.gmane.org/gmane.comp.lang.haskell.cafe/69465 6. http://article.gmane.org/gmane.comp.lang.haskell.cafe/69434 7. http://article.gmane.org/gmane.comp.lang.haskell.cafe/69366 8. http://article.gmane.org/gmane.comp.lang.haskell.cafe/69339 9. http://article.gmane.org/gmane.comp.lang.haskell.general/17753 10. http://article.gmane.org/gmane.comp.lang.haskell.cafe/69322 11. http://article.gmane.org/gmane.comp.lang.haskell.cafe/69250 12. http://blogs.digitar.com/jjww/2009/01/rabbits-and-warrens/ 13. http://article.gmane.org/ 14. http://article.gmane.org/gmane.comp.lang.haskell.cafe/69197 15. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/69428 16. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/69416 17. http://planet.haskell.org/ 18. http://haskell.org/haskellwiki/Blog_articles 19. http://paczesiowa.blogspot.com/2010/01/pure-extensible-exceptions-and-self.… 20. http://neilmitchell.blogspot.com/2010/01/optimising-hlint.html 21. http://kpreid.livejournal.com/21077.html 22. http://blog.darcs.net/2010/01/darcs-weekly-news-52.html 23. http://www.serpentine.com/blog/2010/01/22/new-ghc-io-manager-first-benchmar… 24. http://www.galois.com/blog/2010/01/21/popl-2010-day-1/ 25. http://www.galois.com/blog/2010/01/20/tech-talk-a-scalable-io-manager-for-g… 26. http://www.galois.com/blog/2010/01/20/padlpepm-2010-day-2/ 27. http://chplib.wordpress.com/2010/01/19/the-process-composition-monad/ 28. http://www.galois.com/blog/2010/01/19/padlpepm-2010-day-1/ 29. http://tommd.wordpress.com/2010/01/19/ghc-on-arm/ 30. http://netsuperbrain.com/blog/posts/semantic-editor-combiners/ 31. http://netsuperbrain.com/blog/posts/best-haskell-papers-of-2009/ 32. http://haskell.org/gtk2hs/archives/2010/01/18/compiling-with-ghc-612/ 33. http://archhaskell.wordpress.com/2010/01/18/wxhaskell-packaged-for-arch/ 34. http://blog.darcs.net/2010/01/darcs-weekly-news-51.html 35. http://donsbot.wordpress.com/2010/01/17/playing-with-the-new-haskell-epoll-… 36. http://blog.sigfpe.com/2010/01/target-enumeration-with-euler.html 37. http://www.haskell.org/mailman/listinfo/haskell 38. http://sequence.complete.org/ 39. http://planet.haskell.org/ 40. http://sequence.complete.org/node/feed 41. http://haskell.org/ 42. http://haskell.org/haskellwiki/HWN 43. http://patch-tag.com/r/jfredett/HWN2/pullrepo%20HWN2

1 0

Installing "encoding" package from Hackage
by Chris Wagner 24 Jan '10

24 Jan '10

Hi all, I'm trying to install the "encoding" package using cabal, but I'm getting an error about a missing "HaXml" module... --------------------------------------------------------------------- chris@desktop:~$ cabal install encoding Resolving dependencies... /tmp/encoding-0.6.26119/encoding-0.6.2/Data/Encoding/Preprocessor/XMLMappingBuilder.hs:16:7: Could not find module `Text.XML.HaXml.XmlContent': Use -v to see a list of the files searched for. cabal: Error: some packages failed to install: encoding-0.6.2 failed during the configure step. The exception was: exit: ExitFailure 1 --------------------------------------------------------------------- (I'm actually interested in using the "idiii" package, of which "encoding" is a dependency.) Any help/insight is much appreciated! Thanks! Chris W.

2 1

Re: Space Efficiency When Sorting a List of Many Lists
by Peter Green 24 Jan '10

24 Jan '10

3. Re: Space Efficiency When Sorting a List of Many Lists (Heinrich Apfelmus) >>> Another interesting problem is starting from a file of single >>> wagers and >> trying to compress them (i.e. inverse of 'explosion'. I believe this >> problem is analogous to Set Cover and therefore NP-Complete. >> Heuristics >> are the order of the day here. > > Interesting indeed! What kind of heuristics do you employ there? It > seems quite difficult to me, somehow. One approach I *did* consider early on was to define a kind of 'Pseudo Hamming Distance': i.e. define PHD such that any two single combination wagers differing in one leg's entrant number have PSD = 1 e.g. 1, 2, 3, 4 --A 1, 1, 3, 4 --B 5, 1, 3, 4 --C A and B have PHD = 1 B and C have PHD = 1 A and C have PHD = 2, but PSD > 1 is not of interest here. Construct an adjacency list for all PHD = 1 -> an undirected graph where nodes are combinations and edges link combinations with PSD = 1- > one starts to see various interesting hypercube type structures in the graphviz plots which are analogous to grouped wagers. Easy enough to see this by working backwards from (say) 1+2+2/4+5+6/7+8+9/10+11+12, exploding it by CP, constructing the PHD=1 adjacency list, turning this into a DOT file and firing up graphviz. These graphviz pictures look very lovely to the human eye. Unfortunately, computational extraction of such features in graph spaghetti land, let-alone the issue of extracting maximal features seems to be infeasible before the arrival of Quantum Computers, Maxwell's Demons in every car engine, and Unicorns in the back garden. I also considered a set cover approach, but got a little dispirited when I sat down to compute the number of potential covering sets - CP of power sets of race field sizes! Essentially wager generation in multi-leg races works like this: (1) We have estimated win probabilities for the entrants in each of the race legs. Assume that I have a black box which produces these estimated probabilities. (2) We generate the CP of the entrant numbers in the race fields: Leg 1 Entants x Leg 2 Entrants x.. For each combination generated by the CP, we apply some formula to the probabilities associated with combination entrant numbers to arrive at a decision about whether or not that particular combination represents a viable wager. (3) If a combination does represent a viable wager, we then need to ensure that we have sized its wager amount in a correct and optimal manner. This gives output looking like: 1, 1, 2, 3 - $1.10 1, 1, 2, 4 - $1.20 . . 15, 15, 15, 12, - $0.35 Unfortunately these are hard to compress well - and that is ignoring issues of what to do about differing wager sizes when we can match two combinations for merging together. So one possible hack is to insert step 1(a) where we k-means cluster the win probabilities in each leg, such that we end up with multiple entrant numbers per cluster mean (which is a probability). We then proceed through steps 2 and 3, but instead of generating single combinations, we are now directly producing wagers which look like: 1+6+12/3+4+7/13+14 - $1.20 In this case, k-means clustering will have clustered 1,6,12 together with cluster mean/probability of (say) 0.1, etc. This a bit of a hack, as clearly it will result in the wagers not covering exactly the same single combinations as if we had generated pure single wagers without k-means clustering. Hence my interest in doing set comparisons! Another brute force greedy approach is to start with single combinations and do something like this: def merge_combinations(combinations): """Greedily merge all combinations on one field, then next field, etc. Parameter combinations is structured so: [[[1], [2], [3], [4], [5], [6], [7]], [[1], [2], [3], [4], [5], [6], [8]]] The input combinations are modified *in place*. After calling this method, combinations would become: [[[1], [2], [3], [4], [5], [6], [7,8]]] """ def rotate_left(combi): combi.append(combi.pop(0)) for do_it_seven_times in xrange(7): combinations.sort() k = 1 while k < len(combinations): if combinations[k-1][:6] == combinations[k][:6]: combinations[k-1][6].extend(combinations.pop(k)[6]) combinations[k-1][6].sort() else: k += 1 map(rotate_left, combinations) def merge_demo(): from pprint import pprint combis = [[[1], [1], [1], [2], [2], [12], [1]], [[1], [1], [1], [2], [6], [3], [1]], [[1], [1], [1], [2], [6], [6], [1]], [[1], [1], [1], [2], [6], [7], [1]], [[1], [1], [1], [2], [6], [7], [2]], [[1], [1], [1], [2], [6], [12], [1]], [[1], [1], [1], [2], [8], [7], [1]], [[1], [1], [1], [2], [8], [12], [1]], [[1], [1], [1], [2], [9], [3], [1]], [[1], [1], [1], [2], [9], [7], [1]], [[1], [1], [1], [2], [9], [12], [1]], [[1], [1], [1], [3], [2], [7], [1]], [[1], [1], [1], [3], [2], [12], [1]]] pprint(sorted(combis)) # 13 print len(combis) merge_combinations(combis) pprint(sorted(combis)) # -> # [[[1], [1], [1], [2], [6], [6], [1]], # [[1], [1], [1], [2], [6], [7], [1, 2]], # [[1], [1], [1], [2], [6, 8, 9], [12], [1]], # [[1], [1], [1], [2], [6, 9], [3], [1]], # [[1], [1], [1], [2], [8, 9], [7], [1]], # [[1], [1], [1], [2, 3], [2], [12], [1]], # [[1], [1], [1], [3], [2], [7], [1]]] # 7 print len(combis) print print combis = [[[13], [6, 10, 13], [3, 6], [4], [2, 8], [7], [1]], [[13], [6, 10, 13], [3, 6], [4], [9], [7], [1]], [[13], [6, 10, 13], [3, 6], [4], [6], [12], [1]], [[13], [6, 10, 13], [3, 6], [4], [6], [7], [1]], [[13], [6, 10, 13], [5, 10], [10], [2, 8], [7], [1]], [[13], [6, 10, 13], [5, 10], [10], [9], [7], [1]], [[2, 9], [11], [7, 13], [6, 12], [9], [12], [3]], [[2, 9], [11], [7, 13], [6, 12], [9], [12], [2]], [[2, 9], [11], [7, 13], [6, 12], [9], [12], [1]], [[2, 9], [11], [7, 13], [6, 12], [9], [7], [5, 8, 14]], [[2, 9], [11], [7, 13], [6, 12], [9], [7], [7]], [[2, 9], [11], [7, 13], [6, 12], [9], [7], [4]]] pprint(combis) # 12 print len(combis) merge_combinations(combis) # -> # [[[13], [6, 10, 13], [3, 6], [4], [2, 6, 8, 9], [7], [1]], # [[13], [6, 10, 13], [3, 6], [4], [6], [12], [1]], # [[13], [6, 10, 13], [5, 10], [10], [2, 8, 9], [7], [1]], # [[2, 9], [11], [7, 13], [6, 12], [9], [12], [1, 2, 3]], # [[2, 9], [11], [7, 13], [6, 12], [9], [7], [4, 5, 7, 8, 14]]] pprint(combis) # 5 print len(combis) I wrote a version of this in PLT Scheme also a while back. IIRC, I had to flip everything around and perform actions at the head of lists instead of at the tail of Python lists (more like arrays really). Will have to try my hand at a Haskell version later and see how it goes - although I have the suspicion that feeding such a routine 1M single combinations is going to run into all sorts of space issues. Unfortunately this approach never seems to result in more than a 10 x compression factor. Can do much better with the k-means approach - at the cost of missing some theoretically desirable combinations and betting on some spurious combinations. As always, would be very interested to hear if you have any ideas/ insights on first seeing this kind of problem. I'm almost certainly too fixed in my thinking due to long association with the wagering problem domain and probably cannot see the wood for the trees! > What I mean is that there are two ways to represent sets of wagers: > > * A list of single combinations -> [Row a] > * A list of compressed wagers -> [Row [a]] > > To perform set operations, you first convert the latter into the > former. > But the idea is that there might be a way of performing set operations > without doing any conversion. > > Namely, the compressed wagers are cartesian products which behave well > under intersection: the intersection of two compressed wagers is > again a > compressed wager > > intersectC :: Eq a => Row [a] -> Row [a] -> Row [a] > intersectC [] [] = [] > intersectC (xs:xrow) (ys:yrow) = > intersect xs ys : intersectC xrow yrow > > In formulas: > > intersect (cartesian x) (cartesian y) > = cartesian (intersectC x y) > > for compressed wagers x, y :: Row [a] with > > intersect = list intersection, for instance Data.List.intersect > cartesian = converts a compressed wager to a list of > single combinations > > > This allows you to intersect two lists of compressed wagers directly > > intersect' :: [Row [a]] -> [Row [a]] -> [Row [a]] > intersect' xs ys = filter (not . null) $ > [intersectC x y | x<-xs, y<-ys] > Unfortunately, the result will have quadratic size; but if you have > significantly less compressed wagers than single combinations, this > may > be worth a try. Not to mention that you might be able to compress the > result again. Thank you for the explanation. This approach is going to be quite useful for most reasonable sized input sets. I'm thinking of running this as part of a suite of web services in something like a JEOS instance with relatively limited RAM - so there are advantages to avoiding exploding to singles and using 100+MB with a Map. import Data.List (intersect) import Data.List.Split (splitOn) import System.Environment (getArgs) type Row a = [a] type CombisGroup = Row [Int] -- e.g. ["8+12,11,7+13,10","1+2+3,1+9,3+6,4"] --> -- [[[8,12],[11],[7,13],[10]],[[1,2,3],[1,9],[3,6],[4]]] fromCSV :: [String] -> [CombisGroup] fromCSV = map parseOneLine where parseOneLine = map parseGroup . splitOn "," where parseGroup = map read . splitOn "+" parseInput :: String -> [CombisGroup] parseInput = fromCSV . lines singleCombisCount :: [CombisGroup] -> Int singleCombisCount = sum . map cpCardinality -- e.g. CP xss yss having zero common elements: -- intersectC [[1,2],[1,3,4],[5]] [[1,2],[4,5,6],[3]] --> [[1,2],[4],[]] -- e.g. CP xss yss having two common elements [[1,4,5],[2,4,5]] : -- intersectC [[1,2],[3,4],[5]] [[1,2],[4,5,6],[3,5]] --> [[1,2],[4], [5]] intersectC :: Eq a => Row [a] -> Row [a] -> Row [a] intersectC [] [] = [] intersectC (xs:xrow) (ys:yrow) = intersect xs ys : intersectC xrow yrow -- e.g. no common elements: -- intersect' [[[1,2],[1,3,4],[5]],[[1,2],[7,11],[5]]] -- [[[1,2],[4,5,6],[3]],[[11],[7,11],[3,5]]] --> [] -- e.g. two common elements: -- intersect' [[[1,2],[1,3,9],[5]],[[1,2],[3,4],[5]]] -- [[[1,2],[4,5,6],[3,5]],[[11],[4,5,6],[3,5]]] --> -- [[[1,2],[4],[5]]] --> CP --> [[1,4,5], [2,4,5]] intersect' :: Eq a => [Row [a]] -> [Row [a]] -> [Row [a]] intersect' xs ys = filter (notElem []) [intersectC x y | x <- xs, y <- ys] -- e.g. [[8,12],[11],[7,13],[10]] --> 4 cpCardinality :: Row [a] -> Int cpCardinality = product . map length main = do [arg1, arg2] <- getArgs compressedA <- return . parseInput =<< readFile arg1 compressedB <- return . parseInput =<< readFile arg2 putStrLn $ "A source: " ++ arg1 putStrLn $ "B source: " ++ arg2 putStrLn $ "|A Intersect B|: " ++ show (singleCombisCount $ intersect' compressedA compressedB)

2 1

SYB <<looping>> very, very mysteriously
by Jeremy Shaw 24 Jan '10

24 Jan '10

I have the following program which loops under GHC 6.10.4: http://www.hpaste.org/fastcgi/hpaste.fcgi/view?id=13561#a13561 {-# LANGUAGE DeriveDataTypeable, FlexibleInstances, MultiParamTypeClasses, UndecidableInstances #-} module Main where import qualified Data.Data as Data import Data.Typeable (Typeable) import Happstack.Data.Default import Data.Generics.SYB.WithClass.Basics import Data.Generics.SYB.WithClass.Instances () data Proposition = Proposition Expression deriving (Show, Data.Data, Typeable) data Expression = Conjunction (Maybe Expression) deriving (Show, Data.Data, Typeable) -- instance (Sat (ctx [Expression]), Sat (ctx Expression), Sat (ctx Proposition)) => Data ctx Proposition where instance Data DefaultD Proposition where gunfold _ k z c = case constrIndex c of 1 -> k (z Proposition) instance Default Proposition constrExpr :: Constr constrExpr = mkConstr dataTypeExpr "Conjuction" [] Prefix dataTypeExpr :: DataType dataTypeExpr = mkDataType "Expression" [constrExpr] instance ( Data ctx [Expression] , Sat (ctx Expression) , Sat (ctx (Maybe Expression))) => Data ctx Expression where {- instance Data DefaultD Expression where -} gunfold _ k z c = case constrIndex c of 1 -> k (z Conjunction) dataTypeOf _ _ = dataTypeExpr instance Default Expression e :: Expression e = defaultValueD dict main = print e I wish to explain the *many* ways in which it is mysterious. If you load the program into GHCi and evaluate 'e' it will hang. If you compile the program and run it, it will output <<loop>>. This behavior seems annoying, but not very weird. But, here is where it gets fun: 1. if you load the program into GHCi and eval 'e' it will hang. But, if you load the program and type, '(defaultValueD dict) :: Expression' at the prompt, it works fine! 2. if you remove the (Data DefaultD Proposition) instance, it works fine. (Even though Expression does not refer to Proposition in any way) 3. if you simply change the definition of 'gunfold' in the 'Data ctx Proposition' instance to, error "foo". The application works fine. That's right, if you change the body of a function that isn't even being called, evaluating 'e' starts working. (Even though Expression does not refer to Proposition in any way. And even though that gunfold instance is never actually called). 4. if you change the constraint on, Data ctx Expression, from (Data ctx [Expression]) to (Data ctx Expression) it works fine. (Or remove it all together). 5. if you change 'instance (Data DefaultD Proposition) where' to the line above it which is commented out, it works fine. 6. if you change the type of Proposition to, data Proposition = Proposition (Expression, Expression), then it works fine. So far I have only tested this in GHC 6.10.4. Any idea what is going on here? I can't imagine how changing the body of functions that aren't being called would fix things... - jeremy

3 5

ANN: GlomeTrace raytracing library
by Jim Snow 24 Jan '10

24 Jan '10

I have uploaded to hackage a new version of my ray-tracer, Glome. In previous releases, Glome was a monolithic application. In this release, the core algorithms have been abstracted out into a pair of libraries, GlomeVec and GlomeTrace. GlomeVec is a vector library. It's not necessarily any better than any of the other vector libraries on Hackage, but it's fairly simple and does what I need it to do. Also included with GlomeVec is a solid texture library, which has a perlin noise function that may be useful for other projects. GlomeTrace contains constructors for building scenes out of basic primitives like spheres, cones, triangles, and grouping them, moving them around, and performing boolean operations on them, and functions to test rays for intersection against these objects. It also has some higher-level operations, like tracing a ray corresponding to a certain x/y coordinate within the camera's field of view, and returning the color by recursively tracing rays against the scene. (Recursion comes into play if the object is reflective.) GlomeTrace uses a typeclass called Solid for geometric primitives, and the composite primitives that are built up from other primitives use an existential type SolidItem that is just a container for all types that are instances of Solid, so it should be quite easy for an application to add new primitives of its own and have them integrate nicely with the ones that are already defined, just by making them instances of Solid. One very useful composite primitive is a BIH (Bounding Interval Hierarchy), which is a type of automatically-constructed tree of bounding volumes used to speed up the process of tracing a ray against many objects. Constructing a bih is easy: just pass a list of SolidItems to the bih constructor, and it returns a "Bih" object that is itself a SolidItem, so you can trace rays against it as if it were a single primitive. GlomeTrace and GlomeVec now have (partial) haddock documentation so it should be easier to understand the code than it was previously. There is also a (somewhat out of date) tutorial linked to from the Haskell wiki: http://www.haskell.org/haskellwiki/Glome Glome the application requires OpenGL for display, but GlomeVec and GlomeTrace do not. -jim Packages: http://hackage.haskell.org/package/GlomeVec http://hackage.haskell.org/package/GlomeTrace http://hackage.haskell.org/package/glome-hs

1 0

Web application interface
by Michael Snoyman 24 Jan '10

24 Jan '10

Hi, I recently read (again) the wiki page on a web application interface[1] for Haskell. It seems like this basically works out to Hack[2], but using an enumerator instead of lazy bytestring in the response type. Is anyone working on implementing this? If not, I would like to create the package, though I wouldn't mind some community input on some design decisions: * Hack has been fairly well-tested in the past year and I think it provides the features that people want. Therefore, I would want to model the Environment variable for WAI from Hack. I *could* just import Hack in WAI and use the exact same Environment data type. Thoughts? * If using a different data type for Environment, should I replace the String parts with ByteStrings? On the one hand, ByteStrings are the "correct" data type since the HTTP protocol does not specify a character encoding; on the other hand, Strings are easier to deal with. * It's simple to write a function to convert between a lazy bytestring and an enumerator, meaning it would be very easy to write conversion functions between Hack and WAI applications. This would make it simpler for people to use either backend. If someone else is already working on WAI, please let me know, I don't want to have duplicate implementations. The idea here is to consolidate, not split the community. I have a few Hack handlers (simpleserver, cgi, fastcgi) that I would happily convert to WAI handlers as well. Michael [1] http://www.haskell.org/haskellwiki/WebApplicationInterface [2] http://hackage.haskell.org/cgi-bin/hackage-scripts/package/hack

5 16

Is Haskell capable of matching C in string processing performance?
by John Millikin 23 Jan '10

23 Jan '10

Recently I've been working on a library for generating large JSON[1] documents quickly. Originally I started writing it in Haskell, but quickly encountered performance problems. After exhausting my (meager) supply of optimization ideas, I rewrote some of it in C, with dramatic results. Namely, the C solution is * 7.5 times faster than the fastest Haskell I could write (both using raw pointer arrays) * 14 times faster than a somewhat functional version (uses monads, but no explicit IO) * >30 times faster than fancy functional solutions with iteratees, streams, etc I'm wondering if string processing is simply a Haskell weak point, performance-wise. The problem involves many millions of very small (<10 character, usually) strings -- the C solution can copy directly from string literals into a fixed buffer and flush it occasionally, while even the fastest Haskell version has a lot of overhead from copying around arrays. Dons suggested I was "doing it wrong", so I'm posting on -cafe in the hopes that somebody can tell me how to get better performance without resorting to C. Here's the fastest Haskell version I could come up with. It discards all error handling, validation, and correctness in the name of performance, but still can't get anywhere near C: http://hpaste.org/fastcgi/hpaste.fcgi/view?id=16423 [1] http://json.org/

7 11

hsql won't install due to system.time
by Brian Denheyer 23 Jan '10

23 Jan '10

Database/HaskellDB/HSQL.hs:25:7: Could not find module `System.Time': it is a member of the hidden package `old-time-1.0.0.3' it is a member of the hidden package `old-time-1.0.0.2' Use -v to see a list of the files searched for. old-time-1.0.0.2 is installed (via cabal). Suggestions on how to fix this or which haskell db (for sqlite3) package will actually install greatly appreciated. I get a lot of failures in hackages which involve that dreaded "hidden package" message. Anybody care to enlighten us mere mortals as to why it seems to occur so frequently (at least for me). Brian

4 7

ghci debugger, can't set breakpoint
by Mike Burrell 23 Jan '10

23 Jan '10

I recently upgraded from GHC 6.8.2 to GHC 6.10.4 and can no longer set breakpoints the way I'm used to. It seems I can only set breakpoints in the current module. I Googled around and found someone with a similar error message trying to load modules which aren't interpreted, but that's not the case for me as all modules are interpreted. Here's me loading the TypeInference module (where I want to set the breakpoint) directly and having it work. Then I load the Main module and try to set the breakpoint in TypeInference with no luck. Am I setting breakpoints wrong here? Any help would be appreciated. Cheers, Mike GHCi, version 6.10.4: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer ... linking ... done. Loading package base ... linking ... done. Prelude> :l TypeInference [1 of 3] Compiling Utility ( Utility.lhs, interpreted ) [2 of 3] Compiling Parser ( Parser.hs, interpreted ) Parser.hs:1897:1: Warning: Pattern match(es) are overlapped In a case alternative: _ -> ... [3 of 3] Compiling TypeInference ( TypeInference.lhs, interpreted ) Ok, modules loaded: Parser, TypeInference, Utility. *TypeInference> :break 823 Breakpoint 0 activated at TypeInference.lhs:823:10-14 *TypeInference> :l Main [1 of 7] Compiling Utility ( Utility.lhs, interpreted ) [2 of 7] Compiling Parser ( Parser.hs, interpreted ) Parser.hs:1897:1: Warning: Pattern match(es) are overlapped In a case alternative: _ -> ... [3 of 7] Compiling TypeInference ( TypeInference.lhs, interpreted ) [4 of 7] Compiling Bounds ( Bounds.lhs, interpreted ) [5 of 7] Compiling Interpreter ( Interpreter.lhs, interpreted ) [6 of 7] Compiling PatternMatching ( PatternMatching.lhs, interpreted ) [7 of 7] Compiling Main ( Main.lhs, interpreted ) Ok, modules loaded: Parser, TypeInference, Utility, Main, Bounds, PatternMatching, Interpreter. *Main> :break TypeInference 823 No breakpoints found at that location. *Main>

1 0

Type level splices and instance deriving
by Khudyakov Alexey 23 Jan '10

23 Jan '10

Hello GHC 6.12 introduced type level splices. They are great for instances generation. They allow for much clearer and easier to understand code. However I run into problem with them. It's possible to create instance for type class which doesn't have superclass. If it does have one compiler complains that it could not deduce context. All my attempts to provide context fail. Is that accidental limitation or because of Some Good Reason? Or just due to lack of understanding on my side? Below is simplest example. > {-# LANGUAGE FlexibleInstances #-} > {-# LANGUAGE TemplateHaskell #-} > import Language.Haskell.TH > > -- OK but require FlexibleInstances > makeEq :: Name -> Q [Dec] > makeEq name = > [d| instance Eq $(conT name) where > (==) = undefined > |] > > -- Could not deduce Eq context > makeOrd :: Name -> Q [Dec] > makeOrd name = > [d| instance Ord $(conT name) where > compare = undefined > |] And GHC output: test.hs:14:17: Could not deduce (Eq t) from the context () arising from the superclasses of an instance declaration at test.hs:14:17-32 Possible fix: add (Eq t) to the context of the instance declaration In the instance declaration for `Ord t_aS5' In the Template Haskell quotation [d| instance Ord $(conT name) where { compare = undefined } |] In the expression: [d| instance Ord $(conT name) where { compare = undefined } |] -- Khudyakov Alexey

1 0

ANNOUNCE: Updates and a new member of the Monadic Regions family
by Bas van Dijk 23 Jan '10

23 Jan '10

Hello, I released updates of all the members of the Monadic Regions family: * http://hackage.haskell.org/package/regions-0.3 - Export the 'Dup' and 'ParentOf' classes from 'Control.Monad.Trans.Region'. - Add: 'mapInternalHandle ∷ (Handle resource1 → Handle resource2) → (RegionalHandle resource1 r → RegionalHandle resource2 r)'. - Internal refactoring. * http://hackage.haskell.org/package/regions-monadsfd-0.3 * http://hackage.haskell.org/package/regions-monadstf-0.3 * http://hackage.haskell.org/package/usb-safe-0.5.1 - No major changes in these packages just internal refactoring and support for regions-0.3.*. * http://hackage.haskell.org/package/safer-file-handles-0.3 - 'File' is turned into a GADT, it's parameterized with the IOMode and its constructors are exported. - Added support for bytestring IO. - Internal refactoring. * http://hackage.haskell.org/package/explicit-iomodes-0.2 This package is not a member of the family but it is required for 'safer-file-handles'. I released an update which: - Added support for bytestring IO. - Internal refactoring. Additionally I would like to announce a new member of the family: * http://hackage.haskell.org/package/regional-pointers-0.1 to quote the description: "The library allows you to allocate memory in a region yielding a regional pointer to it. When the region terminates all pointers are automatically freed. Most importantly, a pointer can't be returned from the region. So it's impossible to reference unallocated memory." This package need some more work (primarily documentation and testing) so I consider it a beta release. Note that the darcs repositories for all packages are not updated because I can't seem to get a SSH connection with code.haskell.org :-( Happy Haskell Hacking, Bas

1 1

http://trac.haskell.org down?
by James Britt 23 Jan '10

23 Jan '10

Been trying to reach http://trac.haskell.org for most of the morning, but nothing comes up. Seems I'm not alone: http://downforeveryoneorjustme.com/http://trac.haskell.org James -- Neurogami - Smart application development http://www.neurogami.com james(a)neurogami.com

4 3

GHC bug? Cabal bug? Haddock bug?
by Mark Lentczner 22 Jan '10

22 Jan '10

=== Short Story === If I build syb-with-class-0.6 via cabal (cabal configure; cabal build) in the unpacked tar directory, it builds correctly. If I build it via "cabal install" (either from the unpacked directory, or by letting cabal fetch it), then the resulting package is corrupted. In particular, the .hi interface file for Data.Generics.SYB.WithClass.Instances mentions symbols that aren't in the .a file. (Or rather, they have the wrong names.) I compared verbose logs of both builds and the differ only in temporary file names.... execpt that the "cabal install" version builds haddock, as my .cabal/conf file has documentation: True. Turns out that if turn documentation off, then then "cabal install" builds a .hi file that matches the .a file... and all is well. Is this a bug in cabal? cabal-install? ghc? haddock? I have saved logs of all this if anyone wants. - Mark (MtnViewMark) Lentczner markl(a)glyphic.com http://www.ozonehouse.com/mark/ === Versions === [2373] : cabal -V cabal-install version 0.8.0 using version 1.8.0.2 of the Cabal library [2374] : ghc -V The Glorious Glasgow Haskell Compilation System, version 6.10.4 [2376] : ghc-pkg describe haddock | grep version version: 2.4.2 === Background & Details === I was installing happstack on my Mac with my Haskell Platform (GHC 6.10.4) installation. I have successfully installed dozens of other packages in this environment before, and these results are annomalous. I kicked this off via: cabal install --user happstack This installs many packages, including syb-with-class-0.6, which compiled and installed just fine. When installing happstack-data, and compiling the file Happstack/Data/Proxy.hs, during the Template Haskell step (where things get loaded up in ghci), the build encounters this link error: [ 7 of 16] Compiling Happstack.Data.Proxy ( src/Happstack/Data/Proxy.hs, dist/build/Happstack/Data/Proxy.o ) Loading package ghc-prim ... linking ... done. Loading package integer ... linking ... done. Loading package base ... linking ... done. Loading package syb ... linking ... done. Loading package array-0.2.0.0 ... linking ... done. Loading package bytestring-0.9.1.5 ... linking ... done. Loading package containers-0.2.0.1 ... linking ... done. Loading package packedstring-0.1.0.1 ... linking ... done. Loading package pretty-1.0.1.0 ... linking ... done. Loading package template-haskell ... linking ... done. Loading package syb-with-class-0.6 ... linking ... done. (... many more loads elided...) Loading package HaXml-1.13.3 ... linking ... done. ghc: unknown symbol `_sybzmwithzmclasszm0zi6_DataziGenericsziSYBziWithClassziInstances_dataTypeZMabOQZN_closure' That symbol decodes to something referring to: package: syb-with-class-0.6 module: Data.Generics.SYB.WithClass.Instances reference: dataType[abOQ] The reference turns out to be from Loading Happstack.Data.Default, which in turn imports Data.Generics.SYB.WithClass.Instances. Poking around, I found that the interface (.hi) file for Data.Generics.SYB.WithClass.Instances does indeed export such an object: [2324] : ghc --show-iface Data/Generics/SYB/WithClass/Instances.hi | fgrep dataType[a Data.Generics.SYB.WithClass.Instances.dataType[abOQ]) -} Data.Generics.SYB.WithClass.Instances.dataType[abSm]) -} dataType[abOQ] :: Data.Generics.SYB.WithClass.Basics.DataType dataType[abSm] :: Data.Generics.SYB.WithClass.Basics.DataType But, the library doesn't export it: [2325] : nm libHSsyb-with-class-0.6.a | fgrep dataTypeZMa 0001854c D _sybzmwithzmclasszm0zi6_DataziGenericsziSYBziWithClassziInstances_dataTypeZMaeuiZN_closure 00018604 D _sybzmwithzmclasszm0zi6_DataziGenericsziSYBziWithClassziInstances_dataTypeZMaexTZN_closure These refer to dataType[aeui] and dataType[aexT], which don't exist in the interface file. Something seems amiss here: The interface file is exporting generated names that don't match what the library is exporting. If I do the same investigation with the profiling versions of this module, they match: [2326] : ghc --show-iface Data/Generics/SYB/WithClass/Instances.p_hi | fgrep dataType[a Data.Generics.SYB.WithClass.Instances.dataType[anmx]) -} Data.Generics.SYB.WithClass.Instances.dataType[anq8]) -} dataType[anmx] :: Data.Generics.SYB.WithClass.Basics.DataType dataType[anq8] :: Data.Generics.SYB.WithClass.Basics.DataType markl@mtree ~/Library/Haskell/packages/syb-with-class-0.6/lib/ghc-6.10.4 [2327] : nm libHSsyb-with-class-0.6_p.a | fgrep dataTypeZMa 00032704 D _sybzmwithzmclasszm0zi6_DataziGenericsziSYBziWithClassziInstances_dataTypeZManmxZN_closure 0003282c D _sybzmwithzmclasszm0zi6_DataziGenericsziSYBziWithClassziInstances_dataTypeZManq8ZN_closure These both export something dataType[anmx] and dataType[anq8] These two objects, by the way, stem from the last two lines of Data.Generics.SYB.WithClass.Instances.hs: $( deriveData [''ByteString] ) $( deriveData [''L.ByteString] )

3 5

classes with types which are wrapped in
by Andrew U. Frank 22 Jan '10

22 Jan '10

i encounter often a problem when i have a class with some operations (say class X with push) applied to a type A b. I then wrap A in a type A_sup, with some more type parameters and i cannot write a instance of class A_sup because i have a kind mismatch. any suggestions? (reordering of the type parameters of A_sup is not a solution, because another class operates on this parameter) here a simplistic case (i know that A could be reduced to [], my real cases are more complicated). data A b = A b [b] data Asup x ab y = Asup x ab y class X a b where push :: b -> a b -> a b instance X A Int where push b' (A b bs) = A b' (b:bs) instance X Asup Char Int Float where push b' (Asup a b c) = Asup a (push b' b) c -- this does not compile because the number of type arguments for X is wrong. if i try with a type type A_2 b = Asup Char (A b) Float instance X A_2 Int where push b' (Asup a b c) = Asup a (push b' b) c (and --TypeSynonymInstances) i get: Type synonym `A_2' should have 1 argument, but has been given 0 In the instance declaration for `X A_2 Int' what is the solution? thank you! andrew

5 4

Existential Types (I guess)
by Ozgur Akgun 22 Jan '10

22 Jan '10

Dear Cafe, I can write and use the following, data IntHolder = IntHolder Integer instance Show IntHolder where show (IntHolder n) = show n liftInt :: (Integer -> Integer) -> IntHolder -> IntHolder liftInt f (IntHolder c) = IntHolder (f c) But I cannot generalise it to *Num:* data NumHolder = forall a. Num a => NumHolder a instance Show NumHolder where show (NumHolder n) = show n liftNum :: (Num a) => (a -> a) -> NumHolder -> NumHolder liftNum f (NumHolder c) = NumHolder (f c) The error message I get is the following: Couldn't match expected type `a' against inferred type `a1' `a' is a rigid type variable bound by the type signature for `liftNum' at Lifts.hs:54:16 `a1' is a rigid type variable bound by the constructor `NumHolder' at Lifts.hs:55:11 In the first argument of `f', namely `c' In the first argument of `NumHolder', namely `(f c)' In the expression: NumHolder (f c) Regards, -- Ozgur Akgun

3 3

haddock - error
by Andrew U. Frank 22 Jan '10

22 Jan '10

the problem with me getting no haddock output whatsoever was that i had not reconfigured cabal after replacing haddock 2.4 by 2.5 (by installing it with cabal). it would be helpful if there were an error message - but i can clearly see that this is difficult to achieve in this case. thanks for the help!

1 0

Re: Is Haskell capable of matching C in string processing performance?
by John Lato 22 Jan '10

22 Jan '10

> From: Heinrich Apfelmus <apfelmus(a)quantentunnel.de> > Don Stewart wrote: >> jmillikin: >>> Here's the fastest Haskell version I could come up with. It discards >>> all error handling, validation, and correctness in the name of >>> performance, but still can't get anywhere near C: >>> http://hpaste.org/fastcgi/hpaste.fcgi/view?id=16423 >> >> Thanks for posting the code. >> >> You're not using bytestrings?? >> >> They were invented to deal with the problem of [Char] being a poor >> structure for large scale string processing, and you should have no >> problem getting C-like string performance. >> >> http://www.cse.unsw.edu.au/~dons/papers/CSL06.html > > In my limited experience, ByteStrings are great for reading data, but > not that good for writing data that is being generated on the fly. For > writing, good old difference lists or the Builder monoid / Put monad > from Data.Binary seem to be best. > If find that this is particularly true if you don't know the length of data to write in advance because it involves lots of new allocations and copying. Using Builder/Put is definitely a better approach. Out of curiousity, has anyone ever tried implementing a LinkedListString in C to check what its performance would be? I expect it would be pretty poor compared to standard strings. Cheers, John

1 0