September 2011 - Haskell-Cafe

Solving the configuration problem with parametrized modules
by Joachim Breitner 08 Sep '11

08 Sep '11

Hi Cafe, this is an idea that has been floating in my head for a while, and I’m wondering about its feasibility and, if feasible, complexity (in the range from „trivial“ over “blog post” over “paper” to “thesis”). Application authors in Haskell often face the problem of run-time constants, e.g. values that are expected to be determined once during program start and then stay fixed for the remainder of the run. Good examples are user configuration values (--verbose, certain paths, debugging level, etc.). Usually in Haskell, you’d determine them in the main function and then pass them in an argument to all functions needing them, or you wrap that in a monad. This works and has the advantage of being explicit. One can locally change the value when passing it to another function – this can be a good thing (more powerful) or a bad thing (if the programmer had runtime-constant behavior in mind, an invariant not enforced by the type system or the compiler). The big downside is the verbosity of the approach: A lot of parameters need to be passed, and if one such value is suddenly required in a function where it was not before, this function’s signature and all functions using it have to be modified. Also, I expect that the explicit passing causes a small performance penalty. A less idiomatic approach would be something like this: ====== module M confRef = unsafePerformIO $ newIORef (error "not set yet") conf = unsafePerformIO $ readIORef confRef f = ... conf ... ====== import M main = do ... writeIORef confRef theValue ... f ... ====== This has the nice effect that the programmer can use "conf" as if it is a proper constant value. It could also be quicker, as nothing needs to be passed on the stack. But besides the subtle issues about inlining, this has the big issue of glaring unsafeness: There is no guarantee that confRef is actually set before conf is evaluated, and any code can easily change the confRef anywhere during the runtime. I’d like to propose (on the theoretical level for now, don’t worry) an extension to the language to support _parametrized modules_, which work internally similar to the example above, but with the compiler statically ensuring correctness, where correctness means * The parameter is not evaluated before it is set. * It is set at most once. Rouch syntax sketch: ====== module M parameters (conf1, conf2) f = ... conf ... ====== import M main = do ... M.setParameters someVal1 someVal2 ... f ... ====== where setParameters is a symbol generated by the compiler. The tricky part is, of course, ensuring the correctness. For example while > main = do > M.setParameters 1 2 > ...f... should be easy to check, how about > main = do > id $ M.setParameters 1 2 > ...f... or even > main = do > bool <- someIOAction > guard bool $ M.setParameters 1 2 > ...f... ? I envision a ghc extensions that would allow to one to hook into the type checker and attach bits of information (here: „uses conf“ and „initizalizes M“) to the types, and also check certain conditions (e.g. a >>= b is invalid if a and b both have „initializes M“ attached). If this extension was loaded when id was compiled, it should probably accept the second case, but when no such information is available, or in the third case, reject the code. A framework for such attached bits would surely enable more extensions that statically check some condition that is hard or unwieldy to express in the type system. I have more ideas in that direction, but before I start brabbling, maybe I should first wait for your reactions. Do you think this could be useful (from a user point of view)? Has this idea maybe already been proposed? Thanks, Joachim -- Joachim "nomeata" Breitner mail(a)joachim-breitner.de | nomeata(a)debian.org | GPG: 0x4743206C xmpp: nomeata(a)joachim-breitner.de | http://www.joachim-breitner.de/

8 12

Haskell Weekly News: Issue 198
by Daniel Santa Cruz 07 Sep '11

07 Sep '11

Welcome to issue 198 the HWN, a newsletter covering developments in the Haskell community. This release covers the week of August 28 to September 3, 2011. Announcements Eric Y. Know released the 5th edition of the Parallel Haskell Digest. [1] http://goo.gl/Jqx01 New and Updated Projects * swapper (New - Roman Smrz) Provides a wrapper for functors, which allows their data to be automatically and transparently swapped to disk and loaded back when necessary. [2] http://goo.gl/CwojE * GA (New - Kenneth Hoste) Provides support for working with genetic algorithms in Haskell. [3] http://goo.gl/XjvFA * hledger[-web] (Update - Simon Michael) Bugfix update. [4] http://goo.gl/98ApM * bindings-DSL (Update - Mauricio CA) Provides a language to describe foreign interfaces on top of hsc2hs. [5] http://goo.gl/fFL4Y * hspec (Update - Trystan Spangler) This latest release includes a command line runner that can search through directory trees for specs and support for nested specs. [6] http://goo.gl/rVbF0 * EclipseFP (Update - JP Moresmau) New features and bugfixes. [7] http://goo.gl/dXS6t Quotes of the Week * dankna: I do [gui on osx] by writing a polyglot ObjC/Haskell program * maurer: I'd do it kmc style * hpc: your next challenge: write a definition of fix in java that fits in a twitter post * kmc: i leave for 3 seconds and you're all metaprogramming and shit * acowley: Haskellers are so abstract they can't even spell primitive * edwardk: i was happy that i had the 4th entry on google for a topic... but then i realized it was only for me, not for everyone that searches for it * kmc: oh, my suggestion was to hand off the fd's directly through a UNIX socket ... but that's probably silly ... i just like the fact that fd's can be passed through a unix socket * monochrom: "linear types change the world!" says Wadler. "Record updates change the type!" says Haskell. :) * gosu: Type parameters are covariant. This is not sound, and that does not matter. * Yitzchak Gale: I certainly wouldn't dream of dragging in type-level olegery, unsafe coercion, implicit parameters and other experimental extensions. Simplicity just works. Top Reddit Stories * 8 ways to report errors in Haskell revisited : Inside 233 Domain: blog.ezyang.com, Score: 51, Comments: 33 On Reddit: [8] http://goo.gl/RBcIL Original: [9] http://goo.gl/A2m2Z * Yesod 0.9 Released! A Massive Changelog explained. Domain: yesodweb.com, Score: 43, Comments: 6 On Reddit: [10] http://goo.gl/go7Zo Original: [11] http://goo.gl/dGoMa * EclipseFP 2.1.0 released, with Hoogle and HLint integration, and much more! Domain: jpmoresmau.blogspot.com, Score: 40, Comments: 42 On Reddit: [12] http://goo.gl/6E1QV Original: [13] http://goo.gl/mkx7g * And they say complexity has no philosophical implications Domain: scottaaronson.com, Score: 36, Comments: 9 On Reddit: [14] http://goo.gl/g97k3 Original: [15] http://goo.gl/9OGU7 * data Maybe -- harmful? Domain: blog.dbpatterson.com, Score: 29, Comments: 26 On Reddit: [16] http://goo.gl/fd2QC Original: [17] http://goo.gl/x4ibO * haskell-TLS framework progress report Domain: tab.snarc.org, Score: 29, Comments: 14 On Reddit: [18] http://goo.gl/Mdc8x Original: [19] http://goo.gl/UpyZ6 * I made a simple image upload site using haskell, sqlite3 and lighttpd. Check it out. Domain: self.haskell, Score: 27, Comments: 15 On Reddit: [20] http://goo.gl/vuF8E Original: [21] http://goo.gl/vuF8E * Manuel Chakravarty on Data Parallel Haskell Domain: vimeo.com, Score: 26, Comments: 2 On Reddit: [22] http://goo.gl/CLXJU Original: [23] http://goo.gl/jCf9w * Functional Programming Day : Oct 14th, Cambridge, UK SPJ & D Syme - early book. bef. Aug 31st Domain: fpday.net, Score: 18, Comments: 11 On Reddit: [24] http://goo.gl/wbIWr Original: [25] http://goo.gl/8OPgK * How does AwesomePrelude transform non-strict code to efficient strict code? Domain: self.haskell, Score: 18, Comments: 12 On Reddit: [26] http://goo.gl/VgIQt Original: [27] http://goo.gl/VgIQt * Composable MVCs (or UIs), a google talk by Conal Elliot Domain: youtube.com, Score: 18, Comments: 14 On Reddit: [28] http://goo.gl/9ekmO Original: [29] http://goo.gl/GDEkN Top StackOverflow Questions * Good examples of Not a Functor/Functor/Applicative/Monad? votes: 17, answers: 3 Read on SO: [30] http://goo.gl/F2q5y * Haskell: What monad did I just reinvent? votes: 17, answers: 4 Read on SO: [31] http://goo.gl/HCko4 * Why is Scalas type inference not as powerful as Haskells? votes: 17, answers: 4 Read on SO: [32] http://goo.gl/gvCF1 * Hard to understand Haskell memory allocation behaviour votes: 11, answers: 1 Read on SO: [33] http://goo.gl/X7HtB * Haskell on JVM? votes: 11, answers: 1 Read on SO: [34] http://goo.gl/Ra4kA * Implementations of spatial indexes in Haskell? votes: 11, answers: 1 Read on SO: [35] http://goo.gl/ThqRA * Does Haskell have variadic functions/tuples? votes: 9, answers: 3 Read on SO: [36] http://goo.gl/NLYCE * couldn't match expected type (Int -> Int -> Int) with actual type `(t0, t1, t2)' votes: 9, answers: 2 Read on SO: [37] http://goo.gl/5EmAS * What is the advantage of using a parser generator like happy as opposed to using parser combinators? votes: 9, answers: 3 Read on SO: [38] http://goo.gl/04s9n * manually defining boolean conjunction votes: 8, answers: 1 Read on SO: [39] http://goo.gl/FGbAj About the Haskell Weekly News To help create new editions of this newsletter, please send stories to dstcruz(a)gmail.com. Until next time, Daniel Santa Cruz References 1. http://permalink.gmane.org/gmane.comp.lang.haskell.cafe/92123 2. http://permalink.gmane.org/gmane.comp.lang.haskell.general/18903 3. http://permalink.gmane.org/gmane.comp.lang.haskell.general/18908 4. http://permalink.gmane.org/gmane.comp.finance.ledger.hledger/479 5. http://permalink.gmane.org/gmane.comp.lang.haskell.general/18909 6. http://permalink.gmane.org/gmane.comp.lang.haskell.general/18912 7. http://permalink.gmane.org/gmane.comp.lang.haskell.cafe/92150 8. http://blog.ezyang.com/2011/08/8-ways-to-report-errors-in-haskell-revisited/ 9. http://www.reddit.com/r/haskell/comments/jy34j/8_ways_to_report_errors_in_h… 10. http://www.yesodweb.com/blog/2011/08/announcing-yesod-0.9 11. http://www.reddit.com/r/haskell/comments/jy42c/yesod_09_released_a_massive_… 12. http://jpmoresmau.blogspot.com/2011/09/eclipsefp-210-released.html 13. http://www.reddit.com/r/haskell/comments/k2idm/eclipsefp_210_released_with_… 14. http://www.scottaaronson.com/blog/?p=103 15. http://www.reddit.com/r/haskell/comments/k2kuu/and_they_say_complexity_has_… 16. http://blog.dbpatterson.com/post/9528836599 17. http://www.reddit.com/r/haskell/comments/jxj8i/data_maybe_harmful/ 18. http://tab.snarc.org/post/2011/8/31/tls_progress/ 19. http://www.reddit.com/r/haskell/comments/k12g7/haskelltls_framework_progres… 20. http://www.reddit.com/r/haskell/comments/k2b10/i_made_a_simple_image_upload… 21. http://www.reddit.com/r/haskell/comments/k2b10/i_made_a_simple_image_upload… 22. http://vimeo.com/28477220 23. http://www.reddit.com/r/haskell/comments/k3x2l/manuel_chakravarty_on_data_p… 24. http://www.fpday.net/fpday2011/index.php 25. http://www.reddit.com/r/haskell/comments/jz592/functional_programming_day_o… 26. http://www.reddit.com/r/haskell/comments/k05fo/how_does_awesomeprelude_tran… 27. http://www.reddit.com/r/haskell/comments/k05fo/how_does_awesomeprelude_tran… 28. http://www.youtube.com/watch?v=faJ8N0giqzw 29. http://www.reddit.com/r/haskell/comments/k14k7/composable_mvcs_or_uis_a_goo… 30. http://stackoverflow.com/questions/7220436/good-examples-of-not-a-functor-f… 31. http://stackoverflow.com/questions/7223901/haskell-what-monad-did-i-just-re… 32. http://stackoverflow.com/questions/7234095/why-is-scalas-type-inference-not… 33. http://stackoverflow.com/questions/7247058/hard-to-understand-haskell-memor… 34. http://stackoverflow.com/questions/7261039/haskell-on-jvm 35. http://stackoverflow.com/questions/7270796/implementations-of-spatial-index… 36. http://stackoverflow.com/questions/7220953/does-haskell-have-variadic-funct… 37. http://stackoverflow.com/questions/7232375/couldnt-match-expected-type-int-… 38. http://stackoverflow.com/questions/7269025/what-is-the-advantage-of-using-a… 39. http://stackoverflow.com/questions/7220297/manually-defining-boolean-conjun…

1 0

Fwd: Is there any way to parametrize a value update using record syntax?
by David Barbour 07 Sep '11

07 Sep '11

forgot to CC list. ---------- Forwarded message ---------- From: David Barbour <dmbarbour(a)gmail.com> Date: 2011/9/6 Subject: Re: [Haskell-cafe] Is there any way to parametrize a value update using record syntax? To: Poprádi Árpád <popradi_arpad(a)freemail.hu> 2011/9/6 Poprádi Árpád <popradi_arpad(a)freemail.hu> > But it's ugly. Always the same, only the record selector has another > name. Is it possible to generalize it? > You can generalize using template haskell. I believe Oleg's HList already provides such mechanisms, so you don't need to do this yourself. If you're doing this a lot, try the HList package. (If not, just do the Also, I would say you've too tightly coupled your BigData to the MonadicEnv. I suggest you reduce it instead to: setX :: X -> BigData -> BigData setX x' bd = bd { dataX = x' } updX :: (X -> X) -> BigData -> BigData updX fx bd = bd { dataX = fx (dataX bd) } Then in your state monad you can use: modify (setX x') gets dataX And use of 'updX' is much more composable. That aside, from personal experience, I'm usually okay just using: modify (\ s -> s { dataX = x' }) in the few places I need it.

2 1

ghc-mtl and ghc-7.2.1
by José Romildo Malaquias 07 Sep '11

07 Sep '11

Hello. In order to compile ghc-mtl-1.0.1.0 (the latest released version) with ghc-7.2.1, I would apply the attached patch, which removes any references to WarnLogMonad. ghc-7.2.1 does not have the monad WarnLogMonad anymore. As I do not know the details of the GHC api, I am not sure if this is enough to use ghc-mtl with ghc-7.2.1. I want ghc-mtl in order do build lambdabot. Any thoughts? Romildo

2 1

Is there any way to parametrize a value update using record syntax?
by Poprádi Árpád 07 Sep '11

07 Sep '11

Hi all, i have a record with a lot of items used in a state monad. data BigData = BigData { data1 :: X , data2 :: X -- and so on } type MonadicEnv a = State BigData a I updates the fields in the computation with such functions: updateData1 :: X -> MonadicEnv() updateData1 d = do; env <- get; put env {data1 = d} updateData2 :: X -> MonadicEnv() updateData2 d = do; env <- get; put env {data2 = d} But it's ugly. Always the same, only the record selector has another name. Is it possible to generalize it? Thanks, Árpád

5 5

Data.IArray rant
by Evan Laforge 07 Sep '11

07 Sep '11

I never liked Data.IArray. Recently I had to deal with it again and it kind of touched off a bit of a rant: Array bounds are an inclusive range, which runs counter to almost all other uses of ranges, and for good reason. Length of an array is (high - low + 1). If you want to convert a list to an array it's 'IArray.listArray (0, length xs - 1) xs'. And if you forget to subtract one or add one it's likely to be a runtime crash. Off-by-one errors and runtime crashes... I've gotten accustomed to not having those in haskell! The vast majority of data types have their own set of monomorphic fuctions, and yet array all of the sudden wants to be an interface. That's nice and all, but who actually uses that interface? I recall there used to be a DiffArray and some other variants, but they were decried as being slow and appear to be gone now. So it appears that there's a complexity price paid to support different array types under the same interface, namely giant class contexts that can't be abstracted away and dependence on MPTCs... and appears to be unused. And then another price paid to support arbitrary indices... which I at least have never used. Maybe there is a field where arrays with strange index ranges like 5--20 are useful, but I always just want 0--length-1. Meanwhile, they lack basic features like concatenation. No Monoid instance. In fact, hardly any useful instances. The result is that my first contact with haskell arrays left me with the impression that they were complicated, hard to use, and designed for someone with different priorities than me. Of course, Data.Vector didn't exist back then, but if someone were new to haskell now I would recommend they skip Data.IArray and head straight for vector. If they wanted 2 dimensional arrays then I know there are some matrix libraries. Of course, sometimes IArray is hard to avoid, e.g. Data.Graph is just a type synonym for an array. I wrote a small library of graph utilities, and dealing with the arrays was back to awkward land. So... am I alone in this dislike of IArray? Or maybe someone can explain why it's actually a very useful interface? Admittedly vector is a mere 'cabal install' away, but array has a blessed status as a ghc bundled library and is likely to be the first place people will look, as well as the natural base for other data types that want an array, like Data.Graph. I know it's not as simple as "just debundle IArray" because it's a bootlib and presumably used in ghc itself, though I know there is support for reducing the set of bootlibs. There's also the whole mutable interface, which I haven't used, but maybe I'd like it more. I know the array stuff winds up integrated fairly deeply with ghc's unboxed arrays and whatnot, so there's a lot of detail I don't understand. I just know I dread going back to array using code while I tend to enjoy working with vector, bytestring, text, storablevector, etc.

7 10

Memoization of functions
by Michael Orlitzky 07 Sep '11

07 Sep '11

I'm working on a program where I need to compute a gajillion (171442176) polynomials and evaluate them more than once. This is the definition of the polynomial, and it is expensive to compute: > polynomial :: Tetrahedron -> (RealFunction Point) > polynomial t = > sum [ (c t i j k l) `cmult` (beta t i j k l) | i <- [0..3], > j <- [0..3], > k <- [0..3], > l <- [0..3], > i + j + k + l == 3] Currently, I'm storing the polynomials in an array, which is quickly devoured by the OOM killer. This makes me wonder: how much memory can I expect to use storing a function in an array? Is it possible to save some space through strictness? Does that question even make sense?

2 1

Re: [Haskell-cafe] Configuration Problem and Plugins
by M. George Hansen 06 Sep '11

06 Sep '11

On Sat, Sep 3, 2011 at 12:33 AM, Max Rabkin <max.rabkin(a)gmail.com> wrote: > On Sat, Sep 3, 2011 at 03:15, M. George Hansen <technopolitica(a)gmail.com> wrote: >> Greetings, >> >> I'm a Python programmer who is relatively new to Haskell, so go easy on me :) >> >> I have a program that uses (or will use) plugins to render output to >> the user in a generic way. I'm basing the design of the plugin >> infrastructure on the Plugins library, and have the following >> interface: >> >> data Renderer = Renderer { >> initialize :: IO (), >> destroy :: IO (), >> render :: SystemOutput -> IO () >> } > > How about having initialize return the render (and destroy, if > necessary) functions: > > initialize :: IO (SystemOutput -> IO ()) > > or > > initialize :: IO (SystemOutput -> IO (), IO()) > Thanks for your reply. That does seem like the best solution, I'll give it a try. -- M. George Hansen

2 1

EclipseFP fails to install
by Henk-Jan van Tuyl 06 Sep '11

06 Sep '11

L.S., I have installed Eclipse 3.7 and EclipseFP 2.1.0 on my Windows XP computer After trying to install EclipseFP in Eclipse, I got the following message: C:\Programs\Haskell Platform\2011.2.0.1\bin\ghc-pkg.exe dump --global cabal.0.6.0.exe: failed to parse output of 'ghc-pkg dump' What can I do about this? Regards, Henk-Jan van Tuyl -- http://Van.Tuyl.eu/ http://members.chello.nl/hjgtuyl/tourdemonad.html --

2 1

ANN: Netwire 1.2.0
by Ertugrul Soeylemez 06 Sep '11

06 Sep '11

Hello there, version 1.2.1 of netwire is out. New features include: * Completely reworked event system. Events are now solely based on signal inhibition, which turns out to be much more convenient and faster than AFRP's traditional approach using Maybe-wrapped values. Now instead of checking for the presence of an event, you simply assume that it happened. If it didn't happen, then the signal is inhibited and your wire is not run. This makes your code much simpler, because you get along almost entirely without switches. Example displaying a discrete clock updated at intervals of half a second: system :: forall m. Monad m => Wire m () String system = proc _ -> printf "%8.2f" ^<< hold discreteClock -< () where discreteClock :: Wire m () Double discreteClock = proc _ -> do t <- time -< () repeatedly -< (0.5, t) By using ArrowPlus or Alternative, you can handle the case, where the event did not happen, listening for alternative events or simply having a fallback. By using the various wire combinators you can hold, sample, keep or exhibit event values as shown above. * Lots and lots of Haddock documentation improvements. Now the inhibition and feedback behaviour of every builtin wire is documented. * Apropos feedback: Finally added an ArrowLoop instance, so you can now have recursive values in your wires. There is a catch though: Right now wires which produce values used recursively must not inhibit. There is an inherent problem handling the inhibition case, which you will observe as a pattern match error. Right now I don't have a solution for this problem, but I'm working on it. However, it may well be that it simply cannot be solved with my current internal wire representation. This is the price to pay for signal inhibition. To prevent (or rather catch) inhibition you can use the 'exhibit' or 'event' combinators. * Many small performance improvements. Simple signal networks go well beyond 35000 frames per second on my mobile Intel i7 with 1.7 GHz and their executables cost well less than 5 MiB RAM. The performance passes the 50k FPS mark on my i5 with 2.8 GHz at home. This is more than a 10x speedup compared to my last benchmark on the initial release of version 1.0.0 * Removed some of variations of existing wires and replaced them by wire combinators instead. For example 'executeOnce' and 'executeEvery' are gone. Instead you should use the normal 'execute' wire together with 'swallow', 'sample', 'hold' or other combinators. * Changed the semantics of all wires to support feedback. Now some wires support feedback right away, while for others you need a one-instant delay. The feedback behaviour of all wires is Haddock-documented, as noted. Enjoy and please give me as much feedback as possible. =) Greets, Ertugrul -- nightmare = unsafePerformIO (getWrongWife >>= sex) http://ertes.de/

1 1