October 2013 - Haskell-Cafe

Nested monadic monoids via Traversable?
by Hans Höglund 13 Oct '13

13 Oct '13

I have been experimenting with compositions of monads carrying associated monoids (i.e. Writer-style) and discovered the following pattern: ---------------------------------------------------------------------- {-# LANGUAGE DeriveFunctor, DeriveFoldable, DeriveTraversable, GeneralizedNewtypeDeriving #-} import Control.Monad import Control.Monad.Writer hiding ((<>)) import Data.Semigroup import Data.Foldable (Foldable) import Data.Traversable (Traversable) import qualified Data.Traversable as Traversable newtype Foo m a = Foo (Writer m a) deriving (Monad, MonadWriter m, Functor, Foldable, Traversable) newtype Bar m a = Bar { getBar :: [Foo m a] } deriving (Semigroup, Functor, Foldable, Traversable) instance Monoid m => Monad (Bar m) where return = Bar . return . return Bar ns >>= f = Bar $ ns >>= joinedSeq . fmap (getBar . f) where joinedSeq = fmap join . Traversable.sequence runFoo (Foo x) = runWriter x runBar (Bar xs) = fmap runFoo xs ---------------------------------------------------------------------- That is, given a type that is Monadic and Traversable, we can define a list of the same type as a monad, whose binding action "glues together" the nested Monoid values. A trivial example: ---------------------------------------------------------------------- -- annotate all elements in bar tells :: String -> Bar String a -> Bar String a tells a (Bar xs) = Bar $ fmap (tell a >>) xs -- a bar with no annotations x :: Bar String Int x = return 0 -- annotations compose with >>= y :: Bar String Int y = x <> tells "a" x >>= (tells "b" . return) -- and with join z :: Bar String Int z = join $ tells "d" $ return (tells "c" (return 0) <> return 1) -- runBar y ==> [(0,"b"),(0,"ab")] -- runBar z ==> [(0,"dc"),(1,"d")] ---------------------------------------------------------------------- However, I am concerned about the (Monad Bar) instance which seems ad-hoc to me, especially the use of sequence. Is there a more general pattern which uses a class other than Traversable? Any pointers would be much appreciated. Regards, Hans

1 1

typeclass to select a list element
by Paolino 12 Oct '13

12 Oct '13

Hello, I'm trying to use a type class to select an element from a list. I would like to have a String "CC" as a value for l10'. {-# LANGUAGE MultiParamTypeClasses, GADTs,FlexibleInstances, DataKinds ,TypeFamilies, KindSignatures, FlexibleContexts, OverlappingInstances, StandaloneDeriving, UndecidableInstances #-} import Data.Nat import Data.Monoid data family X (n::Nat) :: * data L (n::Nat) where Q :: (Monoid (X n), Show (X n)) => L (Succ n) -> X n -> L n E :: Monoid (X n) => L n deriving instance Show (L n) data instance X n = String String instance Monoid (X n) where String x `mappend` String y = String $ x `mappend` y mempty = String "" deriving instance Show (X n) class Compose n n' where compose :: L n -> L n -> X n' instance Compose n n where compose (Q _ x) (Q _ y) = x `mappend` y compose _ _ = mempty instance Compose n n' where compose (Q x _) (Q y _) = compose x y compose _ _ = mempty l0 :: L Zero l0 = Q (Q E $ String "C") $ String "A" l0' :: L Zero l0' = Q (Q E $ String "C") $ String "B" l10' :: X (Succ Zero) l10' = compose l0 l0' l00' :: X Zero l00' = compose l0 l0' {- *Main> l00' String "AB" *Main> l10' String "" -} Thanks for help. paolino

3 3

Haskellers in Nashville,TN
by heathmatlock 12 Oct '13

12 Oct '13

Anyone here from the Nashville, TN area? t would be nice to meet regularly with others to discuss and work with Haskell. Also, this group is now available: www.meetup.com/The-Haskell-Study-Group/ If you want to participate remotely, that works too, I'm just looking for others to meet with regularly. -- Heath Matlock +1 256 274 4225

1 0

Haskell Weekly News: Issue 282
by Daniel Santa Cruz 10 Oct '13

10 Oct '13

Welcome to issue 282 of the HWN, an issue covering crowd-sourced bits of information about Haskell from around the web. This issue covers the week of September 22 to October 05, 2013. Quotes of the Week * monochrom: . o O ( error messages are like legalese. it contains complete information so that later lawyers can say "told you so" ) * BetaZiliani: "We escape from monads in the same way that we escape from a building on fire: we run." (#ICFP2013) * ReinH: ReinH's Lemma: There's always a connection to Yoneda's Lemma. * startling: it's sort of depressing when you spend a lot of time with traversals and things to get all your crucial algorithms down to like ten lines and then have like two hundred lines each for pretty-printing and for file I/O. * edwardk: you know that feeling when you wake up in the morning, and your teeth feel fuzzy, you don't know where you've been, but you strongly suspect in 9 months someone will be calling you for library support? * tac: I free the forgotten. * Anonymous: What do you call someone reading a book about category theory? A co-author! * Anonymous: A co-worker == generator of work tasks? * edwardkmett: At one point we had a type in lens where one of its arguments started taking parameters of the form i m a s t a b u. Upon reflection, we let it win the argument and decided not to implement the function. Top Reddit Stories * xkcd - Functional Programming Domain: xkcd.com, Score: 97, Comments: 58 On Reddit: [1] http://goo.gl/XSSTOR Original: [2] http://goo.gl/daH8om * Hackage 2 is up! Domain: hackage.haskell.org, Score: 96, Comments: 20 On Reddit: [3] http://goo.gl/hSyulq Original: [4] http://goo.gl/admcV * Programming - Why Haskell is Great - 10 minutes Domain: youtube.com, Score: 91, Comments: 18 On Reddit: [5] http://goo.gl/SoTDhR Original: [6] http://goo.gl/CwsaFq * GHC: now with -j flag for parallel builds Domain: ghc.haskell.org, Score: 88, Comments: 27 On Reddit: [7] http://goo.gl/tX0YZr Original: [8] http://goo.gl/8SepKH * I can finally say that I'm a Haskell developer (and I'm hiring)! Domain: self.haskell, Score: 85, Comments: 42 On Reddit: [9] http://goo.gl/fr2adZ Original: [10] http://goo.gl/fr2adZ * Hackage 2 switchover imminent Domain: haskell.org, Score: 80, Comments: 14 On Reddit: [11] http://goo.gl/5XyWkt Original: [12] http://goo.gl/Hf3XlL * The Haskell Cast #2 - Don Stewart on Real World Haskell Domain: haskellcast.com, Score: 72, Comments: 11 On Reddit: [13] http://goo.gl/JPukzq Original: [14] http://goo.gl/iOssHA * [ANN] Nomyx, the game where you can change the rules (0.3 beta) Domain: nomyx.net, Score: 57, Comments: 14 On Reddit: [15] http://goo.gl/Z3IFfL Original: [16] http://goo.gl/gmsH5e * The aftermath of my GSoC project Domain: self.haskell, Score: 55, Comments: 6 On Reddit: [17] http://goo.gl/2vAjMs Original: [18] http://goo.gl/2vAjMs * SPJ in need for compelling "lens" use-cases Domain: permalink.gmane.org, Score: 51, Comments: 57 On Reddit: [19] http://goo.gl/6VcDMh Original: [20] http://goo.gl/wEgQUi * Tidal cycles - loops and rhythms made with tweet sized Haskell Domain: twitter.com, Score: 50, Comments: 5 On Reddit: [21] http://goo.gl/zcQqrm Original: [22] http://goo.gl/10bSjH * Who'd have thought: I made a serial monitor with Threepenny-gui Domain: self.haskell, Score: 48, Comments: 14 On Reddit: [23] http://goo.gl/rFBlWl Original: [24] http://goo.gl/rFBlWl * Arrowised Material Shaders in Purely Functional Ray Tracing Domain: tcsavage.org, Score: 44, Comments: 2 On Reddit: [25] http://goo.gl/RHvpLq Original: [26] http://goo.gl/7cDmWq * Work at Erudify Domain: jaspervdj.be, Score: 42, Comments: 5 On Reddit: [27] http://goo.gl/mpFnRa Original: [28] http://goo.gl/gcSs3U * The Universal Properties of Map, Fold, and Filter Domain: jeremykun.com, Score: 41, Comments: 0 On Reddit: [29] http://goo.gl/CksT61 Original: [30] http://goo.gl/j7rSW6 * Conal Elliott : Searching for GUI and Graphics libraries in Haskell Domain: haskell.org, Score: 39, Comments: 29 On Reddit: [31] http://goo.gl/6krMxv Original: [32] http://goo.gl/EYG0af * Type level record syntax... would this make sense? Domain: self.haskell, Score: 39, Comments: 29 On Reddit: [33] http://goo.gl/ysjr3o Original: [34] http://goo.gl/ysjr3o * Haskell Resources List Domain: gist.github.com, Score: 38, Comments: 2 On Reddit: [35] http://goo.gl/vZ9yMH Original: [36] http://goo.gl/Oo75BT * [ANN] yi-0.7 Domain: haskell.org, Score: 38, Comments: 14 On Reddit: [37] http://goo.gl/FihoH9 Original: [38] http://goo.gl/pfx5r5 * Why PVP doesn't work Domain: ro-che.info, Score: 38, Comments: 106 On Reddit: [39] http://goo.gl/XdpZpG Original: [40] http://goo.gl/TkEBzt * Live blog of CUFP 2013 -- a prodigious and growing work of reporting by Anil Madhavapeddy Domain: syslog.cl.cam.ac.uk, Score: 35, Comments: 0 On Reddit: [41] http://goo.gl/JDJJAz Original: [42] http://goo.gl/SvsyBE * What Haskell tools do you use in every project? Domain: self.haskell, Score: 35, Comments: 22 On Reddit: [43] http://goo.gl/9z8Ylr Original: [44] http://goo.gl/9z8Ylr * GHC Core by example, episode 2: Evaluation Domain: alpmestan.com, Score: 35, Comments: 3 On Reddit: [45] http://goo.gl/i7UzV1 Original: [46] http://goo.gl/kuntPM * Oh my laziness! - An introduction to laziness, strictness and unpacking in GHC Haskell Domain: alpmestan.com, Score: 35, Comments: 16 On Reddit: [47] http://goo.gl/Gfwhd0 Original: [48] http://goo.gl/84omB7 * Job opportunity for junior web developer looking to get into Haskell Domain: self.haskell, Score: 35, Comments: 11 On Reddit: [49] http://goo.gl/FqFDff Original: [50] http://goo.gl/FqFDff * Short vlog style video "My Road to Haskell" by jekor Domain: youtube.com, Score: 33, Comments: 1 On Reddit: [51] http://goo.gl/sqjXcT Original: [52] http://goo.gl/GzjCs9 * Arguing against monad transformers for the right reasons Domain: blog.ezyang.com, Score: 32, Comments: 20 On Reddit: [53] http://goo.gl/b2tFy6 Original: [54] http://goo.gl/OfGzvr * Haskell Launchpad Interaction Domain: youtube.com, Score: 32, Comments: 18 On Reddit: [55] http://goo.gl/Qwnhvw Original: [56] http://goo.gl/VlyUqn Top StackOverflow Questions * Sets, Functors and Eq confusion votes: 36, answers: 3 Read on SO: [57] http://goo.gl/2TUC5X * writing a C# version of haskell infinite fibonacci series function votes: 12, answers: 5 Read on SO: [58] http://goo.gl/Nys1p7 * Enforced pattern order votes: 10, answers: 5 Read on SO: [59] http://goo.gl/4yPfrn * Haskell, multiplying Int and Float within a function votes: 10, answers: 4 Read on SO: [60] http://goo.gl/nVsl8Z * What requirements does non-strict semantics of Haskell have on the evaluation strategy? votes: 9, answers: 1 Read on SO: [61] http://goo.gl/wqWIVk * Random access on a huge file in haskell votes: 8, answers: 3 Read on SO: [62] http://goo.gl/gxes9Y * Not using a variable for <- votes: 8, answers: 4 Read on SO: [63] http://goo.gl/jZ9Eqk * A common pattern involving composition of functions (\a b -> f (g a) (g b)) votes: 8, answers: 2 Read on SO: [64] http://goo.gl/FVY6Zl * Haskell data :> constructor meaning votes: 8, answers: 1 Read on SO: [65] http://goo.gl/0ouMUi * In functional reactive programming, how do you share state between two parts of the application? votes: 7, answers: 1 Read on SO: [66] http://goo.gl/vY3Lsq Until next time, [67]+Daniel Santa Cruz References 1. http://xkcd.com/1270/ 2. http://www.reddit.com/r/haskell/comments/1n8gor/xkcd_functional_programming/ 3. http://hackage.haskell.org/ 4. http://www.reddit.com/r/haskell/comments/1n79sh/hackage_2_is_up/ 5. https://www.youtube.com/watch?v=RqvCNb7fKsg 6. http://www.reddit.com/r/haskell/comments/1nfbdm/programming_why_haskell_is_… 7. http://ghc.haskell.org/trac/ghc/ticket/910 8. http://www.reddit.com/r/haskell/comments/1mzfob/ghc_now_with_j_flag_for_par… 9. http://www.reddit.com/r/haskell/comments/1n8zmp/i_can_finally_say_that_im_a… 10. http://www.reddit.com/r/haskell/comments/1n8zmp/i_can_finally_say_that_im_a… 11. http://www.haskell.org/pipermail/haskell-cafe/2013-September/109007.html 12. http://www.reddit.com/r/haskell/comments/1n2rqu/hackage_2_switchover_immine… 13. http://www.haskellcast.com/episode/002-don-stewart-on-real-world-haskell/ 14. http://www.reddit.com/r/haskell/comments/1mymdf/the_haskell_cast_2_don_stew… 15. http://www.nomyx.net/ 16. http://www.reddit.com/r/haskell/comments/1nq468/ann_nomyx_the_game_where_yo… 17. http://www.reddit.com/r/haskell/comments/1n4ips/the_aftermath_of_my_gsoc_pr… 18. http://www.reddit.com/r/haskell/comments/1n4ips/the_aftermath_of_my_gsoc_pr… 19. http://permalink.gmane.org/gmane.comp.lang.haskell.libraries/20400 20. http://www.reddit.com/r/haskell/comments/1nn61l/spj_in_need_for_compelling_… 21. http://twitter.com/tidalcycles/ 22. http://www.reddit.com/r/haskell/comments/1npjrq/tidal_cycles_loops_and_rhyt… 23. http://www.reddit.com/r/haskell/comments/1nidyp/whod_have_thought_i_made_a_… 24. http://www.reddit.com/r/haskell/comments/1nidyp/whod_have_thought_i_made_a_… 25. http://tcsavage.org/blog/2013/06/25/arrowised-materials-in-pfrt/ 26. http://www.reddit.com/r/haskell/comments/1nh0f9/arrowised_material_shaders_… 27. http://jaspervdj.be/posts/2013-09-29-erudify.html 28. http://www.reddit.com/r/haskell/comments/1ngbvi/work_at_erudify/ 29. http://jeremykun.com/2013/09/30/the-universal-properties-of-map-fold-and-fi… 30. http://www.reddit.com/r/haskell/comments/1nqmyg/the_universal_properties_of… 31. http://www.haskell.org/pipermail/haskell-cafe/2013-September/109022.html 32. http://www.reddit.com/r/haskell/comments/1n8c5i/conal_elliott_searching_for… 33. http://www.reddit.com/r/haskell/comments/1nf2ss/type_level_record_syntax_wo… 34. http://www.reddit.com/r/haskell/comments/1nf2ss/type_level_record_syntax_wo… 35. https://gist.github.com/leroux/6395804 36. http://www.reddit.com/r/haskell/comments/1n9z50/haskell_resources_list/ 37. http://www.haskell.org/pipermail/haskell-cafe/2013-October/109156.html 38. http://www.reddit.com/r/haskell/comments/1nn1vw/ann_yi07/ 39. http://ro-che.info/articles/2013-10-05-why-pvp-doesnt-work.html 40. http://www.reddit.com/r/haskell/comments/1ns193/why_pvp_doesnt_work/ 41. http://www.syslog.cl.cam.ac.uk/2013/09/22/liveblogging-cufp-2013/ 42. http://www.reddit.com/r/haskell/comments/1n10u8/live_blog_of_cufp_2013_a_pr… 43. http://www.reddit.com/r/haskell/comments/1nhntw/what_haskell_tools_do_you_u… 44. http://www.reddit.com/r/haskell/comments/1nhntw/what_haskell_tools_do_you_u… 45. http://alpmestan.com/2013/10/01/ghc-core-by-example-episode-2-evaluation/ 46. http://www.reddit.com/r/haskell/comments/1nktj7/ghc_core_by_example_episode… 47. http://alpmestan.com/2013/10/02/oh-my-laziness/ 48. http://www.reddit.com/r/haskell/comments/1nlxlo/oh_my_laziness_an_introduct… 49. http://www.reddit.com/r/haskell/comments/1npram/job_opportunity_for_junior_… 50. http://www.reddit.com/r/haskell/comments/1npram/job_opportunity_for_junior_… 51. http://www.youtube.com/watch?v=mpby_hz8lFc 52. http://www.reddit.com/r/haskell/comments/1n3f8s/short_vlog_style_video_my_r… 53. http://blog.ezyang.com/2013/09/if-youre-using-lift-youre-doing-it-wrong-pro… 54. http://www.reddit.com/r/haskell/comments/1n7mid/arguing_against_monad_trans… 55. http://www.youtube.com/watch?v=7lZ-DcidyMM 56. http://www.reddit.com/r/haskell/comments/1naslm/haskell_launchpad_interacti… 57. http://stackoverflow.com/questions/19177125/sets-functors-and-eq-confusion 58. http://stackoverflow.com/questions/19109152/writing-a-c-sharp-version-of-ha… 59. http://stackoverflow.com/questions/18989916/enforced-pattern-order 60. http://stackoverflow.com/questions/19019093/haskell-multiplying-int-and-flo… 61. http://stackoverflow.com/questions/18944370/what-requirements-does-non-stri… 62. http://stackoverflow.com/questions/18985353/random-access-on-a-huge-file-in… 63. http://stackoverflow.com/questions/19051668/not-using-a-variable-for 64. http://stackoverflow.com/questions/19064694/a-common-pattern-involving-comp… 65. http://stackoverflow.com/questions/19197339/haskell-data-constructor-meaning 66. http://stackoverflow.com/questions/19201901/in-functional-reactive-programm… 67. https://plus.google.com/105107667630152149014/about

1 0

Conditional lens
by Artyom Kazak 10 Oct '13

10 Oct '13

Hello! I am working with TypeReps, and while writing some functions I have noticed that I could use lenses to simplify them; however, I have stumbled upon some difficulties. First I’ll try to clarify which functions I want to write: * a function for converting TypeRep of, say, `Maybe x` to `[x]` (for all x). It requires checking if the TyCon is `Maybe` and replacing it with []-TyCon. If it wasn’t `Maybe`, I return Nothing. * a similar function for replacing `Char`s and `Lazy.Text`s to just `Text`. Again, if the TypeRep-to-be-replaced doesn’t satisfy my conditions, I return Nothing. These two functions (and some others, I suppose) can be written concisely with the help of one combinator. I don’t know how to write it as a composable Lens, so I’ll give it here as an ad-hoc Lens-modifying function instead: ifL :: (a -> Bool) -> Lens s t a b -> Lens s (Maybe t) (Maybe a) b ifL p l = lens getter setter where get s = getConst $ l Const s getter s = let a = get s in if p a then Just a else Nothing setter s b = let a = get s in if p a then Just (set l b s) else Nothing It works like this: > (0, 2) & ifL even fs .~ "hello" Just ("hello",2) > (1, 2) & ifL even fs .~ "hello" Nothing With `ifL`, my initial ugly changeTyCon :: TyCon -> TyCon -> TypeRep -> Maybe TypeRep changeTyCon tc tc' t | t^.tyCon == tc = Just $ t & tyCon .~ tc' | otherwise = Nothing boils down to changeTyCon tc tc' = ifL (== tc) tyCon .~ tc' Why did I call the initial version “ugly”? Well, because a) it manually handles `Maybe`s, and b) it has to perform both getting and setting (two passes). So, my questions are: 1. What would be the idiomatic way to write `ifL`? 2. How can I do something like `t ^. ifL (== tc) tyCon`? Currently it doesn’t work because view’s type has been simplified in lens-3.9. 3. Perhaps it would be better to represent `ifL` as a Traversal which simply ignores values that don’t match the condition? Then I could (?) use `failover` to do what I want. I searched for something filter-like in lens library, but haven’t found anything. 4. If I haven’t missed anything and it indeed can’t be done with bare lens, would `ifL` or something similar be welcome as an addition to the library? Thanks!

1 0

Using Quick Check generators for getting arbitrary value streams
by Luke Evans 10 Oct '13

10 Oct '13

I was hoping I could use Arbitrary instances to generate streams of values for test data. It looks like you're not 'supposed' to be trying this, other than for the specific purpose of then testing some properties on these streams within Quick Check itself. I'm looking for something like the sample' function in Quick Check, only to produce an infinite stream of values (rather than the 11 values that are clearly designed simply to give you a sense of the output of a generator). Am I out of luck, or is there yet some cunning way of doing this without hacking the Quick Check package to export the MkGen constructor? -- Luke

3 2

Suppressing HLint on pattern match of files
by Graham Berks 10 Oct '13

10 Oct '13

Hi, would like to disable 'Top-level binding with no type signature' In my test modules that are prefixed with Test. Is this possible ?? Thanks

2 2

hdbc-odbc not getting any data using mssql stored procedure
by grant 10 Oct '13

10 Oct '13

Hi, I am trying to call a stored procedure (exec gp_somestoredproc 123,22) using microsoft sql server 2008 R2 using hdbc-odbc. Here is the haskell code: import Database.HDBC import Database.HDBC.ODBC import Control.Monad import Data.Maybe import qualified Data.ByteString.Char8 as B test1 = do c <- connectODBC "Driver={SQL Server};Server=?;Database=?;uid=sa;pwd=?;" p <- prepare c "exec gp_somestoredproc 123,22" -- returns no data -- p <- prepare c "exec [sys].sp_datatype_info_100 0,@ODBCVer=4;exec gp_somestoredproc 123,22" -- all is good e <- execute p [] -- returns 6 (number of rows) putStrLn $ "execute " ++ show e r <- fetchAllRows' p putStrLn $ "fetchAllRows' " ++ show r The problem is that this code returns the number of rows correctly but doesn't return data nor are there any errors. However, I ran a perl program using perl dbi and got the data correctly. Here is the perl code: #!/usr/bin/perl use DBI; my $user = 'sa'; my $pw = '????'; my $dsn = '????'; my $dbh = DBI->connect("dbi:ODBC:$dsn", $user, $pw, {PrintError => 1, RaiseError => 1}); if (!$dbh) { print "error: connection: $DBI::err\n$DBI::errstr\n$DBI::state\n"; } my $type_info = $dbh->type_info(93); while(my($key, $value) = each(%$type_info)){ print "$key => $value\n"; }; my $sql = "exec gp_somestoredproc 123,22"; my $sth = $dbh->prepare($sql); my $r = $sth->execute; while (my ($db) = $sth->fetchrow_array) { print $db . "\n===\n"; } $dbh->disconnect if $dbh; I traced both versions and noticed that the perl dbi version first called exec [sys].sp_datatype_info_100 0,@ODBCVer=4 So I prefixed the stored proc call in haskell with "exec [sys].sp_datatype_info_100 0,@ODBCVer=4;" and it worked fine. In short: FAILS p <- prepare c "exec gp_somestoredproc 123,22" -- returns number of rows but no data WORKS p <- prepare c "exec [sys].sp_datatype_info_100 0,@ODBCVer=4;exec gp_somestoredproc 123,22" I have no idea why this works. sp_datatype_info_100 just dumps out the fields types ... More information: The stored procedure returns data with user defined field types. I have managed to do selects against tables with user defined field types without any problems using hdbc-odbc. I couldn't emulate this error on a local older version of mssql server (Microsoft SQL Server 2008 (SP1) - 10.0.2531.0 (X64) express) but the perl dbi prefixed the stored procedure call with "exec sp_datatype_info 0,@ODBCVer=3" I am running this against Microsoft SQL Server 2008 R2 (RTM) - 10.50.1797.0 (X64) I would appreciate any pointers you can give me. Thanks Grant

3 2

Call to Action - Fay FFI Bindings for AngularJS
by Adam Bergmark 10 Oct '13

10 Oct '13

I've talked to a lot of people that would like to be able to use Fay with AngularJS, this seems like a great idea! I don't use AngularJS personally, but I would be happy to help out with designing it and making code reviews. Has anyone started working on this already, and do you have any code/thoughts to share as a starting point? I created a stub repository under the faylang org on github: https://github.com/faylang/fay-angular I'll give anyone who wants to work on this push access, just send me an e-mail with your github account name, you can also create pull requests if you want a more in depth code review. Feel free to open up an issue if you have anything you would like to discuss. https://github.com/faylang/fay-angular/issues Regards, Adam

1 0

RankNTypes + ConstraintKinds to use Either as a "union"
by oleg＠okmij.org 10 Oct '13

10 Oct '13

Thiago Negri wrote: > Why type inference can't resolve this code? > {-# LANGUAGE RankNTypes, ConstraintKinds #-} > > bar :: (Num a, Num b) => (forall c. Num c => c -> c) ->Either a b ->Either a b > bar f (Left a) = Left (f a) > bar f (Right b) = Right (f b) > > bar' = bar (+ 2) -- This compiles ok > > foo :: (tc a, tc b) => (forall c. tc c => c -> c) -> Either a b -> Either a b > foo f (Left a) = Left (f a) > foo f (Right b) = Right (f b) > > foo' = foo (+ 2) -- This doesn't compile because foo' does not typecheck The type inference of the constraint fails because it is ambiguous. Observe that not only bar (+2) compiles OK, but also bar id. The function id :: c -> c has no constraints attached, but still fits for (forall c. Num c => c -> c). Let's look at the problematic foo'. What constraint would you think GHC should infer for tc? Num? Why not the composition of Num and Read, or Num and Show, or Num and any other set of constraints? Or perhaps Fractional (because Fractional implies Num)? For constraints, we get the implicit subtyping (a term well-typed with constraints C is also well-typed with any bigger constraint set C', or any constraint set C'' which implies C). Synonyms and superclass constraints break the principal types. So, inference is hopeless. We got to help the type inference and tell which constraint we want. For example, > newtype C ctx = C (forall c. ctx c => c -> c) > > foo :: (ctx a, ctx b) => C ctx -> (forall c. ctx c => c -> c) -> > Either a b -> Either a b > foo _ f (Left a) = Left (f a) > foo _ f (Right b) = Right (f b) > > foo' = foo (undefined :: C Num) (+2) Or, better > xyz :: (ctx a, ctx b) => C ctx -> Either a b -> Either a b > xyz (C f) (Left a) = Left (f a) > xyz (C f) (Right b) = Right (f b) > > xyz' = xyz ((C (+2)) :: C Num) > xyz'' = xyz ((C (+2)) :: C Fractional)

1 0