- Haskell-Cafe - Haskell.org

Proposal: (.:) operator in base.
by Alex Belanger 24 Aug '16

24 Aug '16

Hi, Some of you might be familiar with (.:) = (.) . (.). It has type :: (c -> d) -> (a -> b -> c) -> a -> b -> d It allows the composition of two functions, the first one, accepting one operand, and the second, two operands. This appears to be a very common pattern, referenced a bit everywhere, almost always defined on lambdabot and found in multiple codebases in the wild. I'd like the know the general sentiment about this operator, as well as how its inclusion in base, probably Data.Function, would be perceived before I actually try to make it happen. Cheers, Alex (nitrix).

7 8

Re: [Haskell-cafe] Proposal: (.:) operator in base.
by David Feuer 23 Aug '16

23 Aug '16

And also sometimes wrong. I don't really understand why, but sometimes eta *expansion* is important when working around RULES. On Aug 23, 2016 5:42 AM, "Alan & Kim Zimmerman" <alan.zimm(a)gmail.com> wrote: Suddenly the eta reduce hlint warning becomes much more important. Alan On 23 Aug 2016 11:22 a.m., "Tom Ellis" <tom-lists-haskell-cafe-2013@ jaguarpaw.co.uk> wrote: > *Very* interesting. Here is the key section of the GHC users guide: > > GHC will only inline the function if it is fully applied, where “fully > applied” means applied to as many arguments as appear (syntactically) > on > the LHS of the function definition > > That this is a good heuristic is *extremely* counterintuitive to me. I > would have supposed that being more explicit, for example > > {-# INLINE (.) f g #-} > > to inline on all static applications of at least two arguments would have > been a much clearer way to communicate this message. > > What's the rationale behind the current behaviour? > > Tom > > > On Tue, Aug 23, 2016 at 09:56:25AM +0100, Matthew Pickering wrote: > > I think the point which no-one has articulated yet is that the > > source-level arity of (.) affects whether GHC will decide to inline it. > > > > Only fully saturated applications are inlined > [...] > > > > On Tue, Aug 23, 2016 at 8:33 AM, Tom Ellis > > <tom-lists-haskell-cafe-2013(a)jaguarpaw.co.uk> wrote: > > > On the contrary, they're exactly the same (on GHC 7.6.3): > > > > > > module Foo where > > > > > > comp1 f g x = f (g x) > > > > > > comp2 f g = \x -> f (g x) > > > > > > > > > % ghc -O0 -dsuppress-all -fforce-recomp -no-link -ddump-prep > test.hs > > > [1 of 1] Compiling Foo ( test.hs, test.o ) > > > > > > ==================== CorePrep ==================== > > > Result size of CorePrep = {terms: 24, types: 36, coercions: 0} > > > > > > comp2 > > > comp2 = > > > \ @ t_aeU @ t1_aeV @ t2_aeW f_sfz g_sfy x_sfx -> > > > let { > > > sat_sfI > > > sat_sfI = g_sfy x_sfx } in > > > f_sfz sat_sfI > > > > > > comp1 > > > comp1 = > > > \ @ t_af5 @ t1_af6 @ t2_af7 f_sfG g_sfF x_sfE -> > > > let { > > > sat_sfJ > > > sat_sfJ = g_sfF x_sfE } in > > > f_sfG sat_sfJ > > > > > > (the same holds for -O2, if you compile them separately) > > > > > > On Tue, Aug 23, 2016 at 05:19:38PM +1000, Ben wrote: > > >> At the semantic level of "does my program compute correct results" > they're > > >> identical. At the operational level of "how fast does my program run" > > >> they're different. > > >> > > >> > > >> On August 23, 2016 5:09:19 PM GMT+10:00, Tom Ellis < > tom-lists-haskell-cafe-2013(a)jaguarpaw.co.uk> wrote: > > >> >On Mon, Aug 22, 2016 at 10:23:07PM -0700, wren romano wrote: > > >> >> (.) f g = \x -> f (g x) > > >> >> > > >> >> vs: > > >> >> > > >> >> (.) f g x = f (g x) > > >> >> > > >> >> has ramifications, though it's fairly easy to guess which one of > > >> >those > > >> >> two will be most performant. > > >> > > > >> >Are these not synonyms? What is the meaning of > > >> > > > >> > fargs var = expr > > >> > > > >> >if not > > >> > > > >> > fargs = \var -> expr > > >> > > > >> >? > _______________________________________________ > Haskell-Cafe mailing list > To (un)subscribe, modify options or view archives go to: > http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe > Only members subscribed via the mailman list are allowed to post. _______________________________________________ Haskell-Cafe mailing list To (un)subscribe, modify options or view archives go to: http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe Only members subscribed via the mailman list are allowed to post.

1 0

Re: [Haskell-cafe] Proposal: (.:) operator in base.
by Mike Ledger 23 Aug '16

23 Aug '16

(Forgot reply all) ---------- Forwarded message ---------- From: "Mike Ledger" <eleventynine(a)gmail.com> Date: 18 Aug 2016 7:52 AM Subject: Re: [Haskell-cafe] Proposal: (.:) operator in base. To: "Csongor Kiss" <kiss.csongor.kiss(a)gmail.com> Cc: IMO: I've never seen a compelling use of this; I think it decreases clarity, and barely shaves off any characters from equivalent "pointful" expressions. On Thu, Aug 18, 2016 at 7:45 AM, Csongor Kiss <kiss.csongor.kiss(a)gmail.com> wrote: > Of course I meant to write ((->) e) > > On 17 August 2016 at 22:39:36, Csongor Kiss (kiss.csongor.kiss(a)gmail.com) > wrote: > > Can you provide a bit more detail on why you think (fmap . fmap) captures > the concept behind (.) . (.) ? > It certainly does generalise it in one way - and I do get how, but is it > the *right* generalisation? > > Or does it just coincide with (-> e)’s fmap being defined as (.), and in > fact we would prefer to > generalise the composition aspect, in which case Category would be a more > ‘obvious’ place to look? > > On 17 August 2016 at 22:15:56, Tony Morris (tonymorris(a)gmail.com) wrote: > > You'd generalise it to: > > fmap . fmap :: (Functor f, Functor g) => (a -> b) -> f (g a) -> f (g b) > > And then, would you do the same for Traversable, Foldable and Applicative? > > On 18/08/16 03:43, Alex Belanger wrote: > > Hi, > > Some of you might be familiar with (.:) = (.) . (.). > > It has type :: (c -> d) -> (a -> b -> c) -> a -> b -> d > > It allows the composition of two functions, the first one, accepting one > operand, and the second, two operands. > > This appears to be a very common pattern, referenced a bit everywhere, > almost always defined on lambdabot and found in multiple codebases in the > wild. > > I'd like the know the general sentiment about this operator, as well as > how its inclusion in base, probably Data.Function, would be perceived > before I actually try to make it happen. > > Cheers, > Alex (nitrix). > > > _______________________________________________ > Haskell-Cafe mailing list > To (un)subscribe, modify options or view archives go to:http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe > Only members subscribed via the mailman list are allowed to post. > > > ------------------------------ > _______________________________________________ > Haskell-Cafe mailing list > To (un)subscribe, modify options or view archives go to: > http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe > Only members subscribed via the mailman list are allowed to post. > > > _______________________________________________ > Haskell-Cafe mailing list > To (un)subscribe, modify options or view archives go to: > http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe > Only members subscribed via the mailman list are allowed to post. >

12 13

library for set of heterogeneous values?
by Anthony Clayden 23 Aug '16

23 Aug '16

I have a (let's call it) database of heterogeneous records. They're not Haskell records, but anonymous/extensible type-labelled rows. (Could be tuples, could be HLists, could be Lens-like, could be something fancier.) There's a small number (dozens) of distinct row types, each with a large number (thousands) of rows. The variety of row-types is not predictable in advance. And indeed a row might 'morph' over time with fields added/removed. So the obvious answer of putting the lot into a giant HList (each element of the list being a row) isn't going to scale. I could have a type-indexed HList in which each element is a Set of homogeneous rows. But performance still suffers from scanning along the list to find the right type index. Is there something better? On hackage there's two packages called HSet, neither giving very much help about their suitability: * `hset` (lower case) [AlekseyUymanov] seems isomorphic to a type-indexed HList. ie Must be unique type in each element (could be a Set type, I guess) * `HSet` (upper case) [athanclark] "Faux heterogeneous sets" seems a lot meatier why the "Faux"? built over hashtables in the ST monad. Has anybody used these? Can give guidance on what they can and can't? Bonus questions: Given a filter specifying a restriction on (some) fields of rows, I want to get a heterog subset: * all rows with at least those fields, matching those restrictions. * the restriction might be merely "has field labelled L". GIven a candidate row for insertion, I want first to scan for quasi-duplicates: * any existing row with a subset of the given fields, and the same value at those fields. * any existing row with a superset of the given fields, and the same value at those fields in common. * ignore records with only a partial overlap of fields. One possible data structure: a "vertical store". Give each row a Globally Unique Id. Have a separate set for each possible field, where the set elements are field value (key) to set of GUId -- records with that value. Then I have a different bonus question: * how to retrieve all field values for a given GUId? Thanks AntC

3 5

Munich Haskell Meeting, 2016-08-25 @ 19:30
by Christian Neukirchen 22 Aug '16

22 Aug '16

Dear all, This week, our monthly Munich Haskell Meeting will take place again on Thursday, August 25 at Hirschgarten Biergarten at 19h30. For details see here: http://muenchen.haskell.bayern/dates.html If you plan to join, please add yourself to this dudle so we can reserve enough seats! It is OK to add yourself to the dudle anonymously or pseudonymously. https://dudle.inf.tu-dresden.de/haskell-munich-aug-2016/ Everybody is welcome! cu, -- Christian Neukirchen <chneukirchen(a)gmail.com> http://chneukirchen.org

1 0

Re: [Haskell-cafe] Using stringize and string concatenation in ghc preprocessing
by Harendra Kumar 22 Aug '16

22 Aug '16

Thanks Mathieu. This works pretty well for gcc ( https://gcc.gnu.org/onlinedocs/cpp/Traditional-macros.html) but sadly it does not work for clang cpp as Brandon too pointed out earlier that clang does not have a traditional mode. -harendra On 22 August 2016 at 02:01, Boespflug, Mathieu <m(a)tweag.io> wrote: > Hi Harendra, > > I ran into this very problem recently. Turns out -traditional knows string > concatenation too. I seem to remember learning this by browsing the GHC > source code, but now I can't find any occurrence of this pattern. But > here's an example of how to do string concatenation with CPP in > -traditional mode: https://github.com/tweag/sparkle/blob/ > a4e481aa5180b6ec93c219f827aefe932b66a953/inline-java/src/ > Foreign/JNI.hs#L274. > > HTH, > > -- > Mathieu Boespflug > Founder at http://tweag.io. > > On 20 August 2016 at 20:33, Brandon Allbery <allbery.b(a)gmail.com> wrote: > >> On Sat, Aug 20, 2016 at 2:27 PM, Harendra Kumar <harendra.kumar(a)gmail.com >> > wrote: >> >>> But "-optP" seems to only append to the flags that GHC already passes >>> and gcc has no "-no-traditional" option to undo the effect of the >>> "-traditional" that GHC has already passed. I think "-optP" should override >>> the flags passed by ghc rather than appending to them. Is there a reason >>> not to do that? >>> >>> Is there any other better way to achieve this? What is the standard way >>> of doing this if any? >>> >> >> Removing -traditional will break much Haskell source. Go look at the >> history of clang with ghc (clang doesn't do -traditional) to see what >> happens. (tl;dr: without -traditional, cpp knows too much about what >> constitutes valid C, and mangles and/or throws errors on valid Haskell that >> doesn't lex the way C does.) >> >> You might want to look at cpphs as an alternative preprocessor. There are >> some ancient K&R-era hacks that could be used if absolutely necessary, but >> cpphs should be a much simpler and cleaner solution. >> >> -- >> brandon s allbery kf8nh sine nomine >> associates >> allbery.b(a)gmail.com >> ballbery(a)sinenomine.net >> unix, openafs, kerberos, infrastructure, xmonad >> http://sinenomine.net >> >> _______________________________________________ >> ghc-devs mailing list >> ghc-devs(a)haskell.org >> http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs >> >> >

1 0

Rewrite without parenthesis
by Carl Petersen 22 Aug '16

22 Aug '16

How do you rewrite head (tail [1,2,3]) without parenthesis. Thanks!

8 10

Re: [Haskell-cafe] Using stringize and string concatenation in ghc preprocessing
by Brandon Allbery 22 Aug '16

22 Aug '16

On Sun, Aug 21, 2016 at 4:31 PM, Boespflug, Mathieu <m(a)tweag.io> wrote: > I ran into this very problem recently. Turns out -traditional knows string > concatenation too. I seem to remember learning this by browsing the GHC > source code, but now I can't find any occurrence of this pattern. But > here's an example of how to do string concatenation with CPP in > -traditional mode: https://github.com/tweag/sparkle/blob/ > a4e481aa5180b6ec93c219f827aefe932b66a953/inline-java/src/ > Foreign/JNI.hs#L274 > > . > That's the hacky K&R way I mentioned earlier. -- brandon s allbery kf8nh sine nomine associates allbery.b(a)gmail.com ballbery(a)sinenomine.net unix, openafs, kerberos, infrastructure, xmonad http://sinenomine.net

1 0

How can I generalize my innerJoinOnId function to innerJoin with Vinyl?
by Cody Goodman 22 Aug '16

22 Aug '16

Hello all! I'll get right to it. As a first step I'd like to generalize innerJoinOnId's type signature to something like: innerJoinOnId :: (Id ∈ fields, Id ∈ fields2, Id ∈ fields2) => [Rec Attr fields] -> [Rec Attr fields2] -> [Rec Attr fields3] Or if possible: innerJoinOnId :: (Id ∈ fields) => [Rec Attr fields] -> [Rec Attr fields] -> [Rec Attr fields] As a next step I'd like to create an innerJoin function with type: innerJoinOn :: (a ∈ fields) => [Rec Attr fields] -> [Rec Attr fields] -> [Rec Attr fields] Where a is supplied and that constraint is carried on to the other inputs. Is this possible in Haskell? Here is both a link and the text of my (working, compilable) code thus far. Link: https://github.com/codygman/vinyl-experiments/blob/master/src/Main.hs Source code: {-# LANGUAGE DataKinds, PolyKinds, TypeOperators, TypeFamilies, FlexibleContexts, FlexibleInstances, NoMonomorphismRestriction, GADTs, TypeSynonymInstances, TemplateHaskell, StandaloneDeriving #-} module Main where import Data.Vinyl import Control.Lens hiding (Identity) import Data.Singletons.TH import Data.Maybe import Control.Monad import Data.Vinyl.TypeLevel (RIndex) data Fields = Id | Name | Age | ActivityName deriving Show type Person = ['Id, 'Name, 'Age] type Activity = ['Id, 'ActivityName] type family ElF (f :: Fields) :: * where ElF 'Id = Int ElF 'Name = String ElF 'Age = Int ElF 'ActivityName = String newtype Attr f = Attr { _unAttr :: ElF f } makeLenses ''Attr genSingletons [ ''Fields ] instance Show (Attr 'Id) where show (Attr x) = "id: " ++ show x instance Show (Attr 'Name) where show (Attr x) = "name: " ++ show x instance Show (Attr 'Age) where show (Attr x) = "age: " ++ show x instance Show (Attr 'ActivityName) where show (Attr x) = "activity: " ++ x (=::) :: sing f -> ElF f -> Attr f _ =:: x = Attr x joy :: Rec Attr ['Id, 'Name, 'Age] joy = (SId =:: 1) :& (SName =:: "Joy") :& (SAge =:: 28) :& RNil jon :: Rec Attr ['Id, 'Name, 'Age] jon = (SId =:: 0) :& (SName =:: "Jon") :& (SAge =:: 23) :& RNil karen :: Rec Attr ['Id, 'Name, 'Age] karen = (SId =:: 2) :& (SName =:: "Karen") :& (SAge =:: 15) :& RNil jonFootball :: Rec Attr ['Id, 'ActivityName] jonFootball = (SId =:: 0) :& (SActivityName =:: "football") :& RNil jonDancing :: Rec Attr ['Id, 'ActivityName] jonDancing = (SId =:: 0) :& (SActivityName =:: "dancing") :& RNil joyRacing :: Rec Attr ['Id, 'ActivityName] joyRacing = (SId =:: 1) :& (SActivityName =:: "racing") :& RNil peopleRows :: [Rec Attr ['Id, 'Name, 'Age]] peopleRows = [joy, jon, karen] activitieRows :: [Rec Attr ['Id, 'ActivityName]] activitieRows = [jonFootball, jonDancing, joyRacing] printActvy :: ('ActivityName ∈ fields) => Rec Attr fields -> IO () printActvy r = print (r ^. rlens SActivityName) isInPplIdx :: ('Id ∈ fields) => [Int] -> Rec Attr fields -> Bool isInPplIdx peopleIdx actvyRow = any (== True) . map (== actvyIdInt) $ peopleIdx where actvyIdInt = actvyRow ^. rlens SId . unAttr mkJoinedRow :: (Eq (ElF r1), RElem r1 ['Id, 'Name, 'Age] (RIndex r1 ['Id, 'Name, 'Age]), RElem r1 ['Id, 'ActivityName] (RIndex r1 ['Id, 'ActivityName]), ElF r1 ~ Int) => sing1 r1 -> [Rec Attr ['Id, 'ActivityName]] -> Rec Attr ['Id, 'Name, 'Age] -> [Rec Attr ['Id, 'Name, 'Age, 'ActivityName]] -- mkJoinedRow :: _ -> [Rec Attr ['Id, 'ActivityName]] -> Rec Attr ['Id, 'Name, 'Age] -> [Rec Attr ['Id, 'Name, 'Age, 'ActivityName]] mkJoinedRow field activities person = do let name = person ^. rlens SName . unAttr age = person ^. rlens SAge . unAttr let filteredActivities = filter (\r -> r ^. rlens field . unAttr == person ^. rlens field . unAttr) activities case listToMaybe filteredActivities of Just _ -> do let activityId actvy = actvy ^. rlens field . unAttr activityName actvy = actvy ^. rlens SActivityName . unAttr (\actvy -> (SId =:: activityId actvy) :& (SName =:: name) :& (SAge =:: age) :& (SActivityName =:: activityName actvy) :& RNil) <$> filteredActivities Nothing -> [] innerJoinOnId :: [Rec Attr ['Id, 'Name, 'Age]] -> [Rec Attr ['Id, 'ActivityName]] -> [Rec Attr ['Id, 'Name, 'Age, 'ActivityName]] innerJoinOnId people activities = do let peopleIdx =(\r -> r ^. rlens SId . unAttr) <$> people let filteredActivites = filter (isInPplIdx peopleIdx) activities join $ map (\p -> mkJoinedRow SId filteredActivites p) people main :: IO () main = mapM_ print $ innerJoinOnId peopleRows activitieRows -- example of main running: -- λ> peopleRows -- [{id: 1, name: "Joy", age: 28},{id: 0, name: "Jon", age: 23},{id: 2, name: "Karen", age: 15}] -- λ> activitieRows -- [{id: 0, activity: football},{id: 0, activity: dancing},{id: 1, activity: racing}] -- λ> main -- {id: 1, name: "Joy", age: 28, activity: racing} -- {id: 0, name: "Jon", age: 23, activity: football} -- {id: 0, name: "Jon", age: 23, activity: dancing} -- Code I wish worked: -- λ> mapM_ print $ innerJoinOn SId peopleRows activitieRows -- λ> mapM_ print $ innerJoinOn SName peopleRows activitieRows -- this line would give a compiler error about activitiesRows not containing 'Name

1 1

Build error with llvm-general and lens
by HMPerson1 21 Aug '16

21 Aug '16

If I have a file > module Lib where > import Control.Lens > data Foo = Foo { _x :: Int , _y :: Int } > makeLenses ''Foo and I compile with `stack build`, it works fine. But if I add `llvm-general` to build-depends, and I run `stack build`, I get > [1 of 1] Compiling Lib ( Lib.hs, .stack-work/dist/x86_64-linux/Cabal-1.22.5.0/build/Lib.o ) > <command line>: can't load .so/.DLL for: /usr/lib/libcurses.so (-lncursesw: cannot open shared object file: No such file or directory) > > -- While building package bug-test-0.1.0.0 using: > /home/user/.stack/setup-exe-cache/x86_64-linux/setup-Simple-Cabal-1.22.5.0-ghc-7.10.3 --builddir=.stack-work/dist/x86_64-linux/Cabal-1.22.5.0 build lib:bug-test --ghc-options " -ddump-hi -ddump-to-file" > Process exited with code: ExitFailure 1 I've made a tiny test case here: https://git.io/v6MVv If I symlink `/usr/lub/libcurses.so` to `/usr/lub/libncursesw.so` (as suggested at https://git.io/v6MV8 and https://ghc.haskell.org/trac/ghc/ticket/9237) I get > [1 of 1] Compiling Lib ( Lib.hs, .stack-work/dist/x86_64-linux/Cabal-1.22.5.0/build/Lib.o ) > > -- While building package bug-test-0.1.0.0 using: > /home/user/.stack/setup-exe-cache/x86_64-linux/setup-Simple-Cabal-1.22.5.0-ghc-7.10.3 --builddir=.stack-work/dist/x86_64-linux/Cabal-1.22.5.0 build lib:bug-test --ghc-options " -ddump-hi -ddump-to-file" > Process exited with code: ExitFailure (-11) (There's no "can't load..." line and the exit code is different) As a workaround, I can separate the lens instances into a separate library, but I would like to avoid doing so if possible. How can I fix this?

2 2