- Haskell-Cafe - Haskell.org

ANN: qtah-0.1.0
by Bryan Gardiner 11 Jul '16

11 Jul '16

Hello all, I'm happy to announce the first Hackage-ready release of Qtah, Qt bindings for Haskell. I announced Qtah earlier this year, but a lot of work was needed to get it to the point where you can just "cabal install" it. That work has now been done, so I invite you to install qtah-qt5 or qtah-examples[1] (currently just the Qt notepad example), and check out: http://khumba.net/projects/qtah https://gitlab.com/khumba/qtah As a small demonstration, I have a good portion of Goatee ported (board rendering clearly missing): http://khumba.net/tmp/20160710-goatee-qt.png However, there is still lots of the Qt API left to cover. Thanks to ezyang for help working with some Cabal internals. Cheers, Bryan [1] cabal install --enable-executable-dynamic qtah-examples

1 0

Practical use of Stream's monad instance?
by Petr Pudlák 11 Jul '16

11 Jul '16

Hi, we've been debating the usefulness of Monad instances for Stream [2] or homogeneous tuples (fixed-length vectors). The Applicative implementation for them is simply zipping, very useful indeed. The "join" of their Monad instances takes the diagonal of a 2-dimensional plane (finite for tuples, infinite for Stream). (See also [1].) The question is, are such monad instances used in the wild? Are they actually useful or interesting for anything? [1] https://mail.haskell.org/pipermail/haskell-cafe/2009-April/thread.html#59079 [2] https://hackage.haskell.org/package/Stream-0.4.7.2/docs/Data-Stream.html Thanks Petr

2 1

What's wrong in my Data.Generic.Zipper code?
by martin 11 Jul '16

11 Jul '16

Hello all, I do not understand why this simple piece of code returns Nothing instead of [1,2,3]. Can anybody help? {-# LANGUAGE BangPatterns,NoMonomorphismRestriction, DeriveDataTypeable #-} import Data.Typeable import Data.Data import Data.Generics.Zipper ex_list1 :: [Int] ex_list1 = [1,2,3] t1 = toZipper ex_list1 -- *Main> getHole t1 -- Nothing

2 3

Re: [Haskell-cafe] Commutative monads vs Applicative functors
by Olaf Klinke 11 Jul '16

11 Jul '16

Anders Kock has a nice paper on commutative monads [1]. He relates commutativity to the Fubini theorem, because that is what commutativity is for the monad of meaures: You can change the order in integration by parts. He works in a symmetric monoidal closed category, so Hask works, too. It is the sort of category theory paper where you can go along the paper and implement most of it in Haskell. In particular, he mentions that MonadPlus has an alternative set of laws: A monad m is a MonadPlus when there are isomorphisms iso0 :: () -> m Void iso2 :: (m a,m b) -> m (Either a b) Indeed, let mzero = liftM absurd (iso0 ()) mplus x y = liftM codiagonal (iso2 (x,y)) where codiagonal (Left x) = x codiagonal (Right y) = y Conversely, let iso0 () = mzero iso2 (x,y) = (liftM Left x) `mplus` (liftM Right y) Then Kock shows that the MonadPlus operations are automatically Eilenberg-Moore algebra homomorphisms, which is the abstract way of saying that addition distributes over multiplication. On a related note, I wonder whether the following is true: Let m be a monad and let r be an object which is an m-algebra. Think of m as the free semigroup monad. Now consider the continuation monad type T x = (x -> r) -> r Observe that (x -> r) inherits the m-algebra structure of r. Let T' x be the subset of maps in T x which are m-linear. Is T' a monad, too? Can this me made concrete in a Haskell function? Regards, Olaf [1] @article{kock12, author="Kock, Anders", title="Commutative monads as a theory of distributions", journal="Theory and Applications of Categories", volume=26, number=4, pages="97-131", year=2012, url="http://www.tac.mta.ca/tac/volumes/26/4/26-04.pdf" }

1 0

Links on https://hackage.haskell.org/package/Cabal-1.24.0.0/docs/Language-Haskell-Extension.html are broken. How can I help?
by Michael Litchard 11 Jul '16

11 Jul '16

Links to documentation are broken. I would be happy to fix. But how? https://hackage.haskell.org/package/Cabal-1.24.0.0/docs/Language-Haskell-Ex…

2 1

Re: [Haskell-cafe] Strange error with cabal install on windows:
by Anton Kholomiov 10 Jul '16

10 Jul '16

For those who wants to help. Pavel discovered the cause of the problem for me. I've got too many modules in my package and it causes the error. The length of the command line strings is limited for windows so if you have too many modules the final build command can not list them all. I need to regroup my modules I guess.. It's solved! 8 июля 2016 г., 20:24 пользователь Anton Kholomiov < anton.kholomiov(a)gmail.com> написал: > Ноо тогда он их не включит в пакет. Не сгенерит объектные файлы для этих > модулей. Я думаю, что мне надо перегруппировать модули. Чтобы их стало > поменьше. Но не слишком сильно, чтобы ghc смог скомпилировать файлы. Если > файл слишком большой он будет крайне долго компилировать. > > 8 июля 2016 г., 13:07 пользователь Pavel Troev <pavel.troev(a)gmail.com> > написал: > > ну да.. у меня давно зуб на виндовую консоль по поводу utf-8 >> >> похоже надо подправить .cabal файл - >> убрать все модули из секции Library / Other-Modules: >> >> >> >> C:\sap\sharc_timbre\sharc-timbre-0.1>cabal install >> Warning: The package list for 'hackage.haskell.org' is 81.2 days old. >> Run 'cabal update' to get the latest list of available packages. >> Resolving dependencies... >> Configuring sharc-timbre-0.1... >> Building sharc-timbre-0.1... >> Preprocessing library sharc-timbre-0.1... >> In-place registering sharc-timbre-0.1... >> Creating package registration file: >> C:\Users\root\AppData\Local\Temp\pkgConf-sharc-timbre-048275436.1 >> Installing library in >> >> C:\Users\root\AppData\Roaming\cabal\x86_64-windows-ghc-7.10.3\sharc-timbre-0.1-Ig1DfnlqQG6G3PN7S26idD >> Registering sharc-timbre-0.1... >> Installed sharc-timbre-0.1 >> >> >> [image: Встроенное изображение 1] >> >> >> 8 июля 2016 г., 12:02 пользователь Anton Kholomiov < >> anton.kholomiov(a)gmail.com> написал: >> >> Похоже слишком много модулей, я этот пакет кодогенерил из БД, стоит >>> пересмотреть структуру пакета, >>> спасибо, теперь я знаю причину.. я что-то слышал о том, что в windows >>> есть лимит на длину строк в командной строке, и мой пакет упёрся в это >>> ограничение >>> >>> 8 июля 2016 г., 12:51 пользователь Pavel Troev <pavel.troev(a)gmail.com> >>> написал: >>> >>>> cabal get sharc-timbre >>>> cabal cd sharc-timbre-0.1 >>>> cabal configure >>>> cabal build -v >>>> видим строку ghc --make .... >>>> удаляем где-то половину нот до "Sharc.Instruments.ViolinMuted.Note15" >>>> начал компилировать, похоже дело именно в длине команды >>>> [ 618 of 1379] Compiling Sharc.Instruments.Cell >>>> arc\Instruments\CelloPizzicato\Note25.o ) >>>> [ 619 of 1379] Compiling Sharc.Instruments.Cell >>>> arc\Instruments\CelloPizzicato\Note26.o ) >>>> ... >>>> >>>> >>>> 8 июля 2016 г., 11:40 пользователь Anton Kholomiov < >>>> anton.kholomiov(a)gmail.com> написал: >>>> >>>>> я пока ещё не изучал вопрос, но вроде работает и там очень много >>>>> обновлений, но судя по тому что они писали в HCAR могут быть сюрпризы, они >>>>> планируют следующие релизы посвятить стабилизации ghc. >>>>> >>>>> >>>>> 2016-07-08 12:27 GMT+04:00 Pavel Troev <pavel.troev(a)gmail.com>: >>>>> >>>>>> а стоит на ghc 8 переходить? >>>>>> >>>>>> 2016-07-08 11:24 GMT+03:00 Anton Kholomiov <anton.kholomiov(a)gmail.com >>>>>> >: >>>>>> >>>>>>> Спасибо, то есть это от версии может зависеть.. >>>>>>> >>>>>>> У меня ghc-8.0.1. со стэком у меня не сложилось под виндой, но это >>>>>>> совсем другая история... >>>>>>> >>>>>>> 2016-07-08 12:19 GMT+04:00 Pavel Troev <pavel.troev(a)gmail.com>: >>>>>>> >>>>>>>> было похожее - пакет не собирался на win ghc 7.10.2 и спокойно >>>>>>>> собрался на 7.10.3, >>>>>>>> если используете stack, то можно просто поменять lts на 7.10.3 >>>>>>>> >>>>>>>> 2016-07-08 11:06 GMT+03:00 Anton Kholomiov < >>>>>>>> anton.kholomiov(a)gmail.com>: >>>>>>>> >>>>>>>>> I've got an error that I don't know how to fix with cabal install. >>>>>>>>> Has anyone experienced such a thing? It installs fine on ubuntu but >>>>>>>>> fails to install on windows 7. It's my own package called >>>>>>>>> `sharc-timbre`: >>>>>>>>> >>>>>>>>> ~~~ >>>>>>>>> Building sharc-timbre-0.1... >>>>>>>>> Preprocessing library sharc-timbre-0.1... >>>>>>>>> C:\Program Files\Haskell Platform\8.0.1\bin\ghc.exe: >>>>>>>>> createProcess: does not >>>>>>>>> exist (No such file or directory) >>>>>>>>> cabal: Leaving directory >>>>>>>>> 'C:\Users\антон\AppData\Local\Temp\cabal-tmp-4484\sharc-timbre-0.1' >>>>>>>>> ~~~ >>>>>>>>> >>>>>>>>> Thanks, >>>>>>>>> Anton >>>>>>>>> >>>>>>>>> _______________________________________________ >>>>>>>>> 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. >>>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> -- >>>>>>>> С уважением, >>>>>>>> >>>>>>>> Павел Троев >>>>>>>> Консультант БерингПойнт >>>>>>>> Проект SOL, Газпром-нефть >>>>>>>> Моб. тел: +7 916 441 48 61 >>>>>>>> pavel.troev(a)gmail.com | troev.pd(a)gazprom-neft.ru >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>> >>>>>> >>>>>> >>>>>> -- >>>>>> С уважением, >>>>>> >>>>>> Павел Троев >>>>>> Консультант БерингПойнт >>>>>> Проект SOL, Газпром-нефть >>>>>> Моб. тел: +7 916 441 48 61 >>>>>> pavel.troev(a)gmail.com | troev.pd(a)gazprom-neft.ru >>>>>> >>>>>> >>>>>> >>>>> >>>> >>>> >>>> -- >>>> С уважением, >>>> >>>> Павел Троев >>>> Консультант БерингПойнт >>>> Проект SOL, Газпром-нефть >>>> Моб. тел: +7 916 441 48 61 >>>> pavel.troev(a)gmail.com | troev.pd(a)gazprom-neft.ru >>>> >>>> >>>> >>> >> >> >> -- >> С уважением, >> >> Павел Троев >> Консультант БерингПойнт >> Проект SOL, Газпром-нефть >> Моб. тел: +7 916 441 48 61 >> pavel.troev(a)gmail.com | troev.pd(a)gazprom-neft.ru >> >> >> >

3 2

Natural keys in Haskell data structures
by Lian Hung Hon 10 Jul '16

10 Jul '16

Dear cafe, What is the idiomatic way to "split" records into their natural keys and content in a data structure? For example, given a user: data User = { username :: ByteString, hash :: ByteString, address :: Text, ... } Using map, a first choice would be Map ByteString User, but this leads to duplication of the username. And it is possible to make mistakes, such as insert "John" (User "Jane" ... What does cafe think? Is there any pattern for this? This is probably just a small nit in the overall architecture, but I'm curious to know the clean way to do it. Regards, Hon

3 3

Re: [Haskell-cafe] Haskell Data Structure design
by Will Yager 10 Jul '16

10 Jul '16

I believe you could use foldM (\sum student -> (sum +) <$> totalFeesOwed student) 0 students Will > On Jul 9, 2016, at 15:09, Guru Devanla <gurudev.devanla(a)gmail.com> wrote: > > OK. Thank you. That is what I ended up doing after I understood how mapM worked. But, since I was going through 2 steps, I was wondering if using foldM directly was possible in this case. > > >> On Sat, Jul 9, 2016 at 11:25 AM, Will Yager <will.yager(a)gmail.com> wrote: >> fees <- mapM totalFeesOwed students >> let total = sum fees >> >> You can use a fold instead of sum if you want. >> >> Will >> >>> On Jul 9, 2016, at 13:10, Guru Devanla <gurudev.devanla(a)gmail.com> wrote: >>> >>> Say, in the above example, I want to add up values returned by `student_totalFeesOwed` by using foldM operation. Is it possible? >>> >>> For example, here is an expression I have >>> >>> L.foldr (\a b-> (evalState (student_totalFeesOwed a) $ env) + b) 0 [(RowId 1), (RowId 2)] >>> >>>> On Sat, Jul 9, 2016 at 9:15 AM, Will Yager <will.yager(a)gmail.com> wrote: >>>> I did the same thing when I was learning to generalize my understanding of monads! Very common mistake. >>>> >>>> I'm not sure I understand your question about #3. Can you give an example using evalState? We'll tell you if you can do it without evalState. >>>> >>>> I suspect you want something like >>>> >>>> "mapM_ addStudentFee students" >>>> >>>> Will >>>> >>>>> On Jul 9, 2016, at 00:56, Guru Devanla <gurudev.devanla(a)gmail.com> wrote: >>>>> >>>>> William/Tom, >>>>> >>>>> (1) Yes, looking into lens and re-factoring my current experimental project in lens will be my next iteration. For now, I plan not to spend time on it. >>>>> >>>>> (2) Agreed. Not sure how I missed that. >>>>> >>>>> (3) I see how foldM works now. I missed the point that foldM not only is a `map` but also does a `sequence` after that. I got stuck earlier, thinking I will end up with a list of state monads. The sequence steps executes this monadic action. >>>>> >>>>> But, how can I do a foldM in a state monad. Say, I need to map over a list of students and add up all their fees, can I get away not `evalState` inside the foldM step function? >>>>> >>>>> Thanks. this is very exciting as I keep simplifying my code! >>>>> >>>>> Guru >>>>> >>>>> >>>>> >>>>> >>>>>> On Fri, Jul 8, 2016 at 7:55 PM, <amindfv(a)gmail.com> wrote: >>>>>> >>>>>>> On Fri, Jul 8, 2016 at 9:57 PM, Guru Devanla <gurudev.devanla(a)gmail.com> wrote: >>>>>> >>>>>> >>>>>>>> 1. I see that almost in every function I deal with state, I have e <- get , expression in the begining. I always ending up having to use the state to query for different values. I guess this is OK. >>>>>> >>>>>> >>>>>> El 8 jul 2016, a las 22:07, William Yager <will.yager(a)gmail.com> escribió: >>>>>> >>>>>>> For #1, look into using the Lens library's support for the State monad. You can often avoid doing a get, and instead write things like `fees += 5`, which will add 5 to the field in the state called "fees". >>>>>> >>>>>> Lens is a pretty heavy extra thing for a beginner to have to learn -- you'll do fine with the 'modify' function: >>>>>> >>>>>> modify :: (s -> s) -> State s () >>>>>> >>>>>> So instead of writing: >>>>>> >>>>>> do >>>>>> s <- get >>>>>> put (s + 5) >>>>>> >>>>>> You say: >>>>>> >>>>>> modify (+5) >>>>>> >>>>>> >>>>>> Tom >

2 1

Commutative monads vs Applicative functors
by Ronald Guida 10 Jul '16

10 Jul '16

I have a few questions about commutative monads and applicative functors. >From what I have read about applicative functors, they are weaker than monads because with a monad, I can use the results of a computation to select between alternative future computations and their side effects, whereas with an applicative functor, I can only select between the results of computations, while the structure of those computations and their side effects are fixed in advance. But then there are commutative monads. I'm not exactly sure what a commutative monad is, but my understanding is that in a commutative monad the order of side effects does not matter. This leads me to wonder, are commutative monads still stronger than applicative functors, or are they equivalent? And by the way, what exactly is a commutative monad?

6 6

Haskell Data Structure design
by Guru Devanla 10 Jul '16

10 Jul '16

Hello All, I am just getting myself to code in Haskell and would like to design advice. Below, I have a made up example: data ClassRoom = ClassRoom { classRoomNo:: Integer, extra_fees::Float, students: Map StudentId Student} data Student = Student {name::String, feesOwed::Float} data StudentId = Integer get_fees_owed classroom student_id = extra_fees + feesOwed $ (students classroom) M.! studentid Here the `get_fees_owed` needs information from the container 'classroom'. Here is my question/problem: I believe I should model most of my code as expressions, rather than storing pre-computed values such as `fees_owed`. But, defining expressions involve passing the container objects all over. For example, deep down in a function that deals with just one `student`, I might need the fees owed information. Without, having a reference to the container, I cannot call get_fees_owed. Also, I think it hinders composing of functions that just deal with one student at a time, but end up with some dependency on the container. I have several questions related to this design hurdle, but I will start with the one above. Thanks! Guru

5 11