- Haskell-Cafe - Haskell.org

Question on Coerce
by Brent Walker 02 Dec '20

02 Dec '20

In the following code, function fmap does not compile because variable 'y' on line marked <***> has type (Expr a) where an (Expr b) is expected. The code can be fixed simply by returning (Val x) on the rhs of the function but then we are allocating a new object for something we could potentially reuse since (Val n) has the same runtime representation in (Expr a) and (Expr b) (it has no dependence on the type variable). I was reading about coerce (Data.Coerce) recently and thought this could be a place it could be used but simply replacing the 'y' on the rhs with 'coerce y' does not compile (error message below). Is it possible to use coerce in this context? Thanks for any help, Brent ======================================= data Expr a = Var a | Val Int | Add (Expr a) (Expr a) deriving Show instance Functor Expr where fmap :: (a -> b) -> Expr a -> Expr b fmap f (Var a) = Var (f a) fmap f (Add e0 e1) = Add (fmap f e0) (fmap f e1) fmap _ y@(Val x) = y -- <***> ======================================= /dev/proj/src/Ex12.hs:69:22: error: • Couldn't match representation of type ‘a’ with that of ‘b’ arising from a use of ‘coerce’ ‘a’ is a rigid type variable bound by the type signature for: fmap :: forall a b. (a -> b) -> Expr a -> Expr b at /dev/proj/src/Ex12.hs:66:11-38 ‘b’ is a rigid type variable bound by the type signature for: fmap :: forall a b. (a -> b) -> Expr a -> Expr b at /dev/proj/src/Ex12.hs:66:11-38 • In the expression: coerce y In an equation for ‘fmap’: fmap _ y@(Val _) = coerce y In the instance declaration for ‘Functor Expr’ • Relevant bindings include y :: Expr a (bound at /dev/proj/src/Ex12.hs:69:10) fmap :: (a -> b) -> Expr a -> Expr b (bound at /dev/proj/src/Ex12.hs:67:3) | 69 | fmap _ y@(Val _) = coerce y | ^^^^^^^^ =======================================

7 11

Textbook example of instance Foldable ((,) a)
by Johannes Waldmann 30 Nov '20

30 Nov '20

Dear Cafe - you might find this entertaining: an error in my code that cost me some hours (most Pictures are equal?) until I finally realized instance Eq a => Eq (Picture a) where p == q = dim p == dim q - && all (\ i -> contents p i == contents q i) (bounds p) + && all (\ i -> contents p i == contents q i) (A.range $ bounds p) Henning already called this "Matlab league" behaviour, https://mail.haskell.org/pipermail/libraries/2016-February/026678.html Well it's just a fun observation now, I am not implying anything. Enjoy! - J. Link to full source: https://gitlab.imn.htwk-leipzig.de/autotool/all0/-/issues/660#note_24567

14 21

Amazonka, conduit and sockets not closing
by Magnus Therning 29 Nov '20

29 Nov '20

I've run into a problem with running out of filedescriptors. The following snippet is a trimmed down version of what I'm doing: #+begin_src haskell main :: IO () main = do awsEnv <- newEnv Discover runAWSCond awsEnv $ sqsSource queueUrl .| C.mapC snd .| sqsDeleteSink queueUrl where runAWSCond awsEnv = runResourceT . runAWS awsEnv . within Frankfurt . C.runConduit sqsSource :: MonadAWS m => T.Text -> C.ConduitT () (T.Text, T.Text) m () sqsSource queueUrl = do (_, msgs) <- C.lift $ recvSQS queueUrl C.yieldMany msgs sqsSource queueUrl sqsDeleteSink :: MonadAWS m => T.Text -> C.ConduitT T.Text o m () sqsDeleteSink queueUrl = do C.await >>= \case Nothing -> pure () Just receiptHandle -> do void $ C.lift $ delSQS queueUrl receiptHandle sqsDeleteSink queueUrl recvSQS queueUrl = do let rm = receiveMessage queueUrl & rmMaxNumberOfMessages ?~ 10 rmrs <- send rm let status = rmrs ^. rmrsResponseStatus msgs = rmrs ^. rmrsMessages & traversed %~ extract pure (status, catMaybes msgs) where extract msg = do body <- msg ^. mBody rh <- msg ^. mReceiptHandle pure (body, rh) delSQS queueUrl receiptHandle = do let dm = deleteMessage queueUrl receiptHandle send dm #+end_src This works fine for a while, but given a queue with enough messages it will fail with something like #+begin_example TransportError (HttpExceptionRequest Request { host = "sqs.eu-central-1.amazonaws.com" port = 443 secure = True requestHeaders = [("Host","sqs.eu-central-1.amazonaws.com"),("X-Amz-Date","20201126T101659Z"),("X-Amz-Content-SHA256","2e4bdf20a857a1416f218b1218670cf019ff53268d0adb34fe06402a62f3271d"),("Content-Type","application/x-www-form-urlencoded; charset=utf-8"),("Authorization","<REDACTED>")] path = "/" queryString = "" method = "POST" proxy = Nothing rawBody = False redirectCount = 0 responseTimeout = ResponseTimeoutMicro 70000000 requestVersion = HTTP/1.1 } (ConnectionFailure Network.Socket.getAddrInfo (called with preferred socket type/protocol: AddrInfo {addrFlags = [AI_ADDRCONFIG], addrFamily = AF_UNSPEC, addrSocketType = Stream, addrProtocol = 0, addrAddress = <assumed to be undefined>, addrCanonName = <assumed to be undefined>}, host name: Just "sqs.eu-central-1.amazonaws.com", service name: Just "443"): does not exist (System error))) #+end_example After some detours I found out that it's actually not a network issue, but rather that the process runs out of filedescriptors. Using =lsof= I can see that it doesn't seem to close /any/ sockets at all, instead they get stuck in a =CLOSE_WAIT= state: #+begin_example COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME wd-stats 88674 magnus 23u IPv4 815196 0t0 TCP ip-192-168-0-9.eu-central-1.compute.internal:60624->52.119.188.213:https (CLOSE_WAIT) wd-stats 88674 magnus 24u IPv4 811362 0t0 TCP ip-192-168-0-9.eu-central-1.compute.internal:43482->52.119.189.184:https (CLOSE_WAIT) wd-stats 88674 magnus 25u IPv4 811386 0t0 TCP ip-192-168-0-9.eu-central-1.compute.internal:60628->52.119.188.213:https (CLOSE_WAIT) wd-stats 88674 magnus 26u IPv4 813527 0t0 TCP ip-192-168-0-9.eu-central-1.compute.internal:43486->52.119.189.184:https (CLOSE_WAIT) ... #+end_example Am I using Amazonka and/or Conduit in a way that results in this? How do I should I use them? Or, is it an issue somewhere "below" my code? What can I do address that? Thanks for any insights or help /M -- Magnus Therning OpenPGP: 0x927912051716CE39 email: magnus(a)therning.org twitter: magthe http://magnus.therning.org/ Action is the foundational key to all success. — Pablo Picasso

4 4

Re: [Haskell-cafe] Textbook example of instance Foldable ((,)a)
by Barak A. Pearlmutter 26 Nov '20

26 Nov '20

On Tue, 24 Nov 2020, Henning Thielemann wrote: > If people would use custom pair types we would not need Foldable on pairs, > at all. To be fair, if people would use custom pair types we would not need *pairs* at all. --Barak Pearlmutter

4 9

network v3.1.2.1
by Kazu Yamamoto 26 Nov '20

26 Nov '20

Hi cafe, network v3.1.2.1 has been released. This version is bug-fixes only: * Increasing base lower bound to 4.9. [#473](https://github.com/haskell/network/pull/473) * Suppressing errors from removeFile in UNIX bind. [#478](https://github.com/haskell/network/pull/478) * Restoring UNIX stub functions on Windows. [#489](https://github.com/haskell/network/pull/489) --Kazu

1 0

Types of partially instantiated polymorphic helper functions
by Todd Wilson 26 Nov '20

26 Nov '20

Dear Cafe, This has got to be a trivial question, but I don't see a solution, nor could I find anything through searching: If I want to write a type declaration for the helper function g here, what do I write? f :: a -> [b] -> [(a,b)] f a bs = g bs where g :: ???? g [] = [] g (x:xs) = (a,x) : g xs Apologies if I'm wasting bandwidth with my ignorance. Todd Wilson

5 6

Looking for idiomatic example of wrapping a Haskell library
by Stuart Hungerford 25 Nov '20

25 Nov '20

Hi Haskellers, In a Haskell learning project I'm looking at using a couple of Haskell packages (from Hackage) for linear algebra and graphics. I've got a strong feeling my choices of these libraries are going to change as the project progresses and to avoid having to change references to these library functions throughout my modules I'd like to "wrap" them in modules of my own. Can anyone point me to a project where a Haskell library has been idiomatically wrapped this way? I can see newtypes and "smart" constructors of the wrapped types will go some way towards this, but I suspect there's more techniques I haven't yet learned. Thanks in advance, Stu

2 2

Re: [Haskell-cafe] What to call Occult Effects
by Olaf Klinke 24 Nov '20

24 Nov '20

> Dear Olaf and everyone, > > the law do {_ <- b; a} = a is equivalent to do {_ <- b; return ()} = > return (), for, assuming the latter, > > do {_ <- b; a} > = do {_ <- do {_ <- b; return ()}; a} > = do {_ <- return (); a} > = a > > and it does actually have some history. It was dubbed > 'discardability' by Thielecke [1] and explored by > Führmann in [2] together with some other important properties of > effects, such as 'centrality' and > 'commutativity'. Combinations of these properties and examples are > explored in the conventional > Haskell-like style in [3] (where 'discardability' is called 'side- > effect freeness' though). > > Cheers, > Sergey > > [1] H. Thielecke, Categorical structure of continuation passing > style, Ph.D. Thesis, University of > Edinburgh, 1997. > > [2] C. Führmann, Varieties of effects, in: M. Nielsen, U. Engberg > (Eds.), Foundations of Software > Science and Computation Structures, FOSSACS 2002, in: Lect. Notes > Comput. Sci., vol. 2303, Springer, > 2002, pp. 144–158. > > [3] L. Schröder, T. Mossakowski, Monad-independent dynamic logic in > HasCasl, J. Logic Comput. 14 (2004) > 571–619 I've got the impression that all three of the above works look at a problem dual to the OP in a certain sense: At least [1] and [2] ask: For a given monad M, what is the subset of computations b with b >> a = a for all a? In contrast, I understood Kim-Ee's question as: What are the monads M where all computations b belong to this class? Assuming this property, the monoid M () is isomorphic to (). Indeed, the monoid M () inherits the property a <> b = b for all a, b :: M () and the only monoid with this property is the one- element monoid. In category theory a monad M with M 1 being isomorphic to the terminal object 1 is called 'affine'. Bart Jacobs in [4] gives a diagram (10) that amounts to return.const = fmap.const and shows that this is equivalent to being affine. Hence we might also use 'affine' for monads with Kim-Ee's property. Bart Jacobs' proof first reduces the general identity return.const = fmap.const :: a -> m b -> m a to the case b = () just as Sergey did above. Then = fmap (const a) = fmap (const a) . (id :: m () -> m ()) = fmap (const a) . return . const () by assumption = return . const a . const () by naturality of return = return . const a property of const I also learned that the school of 'effects' tries to avoid monads (this going back to John Power), so the terms 'discardable' and 'affine' might just be names for the same thing in different communities. Further there is a categorical construction (see [4]) to compute the greatest affine sub-monad of a monad. Interestingly, the affine part of the state monad is the reader monad, that is, functions s -> (a,s) that may use s but not change it. Olaf [4] https://hal.inria.fr/hal-01446033/file/418352_1_En_5_Chapter.pdf

2 1

Virtual Munich Haskell Meeting, 2020-11-25 @ 19:30
by Leah Neukirchen 23 Nov '20

23 Nov '20

Dear all, This week, our monthly Munich Haskell Meeting will take place again on Wednesday, November 25 at 19:30. **Due to meetup limitations in Bavaria, this meeting will take place online!** For details see here: https://muenchen.haskell.bayern/dates.html Everybody is welcome, especially the Haskellers from Bavaria that do not usually come to our Munich meetings due to travel distance! cu, -- Leah Neukirchen <leah(a)vuxu.org> https://leahneukirchen.org/

1 0

xmonad: eyes based navigation between monitors.
by Karel Gardas 20 Nov '20

20 Nov '20

Hello, I'm using xmonad in a setup with 3 monitors and I'm curious if anybody here got kind of eyes-based navigation working between monitors/workspaces. Explanation: I need to type into windows on different monitors but in not specific order. Sometime I got too deep into thinking that I forget to switch workspace and then I type to different window in different workspace than is intended (and which I see with eyes). Basically I'd like some mode where xmonad will switch workspaces based on my eye/head position so only workspace I follow with eyes will be active. I see that such software (following head/eyes) already exist in AI domain (E.g. NV's xavier NX developer kit demo: from 16:00: https://www.youtube.com/watch?v=aUz1bNGm04w) but I'm more interested in amd64/linux setup as this is for my workstation... Any idea highly appreciated! Thanks! Karel

4 3