November 2020 - Haskell-Cafe

Psa : perhaps time for maintainers to start migrating off Travis ci
by Carter Schonwald 15 Dec '20

15 Dec '20

Hey everyone: it looks like, from my perspective and experiences, that Travis ci should perhaps now be viewed as not open source friendly. Or even converging on hostile? 1) crazy long queue times/ latency for oss ci actions to run 2) very low concurrency on oss builds. 3) very low build build minute caps for oss that require high touch customer support contact to adjust. I’ve started moving my own projects slowly to gh actions for now, though there’s also gitlab ci , src hut and other options that may suit different folks. There’s definitely some ways to keep on having the clever cabal caching we know and love that folks like the Haskell-ci folks and others have hacked out for Travis be available on other platforms, though I don’t think there’s consolidated docs for those yet ? Def seen it discussed though. https://github.com/haskell-CI/haskell-ci/issues/411 Heres a url to my dupe ticket where I share an example naive use of the setup Haskell gh actions Config, definitely not perfect. But kinda amazing to have Mac and Linux and windows ci all in one ! :)

7 14

What features should an (fictitious) IDE for Haskell have?
by Gueven Bay 08 Dec '20

08 Dec '20

Greetings, Imagine that you can decide what functionality and features an Integrated Development Environment for Haskell should get. You can make this IDE with your wishlist similar in power as other IDEs for Java for example. Now list as many features as you can, what would make the life of a Haskell developer as comfortable as possible!? I am dying to know how this list of features at the end (of this thread) will look like.

22 38

STM rollback?
by amindfv＠mailbox.org 04 Dec '20

04 Dec '20

I'd like to be able to give up on an STM transaction: roll back and don't retry. I've cooked up something with exceptions but it feels a bit icky to use exceptions for something like this - is there a better way?: data Rollback = Rollback deriving (Show) instance Exception Rollback rollback :: STM x rollback = throwSTM Rollback atomicallyWithRollback :: STM x -> IO (Maybe x) atomicallyWithRollback a = (Just <$> atomically a) `catch` (\Rollback -> pure Nothing) The alternative I've found is something like: otherWay :: STM x -> IO (Maybe x) otherWay a = atomically $ (Just <$> a) `orElse` pure Nothing But this turns any "retry" in "a" into a rollback, and I'd like to have the option to do either (retry or rollback). Thanks, Tom

6 6

Textbook example of instance Foldable ((,) a)
by Johannes Waldmann 29 Nov '20

29 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 25 Nov '20

25 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 25 Nov '20

25 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 24 Nov '20

24 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