April 2015 - Haskell-Cafe

Ord for partially ordered sets
by Ivan Lazar Miljenovic 27 Apr '15

27 Apr '15

What is the validity of defining an Ord instance for types for which mathematically the `compare` function is partially ordered? Specifically, I have a pull request for fgl [1] to add Ord instances for the graph types (based upon the Ord instances for Data.Map and Data.IntMap, which I believe are themselves partially ordered), and I'm torn as to the soundness of adding these instances. It might be useful in Haskell code (the example given is to use graphs as keys in a Map) but mathematically-speaking it is not possible to compare two arbitrary graphs. What are people's thoughts on this? What's more important: potential usefulness/practicality or mathematical correctness? (Of course, the correct answer is to have a function of type a -> a -> Maybe Ordering :p) [1]: https://github.com/haskell/fgl/pull/11 -- Ivan Lazar Miljenovic Ivan.Miljenovic(a)gmail.com http://IvanMiljenovic.wordpress.com

13 30

Haskell Data.Vector, huge memory leak
by Florian Gillard 27 Apr '15

27 Apr '15

Hi, I am trying to make a basic 2D engine with haskell and the SDL1.2 bindings (for fun, I am just learning). Ideally the world is to be procedurally generated, chunk by chunk, allowing free exploration. Right now my chunk is composed of 200*200 tiles which I represent using a type: Mat [Tile] = Vec.Vector (Vec.Vector [Tile]) and these functions: fromMat :: [[a]] -> Mat a fromMat xs = Vec.fromList [Vec.fromList xs' | xs' <- xs] (§) :: Mat a -> (Int, Int) -> a v § (r, c) = (v Vec.! r) Vec.! c I am using cyclic list of tiles in order to allow for sprite animation, and later for dynamic behaviour. Each frame of the game loop, the program reads the part of the vector relevant to the current camera position, display the corresponding tiles and return a new vector in which every of these cyclic lists has been replaced by it's tail. Here is the code responsible for this: applyTileMat :: Chunk -> SDL.Surface -> SDL.Surface -> IO Chunk applyTileMat ch src dest = let m = chLand $! ch (x,y) = chPos ch wid = Vec.length (m Vec.! 0) - 1 hei = (Vec.length m) - 1 (canW,canH) = canvasSize ch in do sequence $ [ applyTile (head (m § (i,j))) (32*(j-x), 32*(i-y)) src dest | i <- [y..(y+canH)], j <- [x..(x+canW)]] m' <-sequence $ [sequence [(return $! tail (m § (i,j))) | j <- [0..wid]] | i <- [0..hei]] --weird :P return ch { chLand = fromMat m' } the first sequence does the display part, the second one returns the new vector m'. At first I was using the following comprehension to get m' let !m' = [id $! [(tail $! (m § (i,j))) | j <- [0..wid]] | i <- [0..hei]] but doing so results in ever increasing memory usage. I think it has to do with lazy evaluation preventing the data to be properly garbage collected, but I don't really understand why. In this particular case, it doesn't really mater since I have to look at the whole vector. But I don't know how I should do if I wanted to only "update" part of my chunk each frame, thus making a new chunk with only part of the data from the previous one. I am probably not using Data.Vector the way it's intended, but it's the simplest data structure I found with O(n) random access. The whole code is there: https://github.com/eniac314/wizzard/blob/master/tiler.hs

3 2

ANN: getopt-generics - easily create command line parsers by declaring a data type
by shahn 27 Apr '15

27 Apr '15

Hi all, getopt-generics is an experimental library that uses generic programming to automatically create command line parsers given a data type. For an example please refer to the README: [1]. You can also browse the API on hackage: [2]. The primary use-case for getopt-generics is to allow to create proper command line interfaces with as little code as possible. Any feedback appreciated! Cheers, Sönke [1] https://github.com/zalora/getopt-generics#getopt-generics [2] http://hackage.haskell.org/package/getopt-generics/docs/System-Console-GetO…

1 0

Function hanging in infinite input
by martin 26 Apr '15

26 Apr '15

Hello all, I was trying to implement >>= (tBind) on my "Temporal" data type and found that it hangs on an operation like takeInitialPart $ infiniteTemporal >>= (\x -> finiteTemporal) I am pretty sure the result is well defined and by no means infinite. Also the code works on finite Temporals. How does one address such problems? I attach the relevant pieces of code, in case someone would be so kind and inspect or run it. Feel free to point out any flaws as I might be completely off-track. If you call ex10 in GHCI, you get no result. The line marked with "< here" is executed over and over, but apparently without contributing to the result. The debugger shows, that "hd" does have a correct, finite value each time and that "tpr" is consumed as expected. -- -- -- -- -- -- -- -- -- examples -- -- -- -- -- -- -- ex1 = Temporal [(DPast, 1), (T 3,3), (T 7, 7)] :: Temporal Int ex10 = tUntil (T 5) $ outer `tBind` \_ -> ex1 :: Temporal Int where outer = Temporal $ (DPast,0):[(T (fromIntegral t), t)| t <- [5,10 ..]] -- -- -- -- -- -- -- -- -- tBind -- -- -- -- -- -- -- -- -- Changed (Temporal a) to (Temporal Int) for debugging tBind :: (Temporal Int) -> (Int -> Temporal Int) -> Temporal Int tBind tpr f -- tpr is infinite in this example, let's forget these cases -- | tNull tpr = error "empty Temporal" -- | tNull (tTail tpr) = laties | otherwise = let hd = (tUntil (tTt tpr) laties) in hd `tAppend` (tTail tpr `tBind` f) -- < here where laties = switchAt (tTh tpr) ( f (tVh tpr)) tTail (Temporal xs) = Temporal (tail xs) tAppend (Temporal as) (Temporal bs) = Temporal (as ++ bs) switchAt t tpx | tNull (tTail tpx) = Temporal (tot tpx) | between t (tTh tpx) (tTt tpx) = Temporal (tot tpx) | otherwise = switchAt t (tTail tpx) where tot (Temporal ((ty,vy):xs)) = ((max t ty, vy):xs) between t x y = t >= x && t < y -- -- -- -- -- -- -- -- -- helpers -- -- -- -- -- -- -- -- data Time = DPast | T Integer deriving (Eq, Show) -- DPast is "distant past" instance Ord Time where compare DPast DPast = EQ compare DPast _ = LT compare _ DPast = GT compare (T t1) (T t2) = compare t1 t2 -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- data Temporal a = Temporal [(Time, a)] deriving (Eq, Show) tVh :: Temporal a -> a tVh = snd . head . toList -- value head tTt, tTh :: Temporal a -> Time tTt = fst . head . tail . toList -- time tail tTh = fst . head . toList -- time head tNull = null . toList tUntil :: Time -> Temporal a -> Temporal a tUntil t (Temporal xs) = Temporal $ (takeWhile (\(tx, vx) -> tx > t)) xs toList :: Temporal a -> [(Time, a)] toList (Temporal xs) = xs

4 8

503 on upload to Hackage?
by Magnus Therning 26 Apr '15

26 Apr '15

Yesterday and today I made a few releases of a package to Hackage. On each upload I get a 503 error. This is the text: ~~~ Error 503 backend read error backend read error Guru Mediation: Details: cache-fra1229-FRA 1430030130 1668038750 Varnish cache server ~~~ I'm guessing this isn't expected behaviour. Has anyone else seen this? /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus(a)therning.org jabber: magnus(a)therning.org twitter: magthe http://therning.org/magnus Finagle's Second Law: Always keep a record of data -- it indicates you've been working.

2 2

Re: [Haskell-cafe] GHCi shows result of (IO a) only if (a) is in class Show
by Sumit Sahrawat, Maths & Computing, IIT (BHU) 26 Apr '15

26 Apr '15

On 26 April 2015 at 06:10, Alexey Vagarenko <vagarenko(a)gmail.com> wrote: > Yes, but ghci shows an error if it can't print a value, except when the > value is in IO monad. Compare: > Prelude> id > <interactive>:6:1: > No instance for (Show (a0 -> a0)) > (maybe you haven't applied enough arguments to a function?) > arising from a use of `print' > In a stmt of an interactive GHCi command: print it > Prelude> > > and > > Prelude> return id > Prelude> > This behavior is necessary. For example, if we used any function with result of type IO (), such as writeFile, we don't want an error as we are interested in the side-effects only. > > воскресенье, 26 апреля 2015 г., 5:48:55 UTC+6 пользователь Sumit Sahrawat, > Maths & Computing, IIT (BHU) написал: >> >> If something can't be shown (converted to a string), then it can't be >> printed (as a string). >> >> On 25 April 2015 at 17:51, Daniel van den Eijkel <dv...(a)gmx.net> wrote: >> >>> I wrote a parser and it took me a while to realize why GHCi suddenly did >>> not show any result nor an error message anymore. >>> >>> My parsing function has type (IO Expression), Expression is in class >>> Show. >>> >>> After changing the parser to (IO Declaration), it did not show anything >>> anymore, because Declaration was not in class Show. >>> >>> When I typed (parseFile "input.txt" >>= print), I got the error message >>> and understood what was going on. But for I while I was really confused >>> what's happening. >>> >>> Just wanted to share this. >>> >>> Best, >>> Daniel >>> _______________________________________________ >>> Haskell-Cafe mailing list >>> Haskel...(a)haskell.org >>> http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe >>> >> >> >> >> -- >> Regards >> >> Sumit Sahrawat >> > -- Regards Sumit Sahrawat

1 1

Re: [Haskell-cafe] GHCi shows result of (IO a) only if (a) is in class Show
by Sumit Sahrawat, Maths & Computing, IIT (BHU) 26 Apr '15

26 Apr '15

On 26 April 2015 at 06:10, Alexey Vagarenko <vagarenko(a)gmail.com> wrote: > Yes, but ghci shows an error if it can't print a value, except when the > value is in IO monad. Compare: > Prelude> id > <interactive>:6:1: > No instance for (Show (a0 -> a0)) > (maybe you haven't applied enough arguments to a function?) > arising from a use of `print' > In a stmt of an interactive GHCi command: print it > Prelude> > > and > > Prelude> return id > Prelude> > > This behavior is necessary. For example, if we used any function with result of type IO (), such as writeFile, we don't want an error as we are interested in the side-effects only. > > > воскресенье, 26 апреля 2015 г., 5:48:55 UTC+6 пользователь Sumit Sahrawat, > Maths & Computing, IIT (BHU) написал: >> >> If something can't be shown (converted to a string), then it can't be >> printed (as a string). >> >> On 25 April 2015 at 17:51, Daniel van den Eijkel <dv...(a)gmx.net> wrote: >> >>> I wrote a parser and it took me a while to realize why GHCi suddenly did >>> not show any result nor an error message anymore. >>> >>> My parsing function has type (IO Expression), Expression is in class >>> Show. >>> >>> After changing the parser to (IO Declaration), it did not show anything >>> anymore, because Declaration was not in class Show. >>> >>> When I typed (parseFile "input.txt" >>= print), I got the error message >>> and understood what was going on. But for I while I was really confused >>> what's happening. >>> >>> Just wanted to share this. >>> >>> Best, >>> Daniel >>> _______________________________________________ >>> Haskell-Cafe mailing list >>> Haskel...(a)haskell.org >>> http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe >>> >> >> >> >> -- >> Regards >> >> Sumit Sahrawat >> > -- Regards Sumit Sahrawat

1 0

GHCi shows result of (IO a) only if (a) is in class Show
by Daniel van den Eijkel 25 Apr '15

25 Apr '15

I wrote a parser and it took me a while to realize why GHCi suddenly did not show any result nor an error message anymore. My parsing function has type (IO Expression), Expression is in class Show. After changing the parser to (IO Declaration), it did not show anything anymore, because Declaration was not in class Show. When I typed (parseFile "input.txt" >>= print), I got the error message and understood what was going on. But for I while I was really confused what's happening. Just wanted to share this. Best, Daniel

2 1

Re: [Haskell-cafe] How to compile git-annex?
by Chris Wong 25 Apr '15

25 Apr '15

On Apr 26, 2015 10:01 AM, "Peng Yu" <pengyu.ut(a)gmail.com> wrote: > > I clean up all Haskell related things in my home directory. > > rm -rf ~/.cabal > rm -rf ~/Library/Haskell > rm -rf ~.ghc/ Is this a typo? You should have removed ~/.ghc, not ~.ghc. > cabal update > cabal install cabal-install > > > Then I try to install git-annex again, which shows me the following. > What I should do next? > > ~$ cabal install git-annex > Resolving dependencies... > In order, the following would be installed: > HUnit-1.2.5.2 (new package) > SafeSemaphore-0.10.1 (new package) > ansi-terminal-0.6.2.1 (new package) > ansi-wl-pprint-0.6.7.2 (new package) > appar-0.1.4 (new package) > async-2.0.2 (new package) > auto-update-0.1.2.1 (new package) > base16-bytestring-0.1.1.6 (new package) > base64-bytestring-1.0.0.1 (new package) > blaze-builder-0.4.0.1 (new package) > blaze-markup-0.7.0.2 (new package) > blaze-html-0.8.0.2 (new package) > bloomfilter-2.0.0.0 (new package) > byteable-0.1.1 (new package) > byteorder-1.0.4 (new package) > bytestring-builder-0.10.6.0.0 (new package) > cereal-0.4.1.1 (new package) > clock-0.4.5.0 (new package) > cryptohash-0.11.6 (new package) > data-default-class-0.0.1 (new package) > data-default-instances-base-0.0.1 (new package) > data-default-instances-containers-0.0.1 (new package) > data-default-instances-old-locale-0.0.1 (new package) > dataenc-0.14.0.7 (new package) > dlist-0.7.1.1 (new package) > data-default-instances-dlist-0.0.1 (new package) > data-default-0.5.3 (new package) > cookie-0.4.1.4 (new package) > easy-file-0.2.0 (new package) > edit-distance-0.2.1.2 (new package) > entropy-0.3.6 (new package) > fast-logger-2.3.1 (new package) > file-embed-0.0.8.2 (new package) > gnuidn-0.2.1 (new package) > hashable-1.2.3.2 (new package) > case-insensitive-1.2.0.4 (new package) > hourglass-0.2.9 (new package) > asn1-types-0.3.0 (new package) > http-types-0.8.6 (new package) > iproute-1.4.0 (new package) > mime-mail-0.4.8.2 (new package) > mime-types-0.1.0.6 (new package) > monad-loops-0.4.2.1 (new package) > nats-1 (new package) > network-info-0.2.0.5 (new package) > network-multicast-0.0.11 (new package) > parallel-3.2.0.6 (new package) > path-pieces-0.2.0 (new package) > prelude-extras-0.4 (new package) > reflection-1.5.1.2 (new package) > safe-0.3.8 (new package) > scientific-0.3.3.8 (new package) > attoparsec-0.12.1.6 (new package) > css-text-0.1.2.1 (new package) > email-validate-2.1.1 (new package) > http-date-0.0.6 (new package) > securemem-0.1.7 (new package) > crypto-cipher-types-0.0.9 (new package) > cipher-aes-0.2.10 (new package) > cipher-des-0.0.6 (new package) > cipher-rc4-0.1.4 (new package) > setenv-0.1.1.3 (new package) > silently-1.2.4.1 (new package) > simple-sendfile-0.2.18 (new package) > socks-0.5.4 (new package) > stm-chans-3.0.0.3 (new package) > stringsearch-0.3.6.6 (new package) > system-filepath-0.4.13.3 (new package) > system-fileio-0.3.16.2 (new package) > tagged-0.8.0.1 (new package) > tagsoup-0.13.3 (new package) > transformers-0.4.3.0 (new version) > StateVar-1.1.0.0 (new package) > crypto-api-0.13.2 (new package) > gsasl-0.3.5 (new package) > mmorph-1.0.4 (new package) > monads-tf-0.1.0.2 (new package) > gnutls-0.1.5 (new package) > mtl-2.2.1 (new version) > IfElse-0.85 (new package) > asn1-encoding-0.9.0 (new package) > asn1-parse-0.9.0 (new package) > crypto-pubkey-types-0.4.3 (new package) > hfsevents-0.1.5 (new package) > hslogger-1.2.8 (new package) > parsec-3.1.9 (reinstall) changes: mtl-2.1.3.1 -> 2.2.1 > bencode-0.5.0.1 (new package) > json-0.9.1 (new package) > network-uri-2.6.0.1 (reinstall) > pem-0.2.2 (new package) > primitive-0.6 (new package) > regex-base-0.93.2 (new package) > regex-posix-0.95.2 (new package) > regex-compat-0.95.1 (new package) > MissingH-1.3.0.1 (new package) > regex-tdfa-1.2.0 (new package) > skein-1.0.9.3 (new package) > streaming-commons-0.1.12 (new package) > torrent-10000.0.0 (new package) > transformers-compat-0.4.0.4 (new package) > MonadRandom-0.3.0.2 (new package) > distributive-0.4.4 (new package) > exceptions-0.8.0.2 (new package) > optparse-applicative-0.11.0.2 (new package) > transformers-base-0.4.4 (new package) > monad-control-1.0.0.4 (new package) > lifted-base-0.2.3.6 (new package) > enclosed-exceptions-1.0.1.1 (new package) > resourcet-1.1.4.1 (new package) > unix-compat-0.4.1.4 (new package) > unix-time-0.3.5 (new package) > unordered-containers-0.2.5.1 (new package) > semigroups-0.16.2.2 (new package) > utf8-string-1 (new package) > language-javascript-0.5.13.3 (new package) > hjsmin-0.1.4.7 (new package) > publicsuffixlist-0.1 (new package) > http-client-0.4.11.1 (new package) > uuid-types-1.0.1 (new package) > uuid-1.3.10 (new package) > vault-0.3.0.4 (new package) > vector-0.10.12.3 (new package) > aeson-0.8.0.2 +old-locale (new package) > crypto-random-0.0.9 (new package) > cprng-aes-0.6.1 (new package) > clientsession-0.9.1.1 (new package) > crypto-numbers-0.2.7 (new package) > crypto-pubkey-0.2.8 (new package) > mwc-random-0.13.3.2 (new package) > resource-pool-0.2.3.2 (new package) > shakespeare-2.0.4.1 (new package) > hamlet-1.2.0 (new package) > void-0.7 (new package) > conduit-1.2.4 (new package) > conduit-extra-1.1.7.2 (new package) > contravariant-1.3.1 (new package) > comonad-4.2.5 (new package) > cryptohash-conduit-0.1.1 (new package) > dns-1.4.5 (new package) > monad-logger-0.3.13.1 (new package) > persistent-2.1.3 (new package) > esqueleto-2.1.3 (new package) > persistent-sqlite-2.1.4.1 (new package) > persistent-template-2.1.3 (new package) > semigroupoids-4.3 (new package) > bifunctors-4.2.1 (new package) > profunctors-4.4.1 (new package) > free-4.11 (new package) > adjunctions-4.2 (new package) > either-4.3.3.2 (new package) > errors-1.4.7 (new package) > kan-extensions-4.2.1 (new package) > lens-4.9.1 (new package) > wai-3.0.2.3 (new package) > wai-logger-2.2.4 (new package) > warp-3.0.12.1 (new package) > word8-0.1.2 (new package) > wai-extra-3.0.7.1 (new package) > wai-app-static-3.0.1 (new package) > x509-1.5.0.1 (new package) > x509-store-1.5.0 (new package) > x509-system-1.5.0 (new package) > x509-validation-1.5.1 (new package) > tls-1.2.17 (new package) > connection-0.2.4 (new package) > http-client-tls-0.2.2 (new package) > http-conduit-2.1.5 (new package) > warp-tls-3.0.3 (new package) > xml-types-0.3.4 (new package) > libxml-sax-0.7.5 (new package) > network-protocol-xmpp-0.4.6 (new package) > xml-conduit-1.2.4 (new package) > aws-0.11.4 (new package) > tagstream-conduit-0.5.5.3 (new package) > authenticate-1.3.2.11 (new package) > xml-hamlet-0.4.0.10 (new package) > DAV-1.0.4 (new package) > xss-sanitize-0.3.5.5 (new package) > yaml-0.8.11 (new package) > yesod-core-1.4.9.1 (new package) > yesod-default-1.2.0 (new package) > yesod-persistent-1.4.0.2 (new package) > yesod-form-1.4.4.1 (new package) > yesod-auth-1.4.4 (new package) > yesod-1.4.1.5 (new package) > yesod-static-1.4.0.4 (new package) > git-annex-5.20150420 -testsuite -feed (new package) > cabal: The following packages are likely to be broken by the reinstalls: > HTTP-4000.2.19 > Use --force-reinstalls if you want to install anyway. > > > On Sat, Apr 25, 2015 at 4:18 PM, Brandon Allbery <allbery.b(a)gmail.com> wrote: > > > > On Sat, Apr 25, 2015 at 3:45 PM, Peng Yu <pengyu.ut(a)gmail.com> wrote: > >> > >> Why cabal install print so many irrelevant messages? Is it better to > >> follow Unix "Rule of Silence" to only print usage error messages? > > > > > > This is not an irrelevant message: > > > >> asn1-parse-0.9.0 (reinstall) changes: text-1.2.0.4 added > > > > > > It's an indication that things are about to go very wrong, as indeed they > > did. In fact, the errors you got indicate quite a lot of problems with your > > Haskell installation; you apparently have a bunch of broken packages. > > > > -- > > brandon s allbery kf8nh sine nomine associates > > allbery.b(a)gmail.com ballbery(a)sinenomine.net > > unix, openafs, kerberos, infrastructure, xmonad http://sinenomine.net > > > > -- > Regards, > Peng > _______________________________________________ > cabal-devel mailing list > cabal-devel(a)haskell.org > http://mail.haskell.org/cgi-bin/mailman/listinfo/cabal-devel -- Chris Wong https://lambda.xyz

1 0

Re: [Haskell-cafe] Ord for partially ordered sets
by Barak A. Pearlmutter 24 Apr '15

24 Apr '15

> I'm confused. What is supposed to be the result of `g1 <= g2` when `g1` and > `g2` are not comparable according to the partial order? Surely it would make things less confusing to simply follow existing precedent already in the standard prelude? $ ghci Prelude> let nan=0/0::Double Prelude> nan==nan False Prelude> compare 0 nan GT Prelude> compare nan 0 GT Prelude> compare nan nan GT Prelude> 0<=nan False Prelude> nan<=0 False Prelude> nan<=nan False Prelude> let infty=1/0::Double Prelude> infty <= nan False Prelude> nan <= infty False Perhaps not.

1 0