December 2012 - Haskell-Cafe

Documentation operator
by Christopher Done 27 Dec '12

27 Dec '12

Hello chums, I've been playing around with an idea, something that has obvious pros and cons, but I'll sell it to you because there might be some positive ideas out of it. Consider the following operator: {-# LANGUAGE TypeOperators, DataKinds, KindSignatures #-} module Docs where import GHC.TypeLits type a ? (sym :: Symbol) = a First I'll describe how I'd want to use this and then what I think are the advantages and disadvantages. I call this (?) operator “the documentation operator”, to be used for: * Things that either don't belong or can't be encoded in the type system, or for things need to be in English. * Things that cannot be encoded in Haddock. The simple case of ye olde days: -- | Lorem ipsum dolor sit amet. Suspendisse lacinia nibh et -- leo. Aenean auctor aliquam dapibus. loremIpsum :: Int -> Int -> String Which has since been somewhat evolved into: loremIpsum :: Int -- ^ Lorem ipsum dolor sit amet. -> Int -- ^ Suspendisse lacinia nibh et leo. -> String -- ^ Aenean auctor aliquam dapibus. But could now be written: loremIpsum :: Int ? "Lorem ipsum dolor sit amet." -> Int ? "Suspendisse lacinia nibh et leo." -> String ? "Aenean auctor aliquam dapibus." Here is a contrived case I'll use later on: data Person = Person describeAge :: Int ? "an age" -> String ? "description of their elderliness" describeAge n = undefined personAge :: Person ? "a person" -> Int ? "their age" personAge = undefined One could also encode previously informal specifications more formally, so that -- | The action 'hFlush' @hdl@ causes any items buffered for output -- in handle @hdl@ to be sent immediately to the operating system. -- -- This operation may fail with: -- -- * 'isFullError' if the device is full; -- -- * 'isPermissionError' if a system resource limit would be exceeded. -- It is unspecified whether the characters in the buffer are discarded -- or retained under these circumstances. hFlush :: Handle -> IO () hFlush handle = wantWritableHandle "hFlush" handle flushWriteBuffer with type Throws ex (docs :: Symbol) = docs could now be written hFlush :: Handle ? "flush buffered items for output on this handle" -> IO () ? Throws IsFullError "if the device is full" ? Throws IsPermissionError "if a system resource limit would be exceeded. It is \ \unspecified whether the characters in the buffer are \ \discarded or retained under these circumstances." hFlush handle = wantWritableHandle "hFlush" handle flushWriteBuffer With this in place, in GHCi you get documentation "lookup" for free: > :t hFlush hFlush :: (Handle ? "flush buffered items for output on this handle") -> (IO () ? Throws IsFullError "if the device is full") ? Throws IsPermissionError "if a system resource limit would be exceeded. It is unspecified whether the characters in the buffer are discarded or retained under these circumstances." And you get function composition, or “documentation composition” for free: > :t describeAge . personAge describeAge . personAge :: (Person ? "a person") -> String ? "description of their elderliness" We could have a :td command to print it with docs, and otherwise docs could be stripped out trivially by removing the ? annotations: > :t describeAge . personAge describeAge . personAge :: Person -> String > :td describeAge . personAge describeAge . personAge :: (Person ? "a person") -> String ? "description of their elderliness" You could even add clever printing of such “documentation types”: > :t hFlush hFlush :: Handle — flush buffered items for output on this handle -> IO () Throws IsFullError if the device is full" Throws IsPermissionError if a system resource limit would be exceeded. It is unspecified whether the characters in the buffer are discarded or retained under these circumstances." Unfortunately it doesn't work with monadic composition, of course. So here are the advantages: * You get parsing for free (and anyone using haskell-src-exts). * You get checking for free (i.e. GHC can check that IsFullError exists for you). * You get a continuity of documentation through your operations including composition. * You can extend the "documentation language" easily by just defining some types (like the Throws I used above). SeeMore, Author, Deprecated, etc. Whatever. * You can print out some helpful looking documentation in GHCi based on these simple types. * There's no longer this informal "it might throw this exception" kind of pros we're forced to write. * It could also be used for annotations other than pure documentation, including testing. E.g. add a Testable "property" and then your test framework can search for functions with this Testable annotation. * Free of Haddock's syntax. Here are the disadvantages: * It doesn't work for types. * Writing big pros inside a string can be boring without a decent editor. * The neat composition trick only goes so far. * There might be a compilation overhead. * It would require an updated GHCi to strip them out when not wanted. * Requires GHC 7.6.1+. Conclusions: What we have now for documentation is pretty good, especially generated documentation. Compared to other languages Haskell is quite well documented, I feel. But we can do more with it. In some languages, documentation is built into the language. You can ask for documentation inside the REPL, it belongs to that piece of code. It shouldn't, I don't think, be left as a code comment which is essentially whitespace as far as the compiler is concerned. Two sweet ideas that I like from the above are: * The checking by GHC. * The extension of the "documentation language", with the ability to formalize things like what exceptions are thrown. * Composing functions generates "new" documentation that still makes sense. Thoughts? Ciao!

4 3

Non polymorphic numerals option -- avoiding type classes
by Rustom Mody 27 Dec '12

27 Dec '12

In haskell, we have Prelude> :t 4 4 :: Num a => a Prelude> This may be nice in its generality but it makes it hard (for me at least) when teaching a beginners course to teach polymorphic vs monomorphic types. The above leads to even more 'advanced' results like this: Prelude> :t [[1],2] [[1],2] :: (Num [t], Num t) => [[t]] Prelude> [[1],2] <interactive>:5:6: No instance for (Num [t0]) arising from the literal `2' Possible fix: add an instance declaration for (Num [t0]) In the expression: 2 In the expression: [[1], 2] In an equation for `it': it = [[1], 2] By contrast in gofer, numeric literals are monomorphic and no such peculiarities arise ? :t [[1],2] ERROR: Type error in list *** expression : [[1],2] *** term : 2 *** type : Int *** does not match : [Int] [[1],2] ERROR: Type error in list *** expression : [[1],2] *** term : 2 *** type : Int *** does not match : [Int] So is there any set of flags to make haskell literals less polymorphic? ie I want 3 to have type Int and 3.0 to have type Float. This is of course for beginning students to not see type classes too early

6 11

Decent docs to my visualization tools
by Eugene Kirpichov 27 Dec '12

27 Dec '12

Hi cafe, I just finally published a decent documentation website and binary distributions for my log visualization tools written in Haskell - http://jkff.info/software/timeplotters/ . I think now they might actually stand a chance of starting to be adopted by a wider circle than my friends :) Feedback and sharing would be very welcome. -- Eugene Kirpichov http://www.linkedin.com/in/eugenekirpichov

1 0

ANNOUNCE: haskell-docs - Given a module name and a name, it will find and display the documentation of that name.
by Christopher Done 27 Dec '12

27 Dec '12

Ahoy hoy, Just thought I'd announce a tool I whipped up these evening to take a module name and a name and output the installed Haddock documentation for it. Examples with my GHCi session: λ> :doc Data.List.Split split Split a list according to the given splitting strategy. This is how to "run" a Splitter that has been built using the other combinators. λ> :doc Control.Concurrent.MVar swapMVar Take a value from an MVar, put a new value into the MVar and return the value taken. This function is atomic only if there are no other producers for this MVar. λ> :doc Data.List sort Ambiguous module, belongs to more than one package: base haskell2010-1.1.0.1 Continuing anyway... Package: base The sort function implements a stable sorting algorithm. It is a special case of sortBy, which allows the programmer to supply their own comparison function. Please have a play with it, the package is at: http://hackage.haskell.org/package/haskell-docs It has installation instructions. Feel free to share any issues that you have, either here, or on the Github page: https://github.com/chrisdone/haskell-docs There are some issues to do with versioning that I'm not sure how to solve in a standard way. The obvious next step is to have a -package-conf flag so that it can be used with cabal-dev. The wizards on #haskell are currently thinking of a way to avoid having to write the module name and just use what's in scope. Ciao!

4 3

Virthualenv/HsEnv in Windows
by Kyle Hanson 27 Dec '12

27 Dec '12

Has anyone gotten this to work on Windows either native or on cygwin? Is there an alternative? Happy Holidays, Kyle Hanson

2 1

Type error when trying to adapt http-proxy to new conduit
by Pieter Laeremans 27 Dec '12

27 Dec '12

Hi, The http-proxy package isn't compatible any longer with the latest conduit. Since it is open source, I thought, I might as well try to adapt it and submit a patch. However I run into some difficulties. For example I get this type error when I'm trying to compile it : Network/HTTP/Proxy.hs:254:15: Couldn't match expected type `ResourceT IO (CIN.Pipe () () ByteString () (ResourceT IO) ())' with actual type `IO (CIN.Pipe () () ByteString () (ResourceT IO) ())' I can't figure it out. To me it reads like : Expected x, actual x. Anyone has a clue? thanks, Pieter -- Pieter Laeremans <pieter(a)laeremans.org> "The future is here. It's just not evenly distributed yet." W. Gibson

4 3

ANNOUNCE: Haskell BOF on 29c3 in Hamburg, Germany
by Matthias Fischmann 27 Dec '12

27 Dec '12

Hi everybody, we will be running a nano-hackathon on the CCC congress in Hamburg on Thursday, 2012-12-27, 5pm German time: https://events.ccc.de/congress/2012/wiki/Haskell_BOF Sorry for the short notice. If you would like to meet up later during the congress please drop us a line. cheers, matthias

2 1

Substituting values
by Radical 26 Dec '12

26 Dec '12

Sometimes I'll need something like: if value == Foo then Bar else value Or some syntactic variation thereof: case value of { Foo -> Bar; _ -> value } Is there a better/shorter way to do it? I'm surprised that it's more complicated to substitute a value on its own than e.g. in a list, using filter. Or perhaps I'm missing the right abstraction? Thanks, Alvaro

6 8

How to fold on types?
by Magicloud Magiclouds 25 Dec '12

25 Dec '12

Say I have things like: data LongDec = LongDef a b c ... x y z values = [ 'a', 'b', 'c', ... 'x', 'y', 'z' ] Now I want them to be "LongDef 'a' 'b' 'c' ... 'x' 'y' 'z'". In form, this is something like folding. But since the type changes, so code like following won't work: foldl (\def value -> def value) LongDef values Is it possible to do this in some way? -- 竹密岂妨流水过山高哪阻野云飞 And for G+, please use magiclouds#gmail.com.

4 6

multi-thread and lazy evaluation
by Corentin Dupont 25 Dec '12

25 Dec '12

Hi all, I have a program where the user can submit his own little programs, which are interpreted using Hint. The user-submitted programs are used to modify a state held in a TVar. As of course those user-submitted programs can't be trusted, I'm trying to protect them, like in Mueval. I installed a watchdog to monitor and kill the user's thread if it doesn't finish. However it doesn't work properly, due to lazy evaluation I believe. I made a little exemple to illustrate the problem. -> The following program doesn't terminate, but if you uncomment the "putStrLn" at the end, it will. Could someone explain me this and how to do it properly?? Merry Christmas to all!!!! Corentin * {-# LANGUAGE ScopedTypeVariables #-} import Control.Concurrent.STM.TVar import Control.Concurrent.MVar import Control.Concurrent import Control.Monad.STM data MyData = MyData { a :: String, b :: String } deriving (Show) main = do tv <- atomically $ newTVar $ MyData "a" "b" protectedExecCommand tv myNewData <- atomically $ readTVar tv putStrLn $ show myNewData protectedExecCommand :: (TVar MyData) -> IO () protectedExecCommand tv = do mv <- newEmptyMVar before <- atomically $ readTVar tv id <- forkIO $ execBlocking mv forkIO $ watchDog' 5 id mv res <- takeMVar mv case res of Nothing -> (atomically $ writeTVar tv before) Just after -> (atomically $ writeTVar tv after) watchDog' :: Int -> ThreadId -> MVar (Maybe x) -> IO () watchDog' t tid mv = do threadDelay $ t * 1000000 killThread tid putStrLn $ "process timeout " tryPutMVar mv Nothing return () execBlocking :: MVar (Maybe MyData) -> IO () execBlocking mv = do let (a::String) = a --If you uncomment the next line, it will work --putStrLn $ show a putMVar mv (Just $ MyData a "toto")*

4 9