June 2010 - Haskell-Cafe

Updates in the regions library
by Bas van Dijk 08 Jun '10

08 Jun '10

Hi Arie (and others who are interested in the regions library), I would like to let you know that I'm working on a new version of my regions package and its reverse dependencies: regions-monadsfd, regions-monadstf, safer-file-handles, regional-pointers and usb-safe. The major change is that I removed the Resource class: class Resource resource where data Handle resource ∷ * open ∷ resource → IO (Handle resource) close ∷ Handle resource → IO () in favor of a much more simple interface, namely: -- | An 'IO' computation that closes or finalizes a resource. For example -- 'hClose' or 'free'. type CloseAction = IO () -- | A handle to a 'CloseAction' that allows you to duplicate the action to a -- parent region using 'dup'. newtype CloseHandle (r ∷ * → *) -- | Register the 'CloseAction' in the region. When the region terminates all -- registered close actions will be perfomed if they're not duplicated to a -- parent region. register ∷ MonadIO pr ⇒ CloseAction → RegionT s pr (CloseHandle (RegionT s pr)) Here's an example how to use this new interface from my updated regional-pointers package: import qualified Foreign.Marshal.Alloc as FMA -- | A regional handle to memory. This should provide a safer replacement for -- Foreign.Ptr.Ptr data RegionalPtr α (r ∷ * → *) = RegionalPtr (Ptr α) (CloseHandle r) {-| Allocates the given number of bytes and returns a regional pointer to them. This should provide a safer replacement for: Foreign.Marshal.Alloc.mallocBytes. -} mallocBytes ∷ MonadCatchIO pr ⇒ Int → RegionT s pr (RegionalPtr α (RegionT s pr)) mallocBytes size = block $ do ptr ← liftIO $ FMA.mallocBytes size let closeAction = free ptr ch ← register closeAction return $ RegionalPtr ptr ch And here's an example from safer-file-handles: -- | A regional handle to an opened file parameterized by the 'IOMode' in which -- you opened the file and the region in which it was created. data RegionalFileHandle ioMode (r ∷ * → *) = RegionalFileHandle (Handle ioMode) (CloseHandle r) openFile ∷ MonadCatchIO pr ⇒ FilePath → IOMode ioMode → RegionT s pr (RegionalFileHandle ioMode (RegionT s pr)) openFile = openNormal E.openFile -- | Opens a file in binary mode then yields a regional handle to it. This -- provides a safer replacement for System.IO.openBinaryFile. openBinaryFile ∷ MonadCatchIO pr ⇒ FilePath → IOMode ioMode → RegionT s pr (RegionalFileHandle ioMode (RegionT s pr)) openBinaryFile = openNormal E.openBinaryFile openNormal ∷ MonadCatchIO pr ⇒ (FilePath → IOMode ioMode → IO (E.Handle ioMode)) → FilePath → IOMode ioMode → RegionT s pr (RegionalFileHandle ioMode (RegionT s pr)) openNormal open filePath ioMode = block $ do h ← liftIO $ open filePath ioMode let closeAction = sanitizeIOError $ hClose h ch ← register closeAction return $ RegionalFileHandle h ch I haven't released it yet because I need to write some more documentation. However all the repositories contain the new code: darcs get http://code.haskell.org/~basvandijk/code/regions darcs get http://code.haskell.org/~basvandijk/code/regions-monadsfd darcs get http://code.haskell.org/~basvandijk/code/regions-monadstf darcs get http://code.haskell.org/~basvandijk/code/regional-pointers darcs get http://code.haskell.org/~basvandijk/code/safer-file-handles darcs get http://code.haskell.org/~basvandijk/code/usb-safe and some examples: darcs get http://code.haskell.org/~basvandijk/code/usb-safe-examples darcs get http://code.haskell.org/~basvandijk/code/safer-file-handles-examples darcs get http://code.haskell.org/~basvandijk/code/usb-safe-and-safer-file-handles-ex… Regards, Bas On Wed, Jun 2, 2010 at 3:13 PM, Bas van Dijk <v.dijk.bas(a)gmail.com> wrote: > On Wed, Jun 2, 2010 at 2:57 PM, Arie Peterson <ariep(a)xs4all.nl> wrote: >>> On Wed, Jun 2, 2010 at 2:28 PM, Bas van Dijk <v.dijk.bas(a)gmail.com> >> wrote: >>>> Before answering your questions I would like to make sure I understand >>>> your Resource type. When I want to create a memory Resource for >>>> example is the following what you have in mind? >>>> >>>> {-# LANGUAGE Rank2Types #-} >>>> >>>> -- from base: >>>> import Foreign.Ptr ( Ptr ) >>>> import Foreign.Marshal.Alloc ( mallocBytes, free ) >>>> >>>> -- from transformers: >>>> import Control.Monad.IO.Class ( liftIO ) >>>> >>>> -- from MonadCatchIO-transformers: >>>> import Control.Monad.CatchIO ( MonadCatchIO, bracket ) >>>> >>>> newtype Resource cap m = Resource { with :: forall a. (cap -> m a) -> m >>>> a } >>>> >>>> type Memory m a = Resource (Ptr a) m >>>> >>>> memory :: MonadCatchIO m => Int -> Memory m a >>>> memory size = Resource $ bracket (liftIO $ mallocBytes size) (liftIO . >>>> free) >> >> Yes, exactly. I also create type aliases for resources providing a >> specific capability. >> >> On Wed, 2 Jun 2010 14:45:08 +0200, Bas van Dijk <v.dijk.bas(a)gmail.com> >> wrote: >>> The previous can also be generalized using my Resource class: >>> >>> -- from regions: >>> import qualified Control.Resource as C ( Resource(..) ) >>> >>> resource :: (MonadCatchIO m, C.Resource resource) >>> => resource -> Resource (C.Handle resource) m >>> resource r = Resource $ bracket (liftIO $ C.open r) (liftIO . C.close) >> >> Yes, definitely. >> >> (This is not a literal generalisation of the 'memory' function, unless you >> make 'Int' an instance of 'C.Resource'; one would probably create a special >> type 'data Memory = Memory Int' instead. This is the difference I alluded >> to in my earlier email.) > > Indeed. > > My regional-pointers package provides this Memory type and an instance > for Resource: > http://hackage.haskell.org/packages/archive/regional-pointers/0.1.0.2/doc/h… > > Now I'm beginning to understand the problem with your 'fallback' > function. If I specialize the capability to Handles to Memory as in: > > -- from regional-pointers: > import Foreign.Ptr.Region ( Memory ) > > fallback :: Resource (C.Handle (Memory a)) IO > -> Resource (C.Handle (Memory a)) IO > -> Resource (C.Handle (Memory a)) IO > fallback (Resource primary) (Resource backup) = Resource l > where > l c = primary c `catch` (\(SomeException _) -> backup c) > > then the only way to create a value of type 'Resource (C.Handle > (Memory a))' is to apply my previously posted 'resource' function to > 'Memory nrOfBytes'. There's no other way to create Handles to Memory. > Is that your problem? > > The problem is that Handle is an associated _data_ type not an > associated _type synonym_: > > class Resource resource where > data Handle resource ∷ * > open ∷ resource → IO (Handle resource) > close ∷ Handle resource → IO () > > So 'Handle' is an injective type function which means that every > 'Handle resource' has only one 'resource'. > > I could try turning 'Handle' into an associated type synonym. > > Regards, > > Bas >

1 0

PDF generation?
by Jim Tittsler 08 Jun '10

08 Jun '10

What is the easiest way to create PDF files from Haskell? Is gtk2hs's PDF output the preferred way?

13 15

Re: [Haskell-cafe] is there a way to prove the equivalence of these two implementations of (Prelude) break function?
by Alberto G. Corona 08 Jun '10

08 Jun '10

> This isn`t a manifestation of the Curry-Howard isomorphism? > > 2010/6/8 Alexander Solla <ajs(a)2piix.com> > > >> On Jun 7, 2010, at 4:10 PM, Alexander Solla wrote: >> >> >>> You might note how much like evaluating the function generating the >>> analysis is. >>> >> >>

1 0

haskell.org down?
by Ozgur Akgun 08 Jun '10

08 Jun '10

http://downforeveryoneorjustme.com/www.haskell.org -- Ozgur Akgun

4 3

Haddock on HP failing
by Chris Dornan 08 Jun '10

08 Jun '10

Hi, I am trying to use Haddock on the Haskell Platform 2010.1.0.0 and getting haddock.exe: can't find a package database at E:\ghc\ghc-6.12.1\lib\package.conf.d I have tried uninstalling and re-installing HP (just in case) to no avail. Has anybody else seen this? Chris

2 2

After `readFile`, why cannot i `writeFile` ?
by zaxis 07 Jun '10

07 Jun '10

hitSSQ :: String -> [Int] -> IO () hitSSQ no hitNum = do let newNum = unwords $ [no] ++ map (\n -> show n) hitNum hitNums <- fmap lines $ readFile "test.txt" writeFile "test.txt" $ unlines $ hitNums ++ [newNum] *** Exception: test.txt: openFile: resource busy (file is locked) Sincerely! ----- e^(π⋅i) + 1 = 0 -- View this message in context: http://old.nabble.com/After-%60readFile%60%2C-why-cannot-i-%60writeFile%60-… Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.

3 4

How to set a GHC option for a single module and a specific version of GHC?
by Bernie Pope 07 Jun '10

07 Jun '10

Hi, I'm looking for a way to set a GHC compile option on a specific module (not every module in the program) but only for a specific version of GHC. Ideally within the confines of cabal, and in a portable way. GHC provides the OPTIONS_GHC pragma, but it does not appear to provide a way for the pragma to fire for specific versions of GHC. Also, I can't use an #ifdef trick because "File-header pragmas are read once only, before pre-processing the file (e.g. with cpp)." Cabal provides conditional statements and a way to check the version of GHC, but I can't see a way to get it to only use certain compiler options for a particular module. In summary, what I really want is a modified version of the OPTIONS_GHC pragma: {-# OPTIONS_GHC ==6.12.1 ... #-} or a way to tell cabal to use certain compiler options for a particular module under certain conditions. Cheers, Bernie.

3 2

A question on existential types and Church encoding
by Cory Knapp 07 Jun '10

07 Jun '10

Hello, A professor of mine was recently playing around during a lecture with Church booleans (I.e., true = \x y -> x; false = \x y -> y) in Scala and OCaml. I missed what he did, so I reworked it in Haskell and got this: >type CB a = a -> a -> a >ct :: CB aC >ct x y = x >cf :: CB a >cf x y = y >cand :: CB (CB a) -> CB a -> CB a >cand p q = p q cf >cor :: CB (CB a) -> CB a -> CB a >cor p q = p ct q I found the lack of type symmetry (the fact that the predicate arguments don't have the same time) somewhat disturbing, so I tried to find a way to fix it. I remembered reading about existential types being used for similar type-hackery, so I added quantification to the CB type and got >type CB a = forall a . a -> a -> a >ctrue :: CB a >ctrue x y = x >cfalse :: CB a >cfalse x y = y >cand :: CB a -> CB a -> CB a >cand p q = p q cfalse >cor :: CB a -> CB a -> CB a >cor p q = p ctrue q which works. But I haven't the faintest idea why that "forall" in the type makes things work... I just don't fully understand existential type quantification. Could anyone explain to me what's going on that makes the second code work? Thanks, Cory

7 10

ANN: random-fu 0.1.0.0
by mokus＠deepbondi.net 07 Jun '10

07 Jun '10

Announcing the 0.1.0.0 release of the "random-fu" library for random number generation[1]. This release hopefully stabilizes the core interfaces (those exported from the base module "Data.Random"). Warning to anyone upgrading from earlier releases: 'Discrete' has been renamed 'Categorical', the entropy source classes have been redesigned, and many things are no longer exported from the root module "Data.Random" (In particular, DevRandom - this is not available on windows, so it will likely move to its own package eventually so that client code dependencies on it will be made explicit). Unfortunately, Hackage appears to have choked on some of the package's dependencies (specifically, 2 dependencies also depend on time, and were built using different versions) so its documentation (which I put quite a bit of work into) is not displayed on the Hackage site. In the past I have dealt with that by uploading new versions with hacks to make sure the thing builds, but I really would rather not continue to do so. Is there any procedure by which I can request a manual rebuild of the package so that its documentation will be generated and displayed? Incidentally, this is a recurring problem I have run into several times for several packages. Can we *please* come up with a way for sdist or similar to just include pre-built documentation? Or if I were to spend some time working on such a thing, would it be accepted (assuming it was done up to all applicable standards of quality)? For now, I have added some pre-built haddock docs to the repository so that they may be browsed online[2] (if code.haskell.org ever starts responding to my HTTP requests. It's just not my day today, I guess). [1] http://hackage.haskell.org/package/random-fu-0.1.0.0 [2] http://code.haskell.org/~mokus/random-fu/doc/haddock/index.html

5 12

Re: [Haskell-cafe] Issue with "connectTo" function in "network" package
by aditya siram 07 Jun '10

07 Jun '10

Doing (1) solves my problem. What did I do and why did it work? Thanks a lot for your help! -deech On 6/4/10, Roman Cheplyaka <roma(a)ro-che.info> wrote: > Confirm if any of the following solves the problem: > > 1. commenting out (by prepending putting hash sign '#' in the beginning > of the line) the following line in /etc/hosts: > > ::1 localhost ip6-localhost ip6-loopback > > 2. executing as root the following command: > > rmmod ipv6 > > * aditya siram <aditya.siram(a)gmail.com> [2010-06-04 13:51:29-0500] >> Here's my GHCI session (with my network card off): >> Prelude Network Network.Socket> connectTo "localhost" (PortNumber 3000) >> *** Exception: connect: does not exist (Connection refused) >> Prelude Network Network.Socket> connectTo "127.0.0.1" (PortNumber 3000) >> {handle: <socket: 7>} >> >> The OS is Ubuntu (Lucid) Linux. The contents of "/etc/hosts" and >> "/etc/nsswitch.conf" are: >> >> > cat /etc/nsswitch.conf >> # /etc/nsswitch.conf >> # >> # Example configuration of GNU Name Service Switch functionality. >> # If you have the `glibc-doc-reference' and `info' packages installed, >> try: >> # `info libc "Name Service Switch"' for information about this file. >> >> passwd: compat >> group: compat >> shadow: compat >> >> hosts: files mdns4_minimal [NOTFOUND=return] dns mdns4 >> networks: files >> >> protocols: db files >> services: db files >> ethers: db files >> rpc: db files >> >> netgroup: nis >> >> > cat /etc/hosts >> 127.0.0.1 localhost >> 127.0.1.1 deech-mac-ubuntu >> >> # The following lines are desirable for IPv6 capable hosts >> ::1 localhost ip6-localhost ip6-loopback >> fe00::0 ip6-localnet >> ff00::0 ip6-mcastprefix >> ff02::1 ip6-allnodes >> ff02::2 ip6-allrouters >> ff02::3 ip6-allhosts >> >> >> -deech >> >> On 6/4/10, Roman Cheplyaka <roma(a)ro-che.info> wrote: >> > * aditya siram <aditya.siram(a)gmail.com> [2010-06-03 22:30:33-0500] >> >> Hi all, >> >> I had an issue where using the connectTo [1] function would fail to >> >> connect to "localhost" if my wireless card was turned off. The moment >> >> I turned on my wireless connection it worked. But if I use connectTo >> >> with "127.0.0.1" [2] with my network off it works. >> >> >> >> I confirmed that "localhost' resolves to "127.0.0.1" by pinging >> >> "localhost" from command line with the Internet off. >> >> >> >> Is this a bug in "connectTo"? >> >> >> >> -deech >> >> >> >> [1] connectTo "localhost" (PortNumber some_number) >> >> [2] connectTo "127.0.0.1" (PortNumber some_number) >> > >> > How exactly did it fail? Did it throw an exception? If so, what was the >> > exception? >> > >> > Which operating system did you observe that under? >> > >> > If that was Linux, can you show the contents of the following files on >> > that system: >> > >> > /etc/nsswitch.conf >> > /etc/hosts >> > >> > -- >> > Roman I. Cheplyaka :: http://ro-che.info/ >> > "Don't let school get in the way of your education." - Mark Twain >> > _______________________________________________ >> > Haskell-Cafe mailing list >> > Haskell-Cafe(a)haskell.org >> > http://www.haskell.org/mailman/listinfo/haskell-cafe >> > > > -- > Roman I. Cheplyaka :: http://ro-che.info/ > "Don't let school get in the way of your education." - Mark Twain >

2 1