- Libraries - Haskell.org

Announcing a solution to the records problem
by Nikita Volkov 18 Jan '15

18 Jan '15

This thing seems to be going viral on social networks already, nonetheless here's a link for those of you, who aren't yet informed: http://nikita-volkov.github.io/record/

1 0

New maintainer for the boxes library
by Andrés Sicard-Ramírez 14 Jan '15

14 Jan '15

Hi, The current maintainers (Eelis van der Weegen and Benno Fünfstück) of the boxes library (http://hackage.haskell.org/package/boxes) don't want to be maintainers any more. Does someone want to become the new maintainer of this library? Best, -- Andrés

2 4

System.Posix.IO.ByteString
by Erik de Castro Lopo 12 Jan '15

12 Jan '15

Hi All, I'm playing around with some low level IO and found the following function: System.Posix.IO.fdRead :: Fd -> ByteCount -> IO (String, ByteCount) Ok, close, but what if I need a ByteString instead. I then found this: System.Posix.IO.ByteString.fdRead :: Fd -> ByteCount -> IO (String, ByteCount) which is identical to the one above. The ByteString in the module name suggests it would be using ByteString instead of String? Can anyone explain this for me? Cheers, Erik -- ---------------------------------------------------------------------- Erik de Castro Lopo http://www.mega-nerd.com/

6 6

Data.ByteString.Unsafe.unsafeWipe
by Erik de Castro Lopo 12 Jan '15

12 Jan '15

Discussion period: one month When handling sensitive information (like a user's password) it is desirable to only keep the data around for as short a time as possible. Specifically, relying on the garbage collector to clean it up is simply not good enough. I therefore propose that the following function to be added to the Data.ByteString.Unsafe module: -- | Overwrites the contents of a ByteString with \0 bytes. unsafeWipe :: ByteString -> IO () unsafeWipe bs = BS.unsafeUseAsCStringLen bs $ \(ptr, len) -> let go i | i < 0 = return () | otherwise = pokeElemOff ptr i 0 >> go (i - 1) in go (len - 1) It is added to the Unsafe module because it break referential transparency but since ByteStrings are always kept in pinned memory, it should not otherwise be considered unsafe. It could be used as follows: main = do passwd <- getPassword doSomethingWith passwd unsafeWipe passwd restOfProgram Cheers, Erik -- ---------------------------------------------------------------------- Erik de Castro Lopo http://www.mega-nerd.com/

6 7

Cabal: Specify multiple package databases in config file
by Jessica Hamilton 11 Jan '15

11 Jan '15

Hi All, Is it possible to specify multiple package databases in Cabal's config file? Currently, I have two installed package databases. One is the system readonly package database, and the other is the system writable package database. This structure is due to how packaging works on Haiku, so can't be changed. A similar structure would also exist for user package databases. My Cabal config file currently looks like: package-db: /packages/ghc/.self/lib/package.conf.d I need this to be something like: package-db: /packages/ghc/.self/lib/package.conf.d; /system/non-packaged/lib/ghc/package.conf.d Is this possible with the Cabal config file? If so, what is the correct syntax to use? Many Thanks, Jessica

2 1

Stream type
by Christopher Done 09 Jan '15

09 Jan '15

Is this type defined anywhere in base or elsewhere standardish? https://gist.github.com/chrisdone/516489f4f27846712225#file-type-hs-L52 data Stream a = Stream a (Stream a) Ciao

3 3

Re: Version of the pretty-printing library in GHC 7.10.1
by Andrés Sicard-Ramírez 09 Jan '15

09 Jan '15

On 9 January 2015 at 13:33, David Terei <d(a)davidterei.com> wrote: > I'd suggest GHC move to pretty 1.1.2.0. I'll shoot an email about that > to see if we can get it changed. > Thanks! -- Andrés

1 0

Proposal: removeDirectoryRecursive should not follow symlinks
by Edward Z. Yang 09 Jan '15

09 Jan '15

Discussion period: one month I propose to backwards incompatibly change the behavior of removeDirectoryRecursive to not follow symlinks. We could optionally add a replacement function under a new name which does follow symlinks. This would bring its behavior inline with rm -rf. I also opened a ticket here: https://ghc.haskell.org/trac/ghc/ticket/9266 Thanks, Edward

26 51

Version of the pretty-printing library in GHC 7.10.1
by Andrés Sicard-Ramírez 09 Jan '15

09 Jan '15

Hi, While preparing Agda for supporting GHC 7.10.1, I found the following situation: * In GHC 7.8.4, the version of the pretty library is 1.1.1.1. * In GHC 7.10.1 RC1, the version of the pretty library is 1.1.1.3. * I need to use conditional compilation when using GHC 7.10.1 RC1 because in the version 1.1.1.2 of the pretty library was added a new Eq instance which is currently defined in the Agda source code. * I cannot use the Cabal macro MIN_VERSION_package_(A,B,C) because this macro only supports three components (A, B, C) in the version number, and the change in the version of the pretty library was in the fourth component, i.e. 1.1.1.1 -> 1.1.1.2. (I could use the symbol __GLASGOW_HASKELL__ instead, but I don't want to use this symbol for changes related to the libraries.) Is going GHC 7.10.1 to be released with a version of the pretty library where I can use the MIN_VERSION_package_(A,B,C) macro? Best, -- Andrés

2 1

Discussion: Can we make evaluate easier to find?
by David Feuer 09 Jan '15

09 Jan '15

The evaluate function is defined in GHC.IO, and exported publicly by Control.Exception. This strikes me as an extremely strange place for it. Can we find another place to put it that would seem more sensible? I don't think it should be removed from Control.Exception, but rather that maybe it should be added somewhere else as well. Digging all around, I've come up with one idea that might work, if we give evaluate a more general type: evaluate :: PrimMonad m => a -> m a evaluate a = primitive (\s -> seq# a s) With this type, we could easily put it in Control.Monad.Primitive. There may be some challenges wiggling module dependencies around, but hopefully not huge ones. David

4 8