altfloat-0.3: altfloat.cabal

Name:               altfloat
Version:            0.3
License:            OtherLicense
License-File:       LICENSE
Cabal-Version:      >= 1.6
Author:             Nick Bowler
Maintainer:         nbowler@draconx.ca
Homepage:           http://repo.or.cz/w/altfloat.git
Stability:          experimental
Category:           Numerical
Build-Type:         Configure
Extra-Source-Files: configure.ac, configure, altfloat.buildinfo.in, cfloat.h
                    config.h.in examples/*.hs
Extra-Tmp-Files:    altfloat.buildinfo, config.status config.log, config.cache
                    config.h
Synopsis:           Alternative floating point support for GHC.
Description:
    A replacement for the standard Haskell floating point types and supporting
    functions.  There are a number of shortcomings which I feel severely hinder
    Haskell's utility for numerical computation.  These shortcomings include
    .
    * There is no way to sanely convert between Haskell's floating types -- not
    even between Double and CDouble.  The implementation of the 'realToFrac'
    function goes through 'Rational', which loses information as 'Rational'
    cannot represent all floating point values.
    .
    * Making floating types an instance of 'Ord' makes no sense.  'Ord' is for
    totally ordered data types, which floats are not.  As a result, a number of
    library functions (such as 'max' and 'sort') produce nonsensical results.
    .
    * The 'Enum' instance for floating types similarly makes little sense.
    While 'fromEnum' and 'toEnum' functions use 'Int' instead of 'Integer',
    limiting their usefulness, 'pred' and 'succ' can be defined in a much more
    useful way.
    .
    * Functions that should care about negative zeros, such as 'signum' and
    'abs', do not.
    .
    * Some functions, such as 'floor', have nonsensical behaviour for
    non-finite input.
    .
    * The selection of floating point library functions pales in comparison to
    C.  This problem is made worse since, as noted above, it is impossible to
    convert losslessly from 'Double' to 'CDouble' in order to use the FFI.
    .
    * There is no mechanism for handling rounding modes or exceptions.
    .
    This package is intended to address all of the above issues, and more.
    Also provided, for convenience, is an alternative to the standard Prelude
    which includes features from this library and the non-overlapping parts of
    the standard Prelude.

Source-Repository head
    Type:     git
    Location: git://repo.or.cz/altfloat.git

Flag SplitInteger
    Description: Use the new split integer packages that come with GHC 6.12.

Flag IntegerGmp
    Description: Use the integer-gmp package as opposed to integer-simple.

Library
    Build-Depends: base >= 4 && < 5, ghc-prim

    if flag(SplitInteger)
        if flag(IntegerGmp)
            Build-Depends: integer-gmp
        else
            Build-Depends: integer-simple
    else
        Build-Depends: integer

    Include-Dirs: .
    C-Sources: cfloat.c c99-compat.c
    Exposed-Modules:
        Data.Floating.CMath,
        Data.Floating.Classes,
        Data.Floating.Types,
        Data.Floating.Types.Double,
        Data.Floating.Types.Float,
        Data.Floating.Prelude,
        Data.Floating.Environment,
        Data.Floating,
        Data.Poset
    Other-Modules:
        Data.Floating.Instances, Data.Poset.Internal, Data.Poset.Instances