patch applied (ghc): Change the representation of the package database

[Duncan Coutts]

On Fri, 2009-09-11 at 22:52 +0100, Ian Lynagh wrote:
> On Fri, Sep 11, 2009 at 08:24:36PM +0000, Duncan Coutts wrote:
> > On Fri, 2009-09-11 at 18:38 +0100, Ian Lynagh wrote:
> > > On Fri, Sep 11, 2009 at 08:59:27AM +0100, Simon Marlow wrote:
> > > > On 10/09/2009 21:16, Duncan Coutts wrote:
> > > >> On Thu, 2009-09-10 at 05:02 -0700, Simon Marlow wrote:
> > > >>
> > > >>>     - the binary package is now a boot package
> > > >>
> > > >> I hope it's not a registered package. We don't want to pre-empt the
> > > >> decision about including binary in the Haskell Platform.
> > > >
> > > > There are several such packages (haskeline, utf8-string, and now binary  
> > > > and bin-package-db).  I don't know how we handle the not-installing part  
> > > > - they all seem to be installed by make install.  Ian, what's the plan 
> > > > here?
> > > 
> > > The ghc package depends on bin-package-db, which in turn depends on
> > > binary, so we don't have much choice but to install them.
> > 
> > Arrg. The binary package authors (let alone anyone else) are not happy
> > with the package going into the platform in its current state.
> > 
> > In the worst case you can just bundle a copy of the binary code
> > privately since I presume it's not exposed in the ghc package's API.
> 
> Well, it exposes Binary instances of various types. And AFAIK the only
> reason it isn't yet used more in GHC is concerns over performance.

That's fine, people just will not be able to use those instances
(because the class will be in a non-exposed module).

Duncan