Add haskell-src as an official machine-readable component of the Haskell standard

[Yitzchak Gale]

Hi Neil,

Neil Mitchell wrote:
> There is nothing to stop an library author doing exactly this, and it
> might even be useful for some people (personally I'm going to stick to
> haskell-src-exts, because it's a brilliant library).

Yes, it is.

I am not proposing changing in any way how we
develop and use Haskell parsers as tools.

I am proposing taking an existing parser and using it
in a different way.

> I've found plenty of HSE/GHC parsing differences in my work,
> and my suspicion is that several of them are probably also
> present in haskell-src.

Indeed. Easily finding and documenting such discrepancies is one
of the major benefits of having a machine-readable
component in the standard.

> I also don't want the Haskell Prime Committee to take
> on the jobs of library maintainership or implementation,
> those are best kept elsewhere.

Agreed. I am not proposing that.

The machine-readable portion of the standard will be based
on a parser library that exists quite apart from the standard.

Thanks,
Yitz