Why Perl needs a VM

[Matt S Trout]

On Wed, Sep 05, 2007 at 10:08:35AM -0700, Simon Cozens wrote:
> Matt S Trout wrote:
> >But assuming we had that, and I think we should just on principle if 
> >nothing
> >else, other backends would become much more easily possible. One I'd love
> >to see is Bigloo Scheme, since that compiles to C, JVM and MSIL already.
> 
> Are you aware that we already have Perl modules which allow you to write 
> compiler backends that traverse the AST of a Perl program? There are 
> even two backends to C already, which you could use as a basis for one 
> to Scheme, rather than reinventing that wheel.
> 
> Although wheel reinvention is indeed fun. :)

I am indeed, but unless I've missed something those traverse the optree and
that's at least to some extent specialised to the interpreter. In any case,
the intermediate layer I'm talking about would go to and from the optree since
that's how I'm intending to get macros going. (and clkao's already working
on something that's pretty much the intermediate layer I want for his decompile
to JS stuff)

-- 
      Matt S Trout       Need help with your Catalyst or DBIx::Class project?
   Technical Director    Want a managed development or deployment platform?
 Shadowcat Systems Ltd.  Contact mst (at) shadowcatsystems.co.uk for a quote
http://chainsawblues.vox.com/                    http://www.shadowcat.co.uk/