Re: FWD:Re: refactoring

David Abrahams wrote:
> Stefan Seefeld <[email protected]> writes:
> 
> 
>>Hi David,
>>
>>I hope we can sort out the current mail dilemma !
>>
>>Meanwhile:
>>
>> (Dave Abrahams) wrote:
>>
>>>Stefan Seefeld <[email protected]> writes:
>>
>>>>There is still a lot of code in the C++ parser module that needs to
>>>>be cleaned up, and I plan to shortly build an alternative python
>>>>extension module for C++ parsing that uses the new symbol lookup
>>>>(and overload resolution, which I'm working on now), instead of
>>>>the old and rather inefficient code.
>>>
>>>Wow, that's great!  Are you really implementing the C++ overloading
>>>and name lookup rules, including all the template stuff?  This is an
>>>unbelievably complicated area.
>>
>>I'v been working on symbol resolution, but without any type analysis,
>>i.e. there is no overload resolution yet.
>>While I would definitely like to work on all this, I'm not sure
>>how much I can do single-handedly, as I realize just how complicated
>>all this gets. I would certainly appreciate any help on any level.
> 
> 
> This is essentially the problem of building a C++ compiler front-end.
> I know some people who might be very interested in working on such a
> thing with you.
> 
> 
>>In particular, I'm looking for an elegant design that helps to generate
>>a representation for C++'s type system, and which can then be used
>>to obtain information about type relationships...
> 
> 
> It would be wonderful to have such a thing... in C++!  Too many C++
> compilers are written in unexpressive C.

Hi,

Hartmut Kaiser noted that Stefan intends to use the Wave C/C++
preprocessor. Wave started out as a challenge I posed that Hartmut
took seriously. Hartmut, with lots of help and advice from Paul
Mensonides (the CPP guru), got it going strong. Prior to Wave,
there were plans for a Spirit based C++ parser. In fact, there
were some early attempts, but were not quite there yet partly due
to lack of a good preprocessor. With a good lexing preprocessor,
the next logical step is the C++ parser. I'd be thrilled in any
collaboration.

WRT templates and the C++ type system, my current thinking is that
I need an embedded runtime interpreter specifically geared to
handling the C++ meta template language. It would be a minimalist
FP language similar to scheme/lisp. I implemented one called "Rave"
last year which turned out to be a runtime-dispatch equivalent of
the Phoenix library with dynamic types, lambda and all those stuff.

Regards,
-- 
Joel de Guzman
http://www.boost-consulting.com
http://spirit.sf.net