[antlr-interest] Re: ANTLR 3.0 and code generation for C++

[lgcraymer]

If things go right, I will need a C or Fortran to OCaml (or Haskell) 
translator for the HTMT project. [I need first class functions and 
lazy evaluation.]  Ter's aware of that and briefly looked into the 
issues before he got going on ANTLR 3 design/development.  An OCaml 
backend for ANTLR 3 should actually not be difficult since OCaml does 
support classes; for Haskell, modules might be the way to go.

--Loring

--- In antlr-interest at yahoogroups.com, katre <katre at h...> wrote:
> lgcraymer wrote:
> > The 3.0 code generator will look very different from the 2.x code 
> > generators.  It will be driven by an output grammar, with 
> > language-specific implementations of the output templates.  
> > Retargeting output to another language should be relatively easy 
(1000 
> > LOC versus the current 150,000 LOC).
> 
> How generalized will this be?  So far, with ANTLR 2.x, code is only
> generated for OO languages (Java, C++, C#).  I know there was 
another
> output in the past (Haskell?) but will ANTLR be able to target more
> functional languages?  My specific favorite target would be OCaml, 
which
> I find I am using more the more I get to know it, but the included 
lexer
> and parser generators are basically translations of lex and yacc and
> lack a lot of features I got used to when using ANTLR.
> 
> katre


 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/