On Fri, 2007-05-11 at 14:47 -0400, Jacques Carette wrote:
> But I don't understand -- Maple is free form too, yet I can write the
> equivalent of
> module Act = class Base = fun clone () = stuff end fun end class end module;
> and it parses just fine. Ok, so Maple is old-style and this would be
> Act := module() Base := class() clone := proc() stuff end fun end class
> end module;
> but it still only needs the one ;
Felix uses an unambiguous LALR(1) grammar.
fun clone () = stuff end fun clone2() = stuff end fun
can certainly be parsed correctly .. but not by an unambiguous
LALR(1) grammar with a parser like Ocamlyacc.
Dypgen or Elkhound could do this since they're GLR, they
work by spawning new threads on each ambiguity, so here
(see the . in the text):
fun clone () = stuff end . fun clone2() = stuff end fun
a thread is forked off to start a new fun, and the old one
reduced, AND, a thread assumes the 'fun' is 'end fun' and
reduces that then continues.
The latter thread may actually parse 'clone2() = stuff' as
an assignment if Maple allows that .. and fail when
it finally hits the unmatched 'end'.
You could also do this with a backtracking Earley parser.
However you can't parse it with recursive descent.
So I'm actually curious how Maple can do this, since it is
older than GLR technology.
--
John Skaller <skaller at users dot sf dot net>
Felix, successor to C++: http://felix.sf.net
-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Felix-language mailing list
Felix-language@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/felix-language
