Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!csd4.milw.wisc.edu!cs.utexas.edu!uunet!mcvax!ukc!stl!pgh
From: pgh@stl.stc.co.uk (Peter Hamer)
Newsgroups: comp.lang.c++
Subject: Lex/Yacc of C++ -- or even BNF
Keywords: Lex, Yacc, C++
Message-ID: <1615@stl.stc.co.uk>
Date: 25 Jul 89 09:45:47 GMT
References: <5581@pt.cs.cmu.edu>
Sender: news@stl.stc.co.uk
Reply-To: "Peter Hamer" <pgh@stl.stc.co.uk>
Organization: STC Technology Limited, London Road, Harlow, Essex, UK
Lines: 24

I too would very much like a Lex/Yacc grammar of C++.

BUT, failing this, I would welcome pointers to reasonably precise descriptions
of the C++ grammar and/or or its differences from that of C. Obviously I would
also appreciate warnings about over-simplified grammars [cf K&R for C :-)].

The only information I have on the C++ grammar at the moment is in Stroustrup's
book (July 87 reprint).

BTW for anybody intending to develop lex/yacc for C++.
 1) I believe that the best source for a C grammar is Harbison & Steele.
 2) A little while back somebody posted a lex/yacc grammar for (ANSA?) C to
    this group.


Has anyone considered starting with the gnu C++ compiler, and doing one of the
following?

 1) Reverse-engineering the grammar.
 2) Modifying the parser to produce the required output.
 3) Using the parser as-is and manipulating (or unparsing) the syntax tree to
    produce the required output.

Peter