Related articles |
---|
LR(n) parsers whatis@ucsd.edu (Steve Boswell) (1991-10-10) |
Re: LR(n) parsers KARSTEN@tfl.dk (Karsten Nyblad, TFL, Denmark) (1991-10-13) |
Re: LR(n) parsers goer@midway.uchicago.edu (1991-10-14) |
Re: LR(n) parsers sra@ecs.soton.ac.uk (1991-10-14) |
Re: LR(n) parsers anw@maths.nott.ac.uk (1991-10-14) |
Re: LR(n) parsers bburshte@pyrps5.eng.pyramid.com (1991-10-14) |
Re: LR(n) parsers mareb@levels.unisa.edu.au (1991-10-15) |
Re: LR(n) parsers nickh@harlqn.co.uk (Nick Haines) (1991-10-16) |
Re: LR(n) parsers mtxinu!angular!jas@uunet.uu.net (1991-10-18) |
Re: LR(n) parsers rockwell@socrates.umd.edu (Raul Deluth Miller-Rockwell) (1991-10-19) |
Re: LR(n) parsers schoebel@bs5.informatik.uni-stuttgart.de (Thomas Schoebel) (1991-10-19) |
Re: LR(n) parsers drw@riesz.mit.edu (1991-10-22) |
Re: LR(n) parsers schoebel@bs5.informatik.uni-stuttgart.de (Thomas Schoebel) (1991-10-24) |
[4 later articles] |
Newsgroups: | comp.compilers |
From: | mareb@levels.unisa.edu.au |
Keywords: | parse, LR(1) |
Organization: | University of South Australia |
References: | 91-10-036 |
Date: | 15 Oct 91 18:21:42 +0930 |
In article 91-10-036, whatis@ucsd.edu (Steve Boswell) writes:
> What sort of easily-describable or commonly-occuring grammars are
> ambiguous for any amount of lookahead? I'm implementing an
> object-oriented parsing engine and I'd like to take all the work of
> resolving ambiguity away from the end programmer. ...
Some time ago I wrote a paper showing how you can build a deterministic 2PDA
by allowing limited inclusion of non-terminals in the follow sets for LR(k)
states. This technique is not particularly difficult and can parse a wide
class of grammars unambiguously. It is O(n) in efficiency, when applicable.
The parser is a generalisation of the LR(1) parser.
The paper was presented at ACSC-10 1987 and is in the proceedings.
I can send a copy to anyone interested. It is called:
Bob Buckley, "Relaxing a Rightmost-First Restriction"
Bob Buckley
mareb@lux.sait.edu.au
--
Return to the
comp.compilers page.
Search the
comp.compilers archives again.