Re: Grammar with precedence rules

In article 02-03-068, joachim.durchholz@gmx.de wrote:
>> When dealing with different parsing algorithms, one would like to have
>> a language specified by a pair (G, P), where G is a traditional
>> grammar, and P is a suitably defined set of precedence rules. Then
>> from that, one should be able to define the language L(G, P), without
>> any dependency on a specific parsing algorithm.


>> Has this been done (if so, ref's, please)?


>The Dragon book has a set of rules that say when a precedence grammar
>is unambiguous (for a quite wide definition of "precedence grammar",
>i.e. with few restrictions on the form of the productions). This is
>almost certainly not what you want, but it may give you a new angle to
>view the problem from.


Right, this not what I want:


The book http://www.cs.vu.nl/~dick/PTAPG.htm has an even more detailed
description of precedence grammars, but one of its authors told me
that neither he knows how to do it. (Even though everybody seems to
agree on that it should be doable, nobody seems to know how or be able
to give a reference.)


    Hans Aberg * Email: Hans Aberg <remove.haberg@member.ams.org>
                                    * Home Page: <http://www.matematik.su.se/~haberg/>
                                    * AMS member listing: <http://www.ams.org/cml/>

Return to the comp.compilers page.
Search the comp.compilers archives again.