From: Philipp Kempgen <lists@...>
Date: 2010-08-20T04:56:57+09:00
Subject: Re: Parsing, BNF, TreeTop, GhostWheel, ...

Clifford Heath wrote:
> Yes. PEG parsers are greedy (iterate as far as possible and never
> backtrack on that), and also take the first alternative that succeeds,
> without trying subsequent alternatives. It's actually quite useful
> and natural, once you get the hang of it.
>
> Pure PEGs have no lookahead, as Treetop does. It's often necessary
> though, in the light of the above.

What I wanted to do is to construct/generate a parser from a bunch
of (A)BNF rules (dozends) from an RFC so backtracking is a
requirement while lookahead is not. The grammar specification for
the parser (/ parser generator) does not need to be exactly like BNF
but should not require me to think about where lookahead is needed
as this manual conversion can be error-prone.
That seems to rule out PEG parsers.

Clifford Heath wrote:
> Christian Surlykke wrote:
>> You may want to look at: http://kanocc.rubyforge.org/
> 
> Actually, I think Citrus does a better job than Treetop (of which I'm primary maintainer and a serious user). It provides a BNF built on ruby functions and operators, and also a Treetop/PEG-like parser for grammar files as an alternative. A brief look at Kanocc indicates Citrus may be preferable over it also.
> 
> Citrus doesn't memoize by default however, you have to enable that. It also doesn't have support for semantic predicates, an important feature (for me) which I added to Treetop.

> <http://github.com/mjijackson/citrus/>

Thanks a lot for the pointers Clifford and Christian!
Will have a look as soon as time permits.


  Philipp