Re: Bison Parser negate production rules

On 29 Mar 2007 00:57:34 -0400, Chris F Clark
<cfc@shell01.TheWorld.com> wrote:


>>>In fact, Yacc++ can be thought of as taking Yacc/Bison and adding in
>>>regular expressions and predicates (then cleaning up the syntax a
>>>little). In it we have a not predicate that allows exactly what you
>>>are asking for. You will find similar functionality in PCCTS/ANTLR
>>>although the underlying technology is LL not LR for them, so you
>>>cannot use left recursion (or precedence rules).


I was already aware of ANTLR's predicates and complement operator, but
it was not clear from the latest manual I had that the complement
operator could be used on token sets in the parser as well as in
regular expressions.


My fault really ... for a long time the software and manual were
separate and I tended to get a new manual only when new features were
announced.


The manual for the current 2.7.7 version does in fact say clearly that
complement sets can be used in parsers and, AFAICT, it works as
advertised. It also seems to work in 2.7.4 (the latest I had) but an
older 2.5.8 version (circa 1998?) choked badly on the example I tried
so I'm guessing that support for set complement is relatively recent.


>
>http://en.wikipedia.org/wiki/Syntactic_predicate
>


ANTLR and it's predicates are mentioned in the page, but the
complement example shows a generic PEG rather than something that
works in ANTLR.




I still think set complement is a blunt instrument. I can't recall
ever really missing it and I can't easily think of a good use for it.
I can't imagine ever wanting to specify a parser rule in the negative
as in the ANTLR example


expr: T1 (~(T1|T2|T3))* (T1|T2|T3);


because it's so very unlikely that I would ever want to accept every
other valid language token in the hole.


But I guess it's handy to know it's there if the need ever does arise.


Thanks,
George

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