[antlr-interest] Re: proposal for 2.7.4: charVocabulary defaults to ascii 1..127

Mike Lischke lists at lischke-online.de
Sat May 1 14:21:50 PDT 2004


> That  begs the question, "why not a table with a mapping"?  

Not as default IMO. But taking that path for an explicit charVocab would be a nice addition. Let the general case be
just Unicode and require the grammar writer to specify if (s)he needs a smaller set.

> Somebody proposed doing 
> this properly for letters and digits in various languages 
> etc... 

That was me.

>i had planned on doing that for 3.0, but certainly 
> something trivial like ASCII and UNICODE would work nicely 
> for the overall vocab, right?

I agree. Assuming Unicode as default vocabulary would allow for quite some optimizations. For instance checking for a
certain character class (which determines whether a code point is a number, identifiert part etc.) is quite simple.

Mike
--
www.soft-gems.net



 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
     http://groups.yahoo.com/group/antlr-interest/

<*> To unsubscribe from this group, send an email to:
     antlr-interest-unsubscribe at yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
     http://docs.yahoo.com/info/terms/
 



More information about the antlr-interest mailing list