[antlr-interest] ANTLR 3 & hidden token management?

[Kay Roepke]

On 7. Aug 2006, at 12:40 Uhr, Loring Craymer wrote:

> I tend to think that it is "early" rather than "late" in the design  
> process.  This is the first point at which people can comment from  
> experience with a working version of ANTLR 3.  ANTLR 3 is not yet  
> implemented in ANTLR 3; I consider that essential for a "mature"  
> implementation, and consequently see 3.1 as the first "real"  
> release.  For that matter, the parsing engine has not yet been  
> fully shaken down (it's getting close).  Ter keeps making  
> adjustments--mostly for the sake of efficiency at this point--in  
> response to user-detected issues.  There are good reasons for beta  
> testing!
>
> Most of the interface suggestions, like symbolic channels, will not  
> get implemented for ANTLR 3.0.  However, they can still be put on  
> the "to do" list and be implemented later.  Features will not be  
> placed on the "to do" list, however, unless someone points out the  
> desirability.

Again, I probably sounded like I didn't want to discuss anything that  
could be made better. This is definitely not the case. Please  
attribute it to my lack of sleep and to those plumbers that are  
tearing down my bathroom at the moment, mere 5 feet away from my head...

I am well aware of the functionality limitations 3.0 will have (no !,  
no incremental parsing, probably not a totally desirable channel  
handling, I'm sure I missed important parts here, too), but my point  
is that getting 3.0 out of the door is an extremely important to  
reach a wider audience than
a 3.0beta ever will. Plus there are a couple of projects waiting that  
are probably not going to use an ANTLR3 beta.

There are many cases where ANTLR could be made better (your paper is  
describing one of them, a very interesting read!), but I think we  
should focus on getting things working now, getting it out to users  
and then adapt later when v3 is self-hosted. Then it should be a lot  
easier to change syntax around
because all the internal grammars must be worked on anyway.

I am/was simply afraid of featuritis and then risk that the timetable  
slips ever further. That's what the original post evoked in me, and  
why I reacted rather strongly to it.

-k

-- 
Kay Röpke <kroepke at classdump.org>
classdump Software