[antlr-interest] Profiler 2

[Martin Probst]

Hi,

> > Well, the ANTLR v2 generated stuff is not the fastest in the 
> > world at  
> > the moment but the tool is more popular than many other tools even  
> > so--I'm not sure we have a huge problem.
> 
> My bad. I should have mentioned that I was addressing the ~20% of potential
> users/applications for whom speed matters at least as much as ANTLR's
> productivity benefits. Ruslan is a member of that group...

I consider myself being part of that group, too. However we don't have
any performance problems using Java ANTLR with a handwritten Lexer and
no ()=> at all. Or at least no real problems - faster is always better,
but our parser munges through all queries in a couple of ms.

So it's definitely possible to get good performance out of ANTLR, at
least in Java. I used it in C++ for some time too, and results were less
favorable, app. a factor 10 slower than a comparable LALR solution.

Martin