[antlr-interest] Re : Re : thoughts on ANTLR v4's use of interpreters

[Stanislas Rusinsky]

Indeed you can always take a few minutes to walk the graph, but there is always 
the option to spend weeks or even months to make that step automate-able :)

Will v4 interpreter-interpreter also generate a parse tree?


----- Message d'origine ----
De : Terence Parr <parrt at cs.usfca.edu>
À : antlr-interest Interest <antlr-interest at antlr.org>
Envoyé le : Mer 20 avril 2011, 22h 23min 33s
Objet : Re: [antlr-interest] Re : thoughts on ANTLR v4's use of interpreters


On Apr 20, 2011, at 12:43 PM, Stanislas Rusinsky wrote:

> This one made my day :)
> 
> How easy will it be to hook external code or reference a position in the 
> PDA/DFA/NFA to observe passage and generate intellisense like functionality? 

That's a good question. I think it will be easy. I will have a pointer to an 
augmented transition network and memory with functions to return the next set of 
tokens etc.  Or, you can simply walk the graph to figure out what you would like 
to do yourself :)

I have tested, just yesterday, and interpreted lexer and parser for simple HTML. 
works really well, including calling out to some actions :) I'm having some evil 
thoughts about parsing at the moment. more later when I have time to formalize 
and try things out.

All indications are that this approach will yield lots of interesting things. 
You can expect such a beast to be in v4.

Ter

List: http://www.antlr.org/mailman/listinfo/antlr-interest
Unsubscribe: 
http://www.antlr.org/mailman/options/antlr-interest/your-email-address