[antlr-interest] how useful would a generic grammar "action" language be?

[Matt Benson]

Okay, also, Ter, you had mentioned using an ANTLR 3
intermediate form for all kinds of goodies... I am
trying to have an idea but I'm too dumb for it.  :)  

We know we have to retain a means of generating
actions even when they are too complex to be specified
in the "action language."  Does it make sense to have
the concept of a target language-specific
implementation of some kind of ANTLR event listener
interface, where the specification is open enough not
to be restricted to OO target languages?  I'm not sure
what the events would consist of; possibly a string
representation of the ANTLR intermediate form... or to
bring the whole thing full circle, a
target-language-specific representation of the
intermediate form, generated according to the same
principles as we are discussing overall?

Does any of that make sense?

-Matt


--- Oliver Zeigermann <oliver at zeigermann.de> wrote:
> Terence Parr wrote:
> > Yeah, I'm semi-skeptical but it's a nice dream. 
> Perhaps it will only 
> > be good for sem preds?
> 
> For sure, because sem preds are part of the analysis
> part of the 
> grammar. Maybe you could think of completly removing
> all other synthesis 
> actions (except for sem preds and setting of
> conditions for them of 
> course specified in your pseudo-language) from the
> grammar and find 
> another way of specifying semantic actions outside
> of it. I know sableCC 
> elaborated some concepts how to do this.
> Unfortunately, I have no deeper 
> insight into this...
> 
> Oliver
> 
> 
> 
>  
> 
> Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/ 
> 
> 


__________________________________
Do you Yahoo!?
Exclusive Video Premiere - Britney Spears
http://launch.yahoo.com/promos/britneyspears/

 

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