[antlr-interest] Re: Building trees with the correct associativity

[lgcraymer]

Sorry--

I miswrote the second case which should have been

addExpr
    :
    mulExpr  (addOp addExpr)?
    ;

and you can do the rewrite within the ()?.

Out of curiosity, why are you avoiding the more compact

addExpr
    :
    mulExpr (addOp^ mulExpr)*
    ;

The binary trees avoid inserting the extra node.

--Loring


--- In antlr-interest at yahoogroups.com, "Paul J. Lucas"
<pauljlucas at m...> wrote:
> On Thu, 2 Dec 2004, lgcraymer wrote:
> 
> > addExpr:
> >     :
> >     mulExpr (addOp mulExpr)*
> >     ;
> 
> 	This elides the tree-building code, a la:
> 
> 		## = #([ADD_EXPR,"ADD_EXPR", ao, e1 e2);
> 
> 	If I add that in inside the ()*, the ## gets its value
> 	overwritten (as Terence noted and I subsequently corrected).
> 
> > addExpr
> >     :
> >     mulExpr (addOp addExpr)*
> >     ;
> 
> 	This is what I actually have now (as I corrected myself in a
> 	previous e-mail).  This doesn't work.
> 
> 	I want an ADD_EXPR to be present in the tree only if an
> 	expression actually contains an addOp thus prodicing "compact"
> 	trees without lots of useless intervening nodes.
> 
> 	- Paul





 
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/