[antlr-interest] brief analysis of java.g's tree building in 2.x vs proposed 3.0

[John D. Mitchell]

>>>>> "Loring" == Loring Craymer <Loring.G.Craymer at jpl.nasa.gov> writes:
[...]

>> Heh, what if we used -> like this:

>> ifStat : (IF->IFEXPR) '('! expr ')'! stat ... ;

YUCK!  I'd much prefer the verbose but very clear version.


>> Then it begs the question: can I allow a tree rewrite not just a token
>> type change on a single node?  Nah...then I would really be getting
>> complicated. ;)

> This gets pretty ghastly.  You need an attribute syntax, with nodes
> having default attributes "text" and "type".  Then something like IF
> @IF.type = IFEXPR might make sense.

That looks pretty complicated, too, as (it seems like) you're driving
expressions with side-effects into the mix.

Take care,
	John