[antlr-interest] java.g parsing order for "extends"?

[Terence Parr]

On Friday, February 22, 2002, at 08:11  AM, Seth Delackner wrote:

> Suppose we have the following relationship:
>
> A.java:
> public class A {
>     public interface I {}
> }
>
> B.java:
> public class B extends A {}
>
> C.java:
>
> public class C extends B {
>     class inner implements I {}
> }
>
> Now, if I try to parse C.java, first I get a typeDefinition for 'C',
> then it starts analyzing the object block for 'C', which contains the
> declaration for 'inner'.  Shouldn't it parse B, then A, then finally get
> to A's inner block?  Or does java.g already do this, and the grammar I'm
> looking at, which is based loosely on java.g, is broken?

java.g only parses and builds a tree.  It is up to you to do symbol 
resolution if you want a compiler or whatever. :)  Typically you would 
collect a list of dependencies as you parse and then go after them or as 
you say recursively track down all dependencies on demand.

For translation, it is usually sufficient to collect a tree for each 
file separately.

Ter
--
Chief Scientist & Co-founder, http://www.jguru.com
Creator, ANTLR Parser Generator: http://www.antlr.org


 

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