Java Doc for TreeScanner.java in  » 6.0-JDK-Modules-sun » javac-compiler » com » sun » source » util » Java Source Code / Java DocumentationJava Source Code and Java Documentation

The default implementation of the visitXYZ methods will determine a result as follows:

• If the node being visited has no children, the result will be null.
• If the node being visited has one child, the result will be the result of calling scan on that child. The child may be a simple node or itself a list of nodes.
• If the node being visited has more than one child, the result will be determined by calling scan each child in turn, and then combining the result of each scan after the first with the cumulative result so far, as determined by the TreeScanner.reduce method. Each child may be either a simple node of a list of nodes. The default behavior of the reduce method is such that the result of the visitXYZ method will be the result of the last child scanned.

Here is an example to count the number of identifier nodes in a tree:

 class CountIdentifiers extends TreeScanner {
 Override
 public Integer visitIdentifier(IdentifierTree node, Void p) { 
 return 1; 
 }
 Override
 public Integer reduce(Integer r1, Integer r2) { 
 return (r1 == null ? 0 : r1) + (r2 == null ? 0 : r2); 
 }
 }