It occurred to me that it would be pretty easy to allow nodes of the expression tree to be named, and then the Jython code built would introduce named variables, making the code easier to read. Also it's about time to make a new visualization of the tree, instead of using a JTree.
result=... could be used to indicate that the last expression should not be used and the node visualized. I often want to look at intermediate nodes when I need to debug the expression.