We present a streaming algorithm for evaluating XPath expressions that use backward axes (parent and ancestor) and forward axes in a single document-order traversal of an XML document. Other streaming XPath processors handle only forward axes. We show through experiments that our algorithm significantly outperforms (by more than a factor of two) a traditional nonstreaming XPath engine. Furthermore, our algorithm scales better because it retains only the relevant portions of the input document in memory. Our engine successfully processes documents over 1GB in size, whereas the traditional XPath engine degrades considerably in performance for documents over 100 MB in size and fails to complete for documents of size over 200 MB.

Tools for the generation of code from model descriptions are valuable in helping developers maintain consistency between a model and its implementation. In this paper, we describe a new method for generating Java implementation code from UML diagrams. Our method accepts UML diagrams specified at a higher-level than current tools, and places fewer constraints on the supported UML constructs, like multiple generalizations and association classes. Unlike current tools, it generates implementation code that shields system implementers from the low-level details of how associations and other UML constructs are mapped to Java. Furthermore, it supports the modular design of systems according to concerns[2, 10] by being able to generate code from a set of related UML diagrams. While our discussion is focused on the special problem of generating Java implementation code, the issues discussed in this paper are applicable more generally to object-oriented implementation languages.

Mockingbird is a prototype tool for developing interlanguage and distributed applications. It compiles stubs from pairs of interface declarations, allowing existing data types to be reused on both sides of every interface. Other multilanguage stub compilers impose data types on the application, complicating development. Mockingbird supports C/C++, Java, and CORBA IDL, and can be extended to other languages. Mockingbird can generate stubs that convert between types whose structural equivalence would be missed by other tools. We show that this kind of tool improves programming productivity, and describe, in detail, Mockingbird's design and implementation.