Hi, i recently applied to join the gsoc program and would like to introduce myself and the proposed project.
I work as software consultant with a focus on OOAD, Java, JEE, Spring with and study Business Informatics at the University of Business Administration in Vienna, Austria. When dicussing Xml OO Mapping we found out that there could be a need to map only certan parts of a (big) xml document. Gsoc Project - Castor Pull Mapping Currently Castor works in the way that it maps 'complete' XML documents into an object tree and hands back the root of it to the caller. For really large documents that are of a 'flat' nature another approach would be better. The idea is to define a certain XPath which is used as root for mapping of certain document parts and a callback mechanism to provide the mapped documents parts to the caller. Doing so helps to reduce the memory footprint during processing and also allows that document parts are handed to the caller without having parsed (and/or mapped) the complete document. Proposed solution * Castor Pull Mapping is implemented on top of the existing implementation * Relevant Structures are found with XPATH * Substructures are extracted and handed over to Castor for mappig * A callback is provided for each mapped object and/or * the relevant, mapped Objects are returend as collection no worries, Bernhard
