>I guess it depends how deep/complex the XML tree goes. I am in a >similar situation. From my previous use of Java my preference would be >to build an array (or some other collection-type structure) of objects >from the XML tree. Performance should be better actually keeping it in a tree, if the data is complex. It certainly will take as much if not more memory to store it in an array. As XML support gets integrated into future versions of Revolution, it can also serve as a way to store and iterate through complex structures.
libxml which is what revxml is built on top of, can access individual nodes and iterate through the tree extremely quickly as is demonstrated by the xml-tree view sample. It supports the sax and dom.The only reason I can think of dumping the xml tree is if your reading a large document and don't need the majority of data in it. In this case I would advise using the sax based callback mechanism (revxmlstarttree,revxmlendtree, ect) to populate an array, or whatever with just the stuff you data you want to keep. Tuviah Snyder <[EMAIL PROTECTED]> <http://www.runrev.com/> Runtime Revolution Limited - Software at the Speed of Thought _______________________________________________ use-revolution mailing list [EMAIL PROTECTED] http://lists.runrev.com/mailman/listinfo/use-revolution
