I would not do that by using a tree based parser like SAX. The parser has to build an object tree in the memory at runtime. So with big xml files this approach can cause an heavy usage of the servers memory resources. The second disadvantage, on the countrary to event based parsers, is the processing time of a xml document caused of its xml element to object mappings. This way the object relations also has to be handled. On the other hand by using a event based parser like expat u have to define very tricky algorithms to process rdf files, because rdf files enables node relationships between nodes spreaded withing the xml document (see: rdfs:subClassOf). And this by filling stacks with recursive functions is not so funny, I guess;).
Cheers, mike
-------------------------------------------------------------------------------------------
join #ezpublish on irc.freenode.org
You must be logged in to post messages in this topic!