On 2011-10-19 18:30, Sherwin Daganato wrote: > Honestly I have an ambitious plan to create a module using Moose that > will start out small and eventually evolve into something like what > Net XML Serialization[1], and JAXB[2] offer -- XML binding and > serialization, and class generation from XML schema.
Actually, class generation was one of the things I had in mind that my W3C::XMLSchema CPAN module based on XML::Rabbit could help to perform. Unfortunately I never found tuits to finish or actually design a full API for it. The xsd_structure script bundled with W3C::XMLSchema is useful when you're analyzing .xsd files to find structure, but to go from that to working code is still quite some distance away. It is also incomplete. XML::Rabbit has a dump_xml_structure script that I also like to use when I'm trying to build classes based on undocumented XML files (which I don't think you'll be doing in your project). I haven't migrated W3C::XMLSchema to the new sugar functions that came with XML::Rabbit 0.1.0. I guess that is probably the first thing that should be done before continued work is done on it. My thought was that a more complete W3C::XMLSchema implementation could be used to build proper classes for XML::Toolkit instead of the heuristics involved now (if an XSD is available). There are other XML Schema parsers available on CPAN, but none of them are based on Moose, as far as I know. When looking at the code for some of them I also found it hard to understand the code, as there was a lot of OO boilerplate in it. If you're on irc we occasionly do some XML/Moose-related discussion on #xml-toolkit on irc.perl.org. Come join us if you're interested. -- Robin
