From: Stefan B. <bo...@ap...> - 2009-06-09 10:54:36
|
On 2009-06-08, Maxim Filimonov <tpa...@gm...> wrote: > Hi Stefan.I'm thinking more about xslt processors there a couple of them, > but for validation huh. Currently I don't plan to do more WRT XSLT than providing the inut builder method that applies a transformation to a different source. > Ok can you give me you next planned steps to be clear what i can > start to implement? API-wise I'd probably look at XPath next since the difference engine is a complex beast and I'd like to feel my way around it first. >From the implementation POV validation isn't really done, yet. The "old" XMLUnit validation code can do a few tricks we don't support in XMLUnit2 for now: * user specified resolvers for entities and/or URIs. This is important IMHO, we need to figure out a nice API for it, though. A setter or two might be enough, not sure. * validation against a DTD for documents with no DOCTYPE declaration at all or even against a different DTD than the one of the DOCTYPE. Not sure how many people use the feature, but I'd like to at least keep the DOCTYPE-less use-case. * tests * some minor API details like the type of exception we want to throw if the user uses validateSchema on a DTD (we don't know how to do that). Stefan |