Looking through the source, I notice a lot of the methods everywhere (in GeomContainer for example) are completely empty and some just throw an unsupported function error. I checked to see if any of these dummy methods are implemented in subclasses of GeomContainer but from what I can see, that’s all they are – dummy methods. I don’t see how this is supposed to be beneficial. Wouldn’t it be better just to delete those lines so that the error would show up as a compile error (method isn’t there)? That would be a lot better than having a bunch of useless methods taking up space in the API and confusing newbies like me. At one point I was trying to use one of these dummy methods for about and hour or two before checking the Xith source to realize the method was completely empty. It’s hard to figure out which method to use because there are so many versions of similarly named methods, and yet many of them have absolutely no use whatsoever, except to confuse. They should simply be added in when they are implemented. Or is there some fundamental API design issue I have overlooked…?
Oh yea, and personally I think the most important thing to improve on in Xith right now is the Javadoc. It’s practically non-existant right now…
Another oh yea… Does anybody know of any good Xith guides besides the one on the offical xith.org site? Didn’t find any with Google…
