> ----- Original Message ----- > From: "Armando Anton" <[EMAIL PROTECTED]> > > yes i suppose the base URL can be returned :)
Nodding... > i supposed that changing the URI constructor to can create an empty URI was necessary because when resolving relative uris, the base and the relative URI objects are created first, but a check can be done in the (URI,String) constructor easily Hmm... An empty URI means no resource. So it's no proper I guess. If it's relative, it's working on a resource. So only URI(URI, String) is the proper one to be fixed. > Armando > > -----Original Message----- > From: Mike Moran [mailto:[EMAIL PROTECTED]] > > Armando Anton wrote: > > >Hi! > > > >I was working in uri test cases too so i send some more cases testing abnormal examples (you can find them in the RFC) :) > >Most of the abnormal examples works well except two of them: > > > >1. when the relative uri is empty > an exception is thrown (i think it is a good idea to avoid empty uris) > > > [ ... ] > > I don't know if this is relevant to the usage within HttpClient, but, if > someone was using this class to fetch/process URLs extracted from a web > page then this behaviour would cause breakage on links like <a href="">. > Similarly, many form tags use an empty URL to mean `submit to this > page'. Can't you just return the base URL? You mean that an anchor refers to a current resource. That means that the anchor is a relative one. So I agree with you. Sung-Gu --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
