Re: [JDEV] Namespaces vs. Attributes

2002-07-15 Thread David Waite
Ari Johnson wrote: Looking at the jabberd code, it does indeed treat these as xmlns and xmlns:stream attributes on the element. Can anyone explain the reasoning behind this method, or if a fix might be possible without severely injuring the Jabber server itself? I really don't want to have to

Re: [JDEV] Namespaces vs. Attributes

2002-07-15 Thread Ari Johnson
That's what I was afraid of, but I can deal with it, I'm sure. It's just frustrating to write something and then realize when you're done debugging it that you overlooked this kind of thing. Then again, jabberd is significantly bigger than my code thus far, and it suffers that same problem,

Re: [JDEV] Namespaces vs. Attributes

2002-07-15 Thread Kriggs
David Waite writes: In short, I don't think it is really anything that can be cleverly hacked in. Either a lot of things will break, or a lot of logic will need to be added for supporting legacy clients/servers. It has been off and on the roadmap for the open-source server, but I don't

Re: [JDEV] Namespaces vs. Attributes

2002-07-15 Thread Ari Johnson
On Mon, 15 Jul 2002, Matthias Wimmer wrote: Hi David! David Waite wrote: This (together with the 'jabber:server' / 'jabber:component:accept' namespace usage) would be rather difficult to fix at this point without breaking interoperability with nearly every client, server, and

[JDEV] Namespaces vs. Attributes

2002-07-14 Thread Ari Johnson
I am writing a distributed computing system based on Jabber, and have been building it from the ground up as I decided that no existing libraries quite suit my purpose. So I started writing an XML stream library on top of expat, and got it working just right. Namespaces work as specified by