Re: Namespace of xmlns attribute

> You've got no prefix, but its in a > namespace, which is inconsistent and requires special casing. > Why couldn't they have just said that if you see an unprefixed attribute named > xmlns, that's all the signal that you need. That would save special-casing during DOM construction, but add spec

RE: Namespace of xmlns attribute

>So when adding namespace declarations through the DOM, you should manually >set the namespaceURI to "http://www.w3.org/2000/xmlns/". Correct? Yep. See the description of when createAttributeNS throws a NAMESPACE_ERR DOMException. ---

Re: Namespace of xmlns attribute

"has an Attr with nodeName = "xmlns" prefix = null (unspecified) localName = "xmlns" namespaceURI = "http://www.w3.org/2000/xmlns/" nodeValue="xxx"" Is the part that I think is bad. You've got no prefix, but its in a namespace, which is inconsistent and requires special c

RE: Namespace of xmlns attribute

> Also note that if you insist on using DOM Level 1 factory methods to build > the nodes, you wind up with DOM Level 1 nodes -- the nodename gets set, > none of prefix, namespaceURI, or localName will. Intermixing level 1 and > level 2 nodes is _NOT_ recommended. So when adding namespace declarat

RE: Namespace of xmlns attribute

>Do you want to boil that down to the raw facts, >just so that there is no confusion? Sorry; overcompensated writer's block ... If you're using the DOM Level 2 namespace-aware calls to build and probe the DOM, and assuming I'm awake, this should work out as follows. > > > Element n

RE: Namespace of xmlns attribute

PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Tuesday, April 03, 2001 7:44 AM To: [EMAIL PROTECTED] Subject: RE: Namespace of xmlns attribute >The tricky part which I didn't see in >the DOM spec was how would prefix and localname work. I would assume that >the default namespace decl would

RE: Namespace of xmlns attribute

(BTW, apologies if I seem to be convolving a SAX question with a DOM question -- my citations of the DOM's solution were just intended to give the perspective of someone who's already had to wrestle this particular alligator. ) ---

RE: Namespace of xmlns attribute

>The tricky part which I didn't see in >the DOM spec was how would prefix and localname work. I would assume that >the default namespace decl would have a null or zero length local name. The DOM's intent was that the xmlns= default declaration would have localname "xmlns", no prefix, and the nam

RE: Namespace of xmlns attribute

--- From: David E. Cleary [mailto:[EMAIL PROTECTED]] Sent: Friday, March 30, 2001 6:21 AM To: [EMAIL PROTECTED] Subject: RE: Namespace of xmlns attribute > its a prefix that maps to the xmlns URL. In the former, if you go by XML > rules, and I don't see why you shouldn't, xmlns=&

RE: Namespace of xmlns attribute

> its a prefix that maps to the xmlns URL. In the former, if you go by XML > rules, and I don't see why you shouldn't, xmlns="" is an > attribute just like > any other attribute and its in the global namespace and I think > it should be > treated the same as any other such, to avoid having to have

RE: Namespace of xmlns attribute

are Geek Extraordinaire >Portal, Inc >[EMAIL PROTECTED] > > > >-Original Message----- >From: Chris Hill [mailto:[EMAIL PROTECTED]] >Sent: Thursday, March 29, 2001 2:13 PM >To: [EMAIL PROTECTED] >Subject: Namespace of xmlns attribute > > >Xerces 1

RE: Namespace of xmlns attribute

ario specially, and force everyone else to as well. -- Dean Roddey Software Geek Extraordinaire Portal, Inc [EMAIL PROTECTED] -Original Message- From: Chris Hill [mailto:[EMAIL PROTECTED]] Sent: Thursday, March 29, 2001 2:13 PM To: [EMAIL PROTECTED] Subject: Namespace of xmlns at

Namespace of xmlns attribute

Xerces 1.4 puts the xmlns attribute in a default namespace declaration (such as ) in the global namespace, like all other attributes without a prefix. I see that a couple other specifications place it in the "http://www.w3.org/2000/xmlns/" namespace, as well as the document at http://www.w3.or