[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Ecrit] Profiles (was: Consensus Call) - namespaces



I'm somehow missing the problem. All namespaces, and thus, the corresponding schema, are already identified in the XML instance, for GML and the other LoST components. What additional information would I need to provide? Why would I need to glue together the schema?

As far as I know, a single XML document can have components from multiple schema. This seems common for XML documents, including the work in XMPP or NetConf. Neither, as far as I know, has the notion of a profile.

On Apr 26, 2007, at 12:45 PM, Andrew Newton wrote:

So you are advocating the creation of a schema that glues the two different schemas together, so it can be validated via the XML namespace? If so, ok. But that is just one step further than a profile identifier.

-andy

On Apr 25, 2007, at 11:14 PM, Henning Schulzrinne wrote:

The point I'm making is that the namespace identifier is sufficient to determine what XML is valid, since you can determine the corresponding schema, whether expressed in RelaxNG or XML Schema. We are not using DTD.

Thus, I fail to see what this has to do with another label. We don't have profile labels in SIP presence, for example, and the problem is pretty much the same.

On Apr 25, 2007, at 9:47 AM, Andrew Newton wrote:

XML validity and XML namespaces are separate features. XML validity, in the form of DTDs, existed before XML namespaces. It just so happens that the more modern XML schema languages, such as Relax NG and XML Schema, account for XML namespaces and allow the validity to check the namespace, which wasn't possible with DTDs.

However, validation checking will not solve this problem, as each XML fragment is valid against its schema. This means such validation must be done by the application.

-andy




_______________________________________________
Ecrit mailing list
Ecrit at ietf.org
https://www1.ietf.org/mailman/listinfo/ecrit