4348 2007-02-23 19:08:46 +0000 Remove single-ID-attribute rule 2007-05-02 01:06:14 +0000 1 1 1 Unclassified XML Schema Structures: XSD Part 1 1.1 only Macintosh All CLOSED FIXED resolved P2 normal --- 1 cmsmcq cmsmcq www-xml-schema-comments oldest_to_newest 14082 0 cmsmcq 2007-02-23 19:08:46 +0000 Like XML Schema 1.0, and like XML DTDs and SGML DTDs before it, XML Schema 1.1 forbids schema authors to declare more than one attribute of an element as having type ID. This restriction never made much sense, and it's currently getting in the way of users as various as the HTML Working Group and the Web Services Policy Working Group, as well as anyone else with an existing XML vocabulary with defined ID attributes who would like to make it possible for authors to use xml:id in addition to, or instead of, their existing ID attributes. As we have done in other cases where we have decided not to enforce restrictions imposed by our predecessor languages, we should add a note mentioning that declarations with multiple ID attributes can't be translated into DTD notation without loss. But we should then drop the current restriction and allow multiple ID attributes on a given element. 14310 1 fsasaki 2007-03-05 04:10:55 +0000 From my experience in the WS Policy WG I can agree with Michael that having no "single-ID-attribute" rule would be very useful. See http://www.w3.org/2007/02/ws-policy.xsd for an example of the need for this functionality: The <Policy> element contains an attribute extension point <xs:anyAttribute namespace="##any" processContents="lax"/>, since the specifications needs to allow for Policy identification ( http://www.w3.org/TR/ws-policy/#Policy_Identification ) with xml:id or the wsu:Id attribute. It would be good to get rid of this extension point and just have two attribute declarations. 14727 2 sandygao 2007-04-13 16:56:49 +0000 During the 2007-03 F2F meeting, the working group resolved this issue by approving a proposal that removed the "single-id-attribute" rule. Marking this issue as "fixed" accordingly. 14969 3 cmsmcq 2007-05-02 01:06:14 +0000 As the originator of this comment, I confirm that I am content with the resolution adopted by the Working Group.