This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
An XPath expression that selects elements or attributes in a particular namespace must use prefixes associated with that namespace. If you want to use identity constraints in a chameleon schema, then, the XPath expression must be modified to use prefixes which are associated with the target namespace of the including schema. See: http://lists.w3.org/Archives/Public/www-xml-schema-comments/2001OctDec/0047.html This is originally an issue against XML Schema 1.0. See bug 2066.
The WG resolved to accept http://www.w3.org/XML/Group/2004/06/xmlschema-1/structures.b2067.200701.htm as a resolution of this issue at its f2f in February of 2007 (http://www.w3.org/XML/Group/2007/01/xml-schema-ftf-minutes#friam.2067)
At the June 2007 f2f: Based on the decisions in 4419, Sandy volunteered to prepare a wording proposal that brings 2067 resolutions into alignment with 4419. This follows a request (in March) from MSM to reopen the issue since actual consensus on this issue has been hard to prove until now.
The proposal mentioned in comment #2 (http://www.w3.org/XML/Group/2004/06/xmlschema-1/structures.b2067-2.200707.html) [member-only link] was discussed and approved at today's WG telcon. It resolves this issue, which I am marking accordingly. As quasi-originator of the bug report, Jeni Tennison should be notified and asked whether this resolves her issue; this has not yet been done.
At http://lists.w3.org/Archives/Member/w3c-xml-schema-ig/2008May/0033.html (member-only link), Sandy Gao points out a problem. We changed how chameleon include works as part of 2067, but there seems to be a problem: unqualified QName references are *not* updated to point to the renamed components so they will all become dangling. (The only exceptions are the XPaths, where we allow an xpathDefaultNamespace attribute with a ##targetNamespace value.) Michael Kay proposes a solution later in the same thread. I'm reopening this issue to ensure that this gets tracked.
A wording proposal intended to resolve this issue is at http://www.w3.org/XML/Group/2004/06/xmlschema-1/structures.b4767.html (member-only link). The 2067- and 4767-related parts are intertwined, but not hopelessly; if necessary the WG can take separate action on the two issues.
The proposal mentioned in comment #5 was adopted by the WG at today's meeting. Sandy, as the originator (and re-opener) of the issue, please indicate by closing or reopening the issue whether you are satisfied with the resolution.