This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
The test modules-15 has an "expected-error" of XQST0088. That is reasonable, but I'm not sure it follows from the spec. err:XQST0088 is: It is a static error if the literal that specifies the target namespace in a module import is of zero length. But the module import specifies a non-zero-length namespace. The namespace is then mapped in the "module datebase" (i.e. XQTSCatalog.xml) to TestSources/test2-lib.xq, which starts out: module namespace test2=""; I.e. a zero-length namespace in the *module declaration*, not in the module import. There is no explicit requirement in "4.2 Module Declaration" that the namespace by non-zero-length. There may be an implicit requirement, of course. If we go by the spec, the error should perhaps be XQST0059, since in "4.11 Module Import" we have: "It is a static error [err:XQST0059] if the implementation is not able to process a module import by finding a valid module definition with the specified target namespace." Or fix the specification.
The meaning of error code XQST0088 was extended by the resolution of bug #2460. Michael Kay
Moreover, the resolution of bug #2460 appears to be correctly reflected in the 8 June 2006 edition of the spec.
(In reply to comment #2) > Moreover, the resolution of bug #2460 appears to be correctly reflected in the > 8 June 2006 edition of the spec. So it does. My apologies. In my defense let me point out that http://www.w3.org/XML/Query/ lists the specification as Candidate Recommendation 3rd November 2005 and no news item lists the June edition(s). Perhaps that could be fixed? Should I resolve this bug, marking it as "fixed" or perhaps "invalid"?
I'm going to mark this as invalid. I've requested that our public page reflect the new drafts of our CR documents. Anticipating your acceptance of this resolution, I will also mark this bug report closed.