- From: <bugzilla@jessica.w3.org>
- Date: Fri, 30 Jul 2010 10:01:03 +0000
- To: public-qt-comments@w3.org
http://www.w3.org/Bugs/Public/show_bug.cgi?id=10275 Summary: [F+O 1.1] The third argument of format-number() Product: XPath / XQuery / XSLT Version: Working drafts Platform: PC OS/Version: Windows NT Status: NEW Severity: normal Priority: P2 Component: Functions and Operators 1.1 AssignedTo: mike@saxonica.com ReportedBy: mike@saxonica.com QAContact: public-qt-comments@w3.org The third argument of the format-number() function is the name of a decimal format, supplied as a string in the form of a lexical QName. Since this is a run-time value, it would be much preferable for this argument to be an xs:QName value, thus avoiding a run-time dependency on the static context (which is messy for implementations to manage, and messy for users when code is copied-and-pasted from one module into another). The problem of course is backwards compatibility: this function has been in XSLT since version 1.0, but is new to XQuery. It would be nice to do the right thing for XQuery users while retaining compatibility for XSLT. We could tackle this by providing two signatures for the function: a "current" signature and a "legacy" signature. The "current" signature would declare this argument as xs:QName. The "legacy" signature would declare it as xs:anyAtomicValue and the semantics would require that it is either a string in the form of a lexical QName, or an xs:QName value. The host language would choose which form to offer. An alternative and perhaps simpler solution would be for decimal format names in XQuery to be NCNames rather than QNames. Because decimal formats are scoped to a module in XQuery, rather than being global, using a QName seems rather unnecessary. This approach solves the problem for XQuery, though it doesn't help us move XSLT forwards, where it would be nice to allow xs:QName values in this and other similar interfaces (for example, the first argument to key()). Incidentally, the XQuery definition of decimal formats does not say what namespace the QName is in if it is unprefixed. It should match the rule for format-number(), where no prefix means no namespace. -- Configure bugmail: http://www.w3.org/Bugs/Public/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the QA contact for the bug.
Received on Friday, 30 July 2010 10:01:07 UTC