1458 2005-05-17 14:14:34 +0000 XQST0082 should be a dynamic error not a static error 2005-07-28 18:35:19 +0000 1 1 1 Unclassified XPath / XQuery / XSLT XQuery 1.0 Last Call drafts PC Windows XP CLOSED DUPLICATE 1570 P2 normal --- 1 mike chamberl public-qt-comments oldest_to_newest 3280 0 mike 2005-05-17 14:14:34 +0000 This relates to attribute constructors for the xml:id attribute within a direct element constructor. List item 5 in 3.7.1.1 reads: If the attribute name is in namespace http://www.w3.org/XML/1998/namespace and its local part is id (informally, if the attribute name is xml:id), the string-value of the attribute must be in the lexical space of the type xs:NCName [err:XQST0082], and the is-id property of the resulting attribute node is set to true; otherwise the is-id property is set to false. In general it's not known until run-time whether the value of the attribute is a valid NCName, so this should be a dynamic error not a static error. Michael Kay 3281 1 mike 2005-05-17 14:19:46 +0000 Also affects 3.7.3.2 Computed Attribute Constructors 3283 2 mike 2005-05-17 14:30:44 +0000 I also note that XSLT explicitly says there are no constraints on the value of an xml:id attribute, whereas XQuery says it's an error if it isn't an NCName. I can't see any good reason why the two specs should be different. I think the XSLT decision was made on the basis that invalid values should only be an error if you are validating. Michael Kay 5072 3 jim.melton 2005-07-21 16:27:42 +0000 At the July 2005 F2F, Don Chamberlin observed that this bug is a duplicate of #1570. *** This bug has been marked as a duplicate of 1570 ***