1826 2005-07-28 16:23:44 +0000 Editorial: op:union/op:intersect defined in F&O, but error handling in XPath 2.0 2005-09-29 12:51:33 +0000 1 1 1 Unclassified XPath / XQuery / XSLT Functions and Operators 1.0 Last Call drafts PC Linux CLOSED WONTFIX P2 normal --- 1 frans.englich ashok.malhotra public-qt-comments oldest_to_newest 5268 0 frans.englich 2005-07-28 16:23:44 +0000 Hello, This is maybe intentional, but I find it a bit confusing that error handling for op:union, op:intersect and op:intersect is defined in XPath 2.0, "If an operand of union, intersect, or except contains an item that is not a node, a type error is raised [err:XPTY0004].", while their definitions are in F & O "15.3 Equals, Union, Intersection and Except." Should the error handling specification be moved from XPath20 to F & O? Cheers, Frans 5269 1 mike 2005-07-28 17:52:58 +0000 You're not the only one that's confused by the way the operator specifications are split between the language book and the F+O book, but there is some method in the madness: in general operators are polymorphic, and the language book describes the logic that decides which of the various (non-polymorphic) functions needs to be despatched. In this case there is no suitable function to handle non-node-operands, so a type error is raised during the despatch process, not by the function itself. Michael Kay (personal response) 5272 2 ashok.malhotra 2005-07-28 18:38:54 +0000 Shall we close this based on Michael Kay's explanation of the situation? 5274 3 frans.englich 2005-07-28 18:42:11 +0000 Yes, I wouldn't object at least. Frans 5275 4 ashok.malhotra 2005-07-28 18:49:40 +0000 Closed with approval of the commentor.