17173 2012-05-24 21:21:56 +0000 Inconsistent is-id behavior for updated attributes 2013-07-15 10:05:34 +0000 1 1 1 Unclassified XPath / XQuery / XSLT Update Facility Recommendation All All NEW P2 normal --- 1 josh.spiegel jonathan.robie josh.spiegel public-qt-comments oldest_to_newest 68087 0 josh.spiegel 2012-05-24 21:21:56 +0000 Whenever an XDM attribute is brought into the world, the XQuery and XDM specifications ensure that if its name is xml:id: (a) the is-id property is set to true and (b) the attribute value has undergone xml:id processing. Example 1: XDM Attribute Conversion http://www.w3.org/TR/xpath-datamodel/#const-infoset-attribute is-id "If the attribute is named xml:id and its [attribute type] property does not have the value ID, then [xml:id] processing is performed. This will assure that the value does have the type ID and that it is properly normalized" Example 2: Attribute construction http://www.w3.org/TR/xquery/#id-attributes (bullet 5) "If the attribute name is xml:id, then xml:id processing is performed as defined in [XML ID]. This ensures that the attribute has the type xs:ID and that its value is properly normalized. " It looks like the XQuery Update Facility also tries to maintain this policy. Example 3: http://www.w3.org/TR/xquery-update-10/#id-upd-rename (bullet 2. c.) "If $newName is xml:id, the is-id property of $target is set to true." Example 4: http://www.w3.org/TR/xquery-update-10/#id-upd-set-to-untyped (bullet 2. d.) "is-id is set to false if the attribute name is not xml:id." But, I think there might be some holes in the XQUF specification with respect to this. Problem 1: It is possible to cause an attribute named xml:id to have is-id set to false. ---------------------------------------- copy $doc := document { <e xml:id="bar"/> } modify replace value of node $doc/e/@xml:id with "foo" return fn:id("foo", $doc) I think that the result of this query should be <e xml:id="foo"/> but according to the spec the answer should be (). The replace expression will create a upd:replaceValue primitive which in turn will use upd:removeType on the attribute node. upd:removeType will set is-id to false, creating an attribute named xml:id with is-id false. Problem 2: It is possible to produce an attribute named xml:id with is-id set to true where the value has not undergone xml:id processing ---------------------------------------- copy $e := <e a="foo bar"/> modify rename node $e/@a as xs:QName("xml:id") return $e Possibly the result of this query should be <e xml:id="foo bar"/> but currently I think it would be <e xml:id="foo bar"/>. That is, the XQUF does not specify that xml:id processing should be performed on the value. And you can also combine problem 1 and problem 2 (an attribute named xml:id with is-id set to false and no xml:id normalization):     copy $e := <e xml:id="x"/>     modify replace value of node $e/@xml:id with "foo    bar"     return $e