30051 2017-01-20 16:30:42 +0000 "DirElemConstrutor" erroneously displayed in XPath 3.1 Proposed Rec 17.01.2017? 2017-02-07 18:26:12 +0000 1 1 1 Unclassified XPath / XQuery / XSLT XPath 3.1 Proposed Recommendation All All RESOLVED FIXED P2 editorial --- 1 todd.ditchendorf jonathan.robie andrew_coleman josh.spiegel todd.ditchendorf public-qt-comments oldest_to_newest 128403 0 todd.ditchendorf 2017-01-20 16:30:42 +0000 In section "3.11.3 The Lookup Operator ("?") for Maps and Arrays" (https://www.w3.org/TR/2017/PR-xpath-31-20170117/#id-lookup), there is a list of "Examples" explaining use of the Lookup Operator which includes the following item: > • If the context item is an array, let $x:= <node i="3"/> return ?($x/@i) does not raise a type error because the attribute is untyped. But let $x:= <node i="3"/> return ?($x/@i+1) does raise a type error because the + operator with an untyped operand returns a double. This example contains two "Direct Element Constructors" or "DirElemConstrutor"s that seem to come from the XQuery specification (https://www.w3.org/TR/xquery-31/#doc-xquery31-DirElemConstructor). Since XPath 3.1 does not contain DirElemConstrutors (but XQuery 3.1 does), is this example really meant to appear in the XPath 3.1 Rec? BACKGROUND ======== I came upon this apparent issue as I was browsing the XPath 3.1 Proposed Rec to learn about the new JSON support in XPath. I saw the above example, and momentarily thought that "Direct Element Constructors" had been added to XPath as well. That surprised and confused me for a while, until I dug thru the XPath 3.1 EBNF and couldn't find any trace of DirElemConstrutors there. So then I went to the XQuery 3.1 EBNF to compare, and *did* find DirElemConstrutors there. Anyhow, maybe I'm being overly picky, but I feel like the above-cited example should perhaps not appear in the XPath 3.1 Rec, since DirElemConstrutors are not allowed in XPath. I found it confusing, and suspect others might too. 128437 1 andrew_coleman 2017-02-06 09:01:54 +0000 ACTION-666-02 - Josh to address editorial bugs 30051, 30050