9787 2010-05-21 11:05:02 +0000 [XQ31ReqUC] Default values for function arguments 2014-05-20 16:54:08 +0000 1 1 1 Unclassified XPath / XQuery / XSLT Requirements for Future Versions Working drafts All All ASSIGNED P2 normal --- 9062 1 nunojobpinto jim.melton jim.melton john.snelson jonathan.robie mike public-qt-comments oldest_to_newest 35791 0 nunojobpinto 2010-05-21 11:05:02 +0000 define function local:foo($bar, $baz=fn:current-date-time()) { ... } ; instead of define function local:foo($bar) { local:foo($bar, fn:current-date-time()) } ; define function local:foo($bar, $baz) { ... } ; 35792 1 john.snelson 2010-05-21 13:17:21 +0000 I think this is an interesting future direction, since it will also allow users to safely define functions that take the context item as an implicit argument: define function local:findIDs($node as node() := .) { $node//@id }; This kind of default argument is evaluated in the calling context, rather than the function's context. That said, this is almost certainly a feature that should be considered for XQuery 1.2 rather than XQuery 1.1. 35947 2 jonathan.robie 2010-06-03 19:53:07 +0000 Polymorphism is already listed as a potential feature for future versions, as are generics: Bug 9062 This is related to polymorphism, but you ask for defaults as well. I'd like to move this to future versions. 71449 3 mike 2012-07-25 14:26:29 +0000 Renamed the requirement to emphasize its relatively modest ambitions; as described, this is simply a convenient way of defining a family of functions with the same name and different arity using fewer lines of code than currently needed. 71450 4 john.snelson 2012-07-25 14:52:11 +0000 There's a little more to this feature request. Default arguments would be evaluated in the static and dynamic context of the caller, which would allow the definition of functions which took a default context item, for instance. 71468 5 mike 2012-07-25 21:28:14 +0000 The static environment of the caller? So $x:var accesses different variables depending on the namespace binding of x? Surely not. 71504 6 john.snelson 2012-07-26 10:55:38 +0000 Interesting point - this clearly needs more thought. I do think it's reasonable to have a default argument of fn:default-collation() however, which is what I was thinking of when I said it should use the static context of the caller. 106476 7 jonathan.robie 2014-05-20 16:54:08 +0000 Assigning to future requirements per Working Group decision (https://lists.w3.org/Archives/Member/w3c-xsl-query/2012Oct/0087.html).