1741 2005-07-18 23:46:50 +0000 [FS] editorial: 5.15 Function Declaration 2007-01-16 17:40:18 +0000 1 1 1 Unclassified XPath / XQuery / XSLT Formal Semantics 1.0 Last Call drafts All All CLOSED FIXED P2 minor --- 1 jmdyck simeon public-qt-comments oldest_to_newest 4853 0 jmdyck 2005-07-18 23:46:50 +0000 5.15 Function Declaration Notation "Parameters without a declared typed" s/typed/type/ Norm / rule (1|2|3|4) "[[ ParamList? ]]_Param" According to the Notation section, []_Param only applies to an individual parameter, not the whole list. Norm / rule (1|2) "[[ EnclosedExpr ]]_Expr" The []_Expr form isn't defined on EnclosedExprs. Rewrite "EnclosedExpr" as "{ Expr }", and put []_Expr around "Expr". 5.15 / Norm / rule (2|4) (return type is not provided) For these two rules, it might be simpler for the RHS to be a recursive call to []_PrologDecl on a FunctionDecl with the "as item*" inserted. Then only rules 1 and 3 have to be concerned with the detailed normalization. Norm / rule 4 "[[ QName ]]" Delete square brackets. 5.15 / SCP / para 1 "Because functions are mutually referential" s/are/can be/ "This rules also updates" s/also/therefore/ (or just delete "also") STA "The static typing rules for function bodies follows normalization ..." s/follows/follow/ "The typing rules below constructs a new environment" s/rules/rule/ "in which each variable has the given expected type" s/variable/parameter/ "the local functions component of the static context" s/local functions/function signatures/ (funcType isn't just local functions) "If type checking fails, a static error is raised." s/static/type/, I think. (err:XPTY0004) DCP / rule 2 / conclusion "Variable1 as SequenceType1" "Variablen as SequenceTypen" You probably want to change 'Variable' to 'VarRef'. (leftover from last year, comment #182) "when the function is called and is described in ..." s/and is/, as/ 9225 1 simeon 2006-04-13 18:49:55 +0000 Fixed as suggested. - Jerome