1690 2005-07-17 21:44:22 +0000 [FS] need to know what function is invoked before looking for special rules 2007-01-16 17:34:13 +0000 1 1 1 Unclassified XPath / XQuery / XSLT Formal Semantics 1.0 Last Call drafts PC Windows 2000 CLOSED FIXED P2 minor --- 1 fred.zemke simeon public-qt-comments oldest_to_newest 4776 0 fred.zemke 2005-07-17 21:44:22 +0000 4.1.5 Function calls Under "Static type analysis", the first sentence says "to typecheck a Core function call we first check in Section [7 Additional semantics of functions] if there is a specialized typing rule for the function." Then the second sentence says "Otherwise, the function signatures matching the function name and arity are retrieved...". This seems to be logically inverted. The first sentence assumes that you know whether a function call matches something in section 7, yet you have not even decided what function is being invoked, which comes in the second sentence. This can be fixed by clarifying that the check of Section 7 is done purely on the basis of the function's namespace and name, without needing to consult its signature. Thus the first sentence would be "If the namespace of the function is either fs or fn, and if the name matches one of the functions named in Section 7, then we use the rules in the subsection of Section 7 for the function on the basis of its namespace and name." That's probably what you meant; I think this just expresses it more precisely. 4910 1 simeon 2005-07-19 17:56:57 +0000 I believe this is editorial. - Jerome 9169 2 simeon 2006-04-11 22:47:24 +0000 This bug has been taken over by events. The static typing rules for function calls have been refactorized. Their is now a rule which applies to all cases which deals with the ordering and namespace problem mentioned here. - Jerome