1701 2005-07-17 23:13:23 +0000 [FS] what is [For/FLWR]? 2005-09-29 10:48:04 +0000 1 1 1 Unclassified XPath / XQuery / XSLT Formal Semantics 1.0 Last Call drafts PC Windows 2000 CLOSED FIXED P2 trivial --- 1 fred.zemke simeon public-qt-comments oldest_to_newest 4787 0 fred.zemke 2005-07-17 23:13:23 +0000 4.8 [For/FLWR] expressions What is a "[For/FLWR] expression"? Don't you mean FLWORExpr? 4912 1 simeon 2005-07-19 18:02:06 +0000 The reason for the notation is that FLWOR expressions appear in XQuery, and For expressions appear in XPath. The formal semantics serves as one formal semantics specification for both languages, hence the [For/FLWR] notation. 4944 2 fred.zemke 2005-07-19 23:12:48 +0000 Perhaps we could reword the first sentence of 4.8 as follows: "[XPath] provides ForExpr, and [XQuery] provides FLWORExpr, which will be collectively referred to as [For/FLWOR]. [For/FLWR] expressions are used for iteration, ...". Note that I propose to respell it [For/FLWOR] to align with [XQuery]. 5029 3 simeon 2005-07-20 20:43:44 +0000 I agree we should make sure to respell [For/FLWR] to [For/FLWOR]. We should also explains what that mean. We can do it as you suggest, possibly we could also indicate this right at the beginning in the preliminaries. Any preference about how we do this? - Jerome 5239 4 fred.zemke 2005-07-26 22:49:26 +0000 (In reply to comment #3) > I agree we should make sure to respell [For/FLWR] to [For/FLWOR]. > > We should also explains what that mean. We can do it as you suggest, possibly we > could also indicate this right at the beginning in the preliminaries. > > Any preference about how we do this? If you put it as a general statement in the preliminaries, then it would something like "[x/y] refers to something that is called 'x' in [XPath] and called 'y' in [XQuery]." However, [expression/query] is not actually called "expression" in either specification; it is called Expr in both, though I see XPath has XPath ::= Expr and XQuery has QueryBody ::= Expr. Similarly [For/FLWOR] is not actually called For or FLWOR. So there is no syntactic transform from [x/y] to its referents. This good be arranged, but it is probably simpler to just supply a definition whenever one of these symbols is introduced. Fred 6594 5 simeon 2005-09-29 10:37:26 +0000 The [For/FLOWR] case has been fixed, but this particular comment does not touch any formal notation. This is automatically generated text for shared specifications (which FS is). [expression/query] for instance is intended in the english sense, an 'expression' or a 'query'. I don't think there is necessity to add formal description of that unformal written form. - Jerome