19365 2012-10-08 20:28:32 +0000 [XQuery30] add statically known decimal formats to C.1 Module Static Context 2013-07-01 22:21:46 +0000 1 1 1 Unclassified XPath / XQuery / XSLT XQuery 3.0 Last Call drafts PC Windows XP CLOSED FIXED https://www.w3.org/Bugs/Public/show_bug.cgi?id=20828 P2 normal --- 1 andrew.eisenberg jonathan.robie jmdyck mike public-qt-comments oldest_to_newest 75623 0 andrew.eisenberg 2012-10-08 20:28:32 +0000 Statically known decimal formats are defined in 2.1.1 Static Context. A row should be added to the table in C.1 Module Static Context, defining the initialization and alteration of this component. 77410 1 jonathan.robie 2012-10-30 12:58:31 +0000 Here's my first shot. * Default initial value output:method = "xml"; * Can be overwritten or augmented by implementation? Yes * Can be overwritten or augmented by prolog? Yes, using 2.2.4 Output Declarations * Can be overwritten or augmented by expressions? No * Consistency rules Must conform to a valid set of <termref>output declarations</termref> 77412 2 mike 2012-10-30 13:05:21 +0000 >output:method = "xml"; Eh? statically known decimal formats have nothing to do with serialization. 77425 3 jonathan.robie 2012-10-30 14:49:22 +0000 Here's my second shot. * Default initial value Specified in http://www.w3.org/TR/xquery-30/#dt-static-decimal-formats * Can be overwritten or augmented by implementation? Yes * Can be overwritten or augmented by prolog? Yes, Decimal Format Declarations * Can be overwritten or augmented by expressions? No * Consistency rules See error conditions in http://www.w3.org/TR/xquery-30/#dt-static-decimal-formats 77434 4 jonathan.robie 2012-10-30 16:13:46 +0000 The Working Group has adopted the resolution in comment #3. 79965 5 jmdyck 2012-12-11 21:48:41 +0000 You added the row to the table in C.2 (dynamic context), should have added it to C.1 (static context). (Also, note that in rev 1.53 of src/back-matter.xml, I reordered the rows of the tables in appendix C so that components are grouped by overwriteability: first the ones that can be overwritten/augmented by expressions, then the ones that don't vary at the expression level but can be overwritten/augmented by (decls in) prologs, and then the ones that can't be overwritten in either. [I should have put a comment to that effect in the XML source, sorry.] Anyhow, statically known decimal formats should be in the second group.) 81017 6 jonathan.robie 2013-01-07 23:24:22 +0000 Fixed in current build. 81046 7 andrew.eisenberg 2013-01-08 14:03:02 +0000 Looking at the latest draft, saying "yes" to "Can be overwritten or augmented by implementation?" seems wrong. Perhaps statically known collations and default collation could act as our guide. Default decimal format * Default initial value Implementation-dependent * Can be overwritten or augmented by implementation? No * Can be overwritten or augmented by prolog? No * Can be overwritten or augmented by expressions? No * Consistency rules None. Statically known decimal formats * Default initial value only the default decimal format * Can be overwritten or augmented by implementation? no * Can be overwritten or augmented by prolog? augmentable, by Decimal Format Declarations * Can be overwritten or augmented by expressions? No * Consistency rules Each URI uniquely identifies a decimal format. 81170 8 andrew.eisenberg 2013-01-10 02:09:56 +0000 There seemed to be sentiment at the joint meeting on Tuesday to stay with one row in C.1 Static Context Components. Better understanding decimal formats, let me suggest: Statically known decimal formats * Default initial value the default (unnamed) decimal format, which has an implementation-dependent value * Can be overwritten or augmented by implementation? augmentable * Can be overwritten or augmented by prolog? augmentable, by Decimal Format Declarations * Can be overwritten or augmented by expressions? No * Consistency rules Each URI uniquely identifies a decimal format. 81176 9 mike 2013-01-10 08:25:28 +0000 One reservation - decimal format names are QNames, not URIs. 82280 10 jonathan.robie 2013-01-29 17:05:45 +0000 In today's telcon we agreed to adopt Comment #8 modulo Comment #9.