This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
How are multiple pattern facet children represented in the schema component infoset? The pattern facet schema component has a value which is a regular expression. So presumably, that value is the disjunction of all patterns among the children. Is that so? How is the disjunction written? "p1|p2|p3", for instance? What about pattern facets at different levels in the derivation? The rec says they are effectively ANDed together, but how is that represented in the schema infoset, as there is no conjunction operator in the regex language? Is the schema processor to figure out the conjunction? Or do the base and derived patterns stay separate, with the processor required to walk the base chain for the ANDing? Or are there multiple pattern facets in the component model? I'm guessing the patterns at different derivation steps stay separate, but it's only a guess. Relevant part of rec (part 2): Schema Representation Constraint: Multiple patterns If multiple <pattern> element information items appear as [children] of a <simpleType> the [value]s should be combined as if they appeared in a single regular expression as separate branches. Note: It is a consequence of the schema representation constraint Multiple patterns (4.3.4.3) and of the rules for restriction that pattern facets specified on the same step in a type derivation are ORed together, while pattern facets specified on different steps of a type derivation are ANDed together. Related member-only thread: Facet equality: questions/issues http://lists.w3.org/Archives/Member/w3c-xml-schema-ig/2004Mar/0187.html See http://lists.w3.org/Archives/Public/www-xml-schema-comments/2004AprJun/0001.html
Looks like too big a change for 1.0 to straighten this out. Propose to handle it in 1.1. We may want to refer to bug 1929, which is about the same issue in the context of 1.1.
We discussed this briefly during the call of 2005-09-16, and agreed that we will discuss the correct way to reword the spec, when we resolve issue 1929.
In our telcon of 16 September 2005 the WG classified this item as a 'clarification requiring corrigendum'.