Consider element information items ES and EL, where ES matches a skip wildcard and EL is laxly assessed. And another two element information items CS and CL who are children of EL and ES respectively. I think the intention is that we don't try to assess CS, but do try to assess CL. But I don't think the spec spells that out clearly. Another case to consider is for Identity Constraints. What happens if a selector or a field result in an EII or AII that's within a subtree headed by an EII that matches a skip wildcard? Strictly following the current spec, it seems that IDCs still apply. Is this the intention? Thanks, Sandy Gao Software Developer, IBM Canada (1-905) 413-3255 sandygao@ca.ibm.comReceived on Monday, 7 March 2005 05:53:10 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 23 October 2007 06:13:36 GMT