[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: RE: [provision] UPA Improvements...
We recently encountered a UPA violation error in SPML 1.0 when we used XMLSpy or apache's xerces to validate SPMLv1.0 document: [##any] (or elements from their substitution group) violate "Unique Particle Attribution". During validation against this schema, ambiguity would be created for those two particles. This error is caused by <xsd:any> tags without namesapace attribute specified. To pass the validation, the published SPML1.0 schema file has to be changed to include namespace="##other" attribute in <any> tags. As a result of the change, we have to host the changed xsd file ourselves instead of using the one publicly hosted by OASIS. Regards, James -----Original Message----- From: Kent.Spaulding@Sun.COM [mailto:Kent.Spaulding@Sun.COM] Sent: Monday, March 12, 2007 1:21 PM To: Gary.P.Cole Cc: Bohren, Jeff; provision@lists.oasis-open.org Subject: Re: [provision] UPA Improvements... Does anyone else feel that if we can't fix all the UPA issues, then we're just tilting at windmills; or whatever that expression is? --Kent Gary.P.Cole wrote: > Jeff, thanks for doing the work to propose these changes. Since I'm a > bit rusty on the XSDs and since Diffs.txt highlights the "after" version > of changed definitions (but does not show the "before" versions), I'd > like to check my understanding with you: > > 1) ExtensibleType still allows 'anyAttribute', but no longer allows > the > 'sequence-of-any-element' that it previously did. > > 2) A new 'ExtensibleDataType' extends 'ExtensibleType' to allow both > 'anyAttribute' and 'sequence-of-any-element'. > > 3) The new 'ExtensibleDataType' is used as the data type for elements > within several types (that previously used 'ExtensibleType'): > - PSOType#data > - ReferenceType#referenceData > - HasReferenceType#referenceData > > 4) A number of Types that extend 'ExtensibleType' now explicitly add > 'sequence-of-any-element' (whereas in the past they would have inherited > 'sequence-of-any-element'): > - CapabilityDataType > - PSOIdentifierType > - QueryClauseType > - BatchRequestType > - BatchResponseType > - SearchQueryType > > Do the above (1-4) correctly summarize the proposed changes? > > Bohren, Jeff wrote: >> >> Attached is a some possible changes we could put into the XSDs to >> reduce some of of UPA issues. I am reluctant to making any changes in >> the XSDs as part of the errata unless it really makes a difference. >> I’m not sure that these changes are worth it since they don’t resolve >> all the UPA issues, just some of them. >> >> >> >> That said I believe that these changes would still be compatible with >> the specification, but we need to evaluate the likelihood of it >> causing problems with existing SPML 2.0 implementations. >> >> >> >> Please take some time to evaluate these possible changes. The >> attached >> file contains the affects XSD files and a diffs.txt that summarizes >> the changes. >> >> >> >> *Jeff Bohren* >> >> 13577 Feather Sound Drive, Suite 200, Clearwater, FL 33762 >> >> tel: 813.433.5719 >> >> >> >> Jeffrey_Bohren@bmc.com <mailto:Jeffrey_Bohren@bmc.com> >> >> www.bmc.com <http://www.bmc.com> >> >> Blog: http://talk.bmc.com/blogs/blog-bohren/jeff-bohren/ >> >> >> >> >> >
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]