[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [wsbpel] wsbpel 3/31/2006: propertyAlias and Return of MultipleNodes
> yiu: Yes, we still need the text around "bpel:selectionFailure" in > chapter 9. The usage context was different. In chapter 9, we are using > propertyAlias in the context of correlationSet. It was not clear > before that whether we should throw a selectionFailure or > correlationViolation before we passed a related resolution. (I opened > that clarification issue last summer - Issue 219) mm3: Thanks for the additional information Alex. I had thought the reference would stay. What about the hint then that was discussed? Do we still need it? >>> mm1: In the F2F when we discussed query and expressions, we touched >>> on the application of the XPATH in the propertyAlias to the variable >>> value and what happens if this results in a response that contains >>> anything other than exactly one information item and/or a collection >>> of CII, i.e. a bpel:selectionFailure fault MUST be thrown in the >>> context of <copy> operation. I believe the text was rewritten in >>> Section 8 although at the time we suggested deletion. Part of the >>> reason for the rewording or deletion was because: In queries, is it >>> acceptable to return multiple nodes with this function and the >>> selections are filtered in the expression and end up with one node. >>> I don't see that we have specifically provide such a hint later in >>> Section 8.2.6 as suggested. Is there still a need to do so? (Note: >>> The text provided are brief notes, not proposed specification >>> language if appropriate). Thanks. >> >> mm2: In looking further I see the propertyAlias reference is in >> Section 9.2. Do we still need the hint? Is there any reason to still >> consider why we originally wished to delete this text in the F2F and >> that is still relevant. Thanks. >
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]