OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

wsbpel message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Re: [wsbpel] Issue 305 proposal for discussion


Thanks for the proposal.

I have taken a look of the proposal.
I have uploaded a reviewed version of your proposal to OASIS website:

Most of my review changes are relatively minor rewording for clarity, except the following two which have a real semantics changes:

[a] Consistent LValue Restriction:

There is no requirement that <query> return lvalues. When the <query> used in a variable variant to-spec or the <query> of <propertyAlias> used in a property variant to-spec does not return an lvalue, an attempt to assign to such a to-spec MUST fail with a bpel:selectionFailure (as defined in section 8.4. Assignment). Multiple nodes may be selected with this <propertyAlias> feature. However, those selections may be then filtered in the rest of expression and result in one node returned.

[b] Opaque from-spec and to-spec:

In your original proposal, you introduced two new forms of opaque of from-spec and to-spec. Those two original suggestions do not provide any new additional values, when compared with the opaque from-spec and to-spec based on expression. I would suggest we should either choose not to add new opaque forms or add opaque forms with different level of abstraction. For example:


·        <from> element incorporating a query:

<from variable="NCName" part="NCName"?>

   <query queryLanguage="anyURI"? opaque="yes" />





·        <to> element incorporating a query:

<to variable="NCName" part="NCName"?>

   <query queryLanguage="anyURI"? opaque="yes" />




These two forms of from-spec and to-specs are edited in the reviewed proposal that I have uploaded.


Alex Yiu

Mark Ford wrote:

I have prepared the changes for Issue 305 and added them to the TC documents page:


Here are the links to the document:


View Document Details:



Download Document: 




Summary of changes:

1. restored the <query> option to the first from-spec and to-spec.

2. various edits to Section 8 to establish that a query has a context while an expression does not

3. establish XPath 1.0 as the default value for a <propertyAlias> query


As it turns out, this was probably the fewest number of edits in order to achieve consistency between queryLanguage and expressionLanguage. Restoring the <query> to the from-spec and to-spec resulted in some overlap with the expression variants but this hopefully won’t cause any concern as we had this in the past and have agreed that redundancy was ok.

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]