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: Conflicting partnerlinks


The latest spec (an olders) describe in 14.5

> If during the execution of a business process instance, two or more
> activities for the same partner link, portType, operation and
correlation set(s)
>  are in fact simultaneously enabled, then the standard fault
> MUST be thrown by a compliant implementation. 

Other parts of the spec describe scoped partnerLinks and name overriding
rules.  So my question is, if a partnerLink has the same name is it the
same partnerlink or only if it has the same declartion? If it is bound
to the same declaration, how does one compare partnerlinks in different
process templates or multiple versions?

So for describing the intention behind the above definition, it would be
good to define a "formal" identification string for a partner link
(where we can do string compare). For this I introduce:

<processnamespace>=the namespace of the process definitionm
<processname>=the local name of the process
<linkid>=unique string id for the declaration element (i.e. differs for
scopes and process template instances!)
<scopepath>=name of all parent scopes concatenated
<linkname>=ncname of the partnerlink (could be multiple nested instances
with the same name)

So the question is, what would be a "virtual" fully qualified
partnerLink name?


Thinks we need to consider:

Bla:Proc1 and Fasel:Proc2 both define a PartnerLink name="initial"  -
are outstanding receives allowed?

Bla.Proc1 defines two PartnerLink name="initial" in <scope><partnerLink
name="initial" /><scope><partnerLink name="initial"
/>...</sceope></scope> - are two conflicting receives allowed?


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