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

 


Help: OASIS Mailing Lists Help | MarkMail Help

bpel4people message

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


Subject: [bpel4people] Issue 19 Proposal


All,

                Issue 19 reads:

SUBMITTED BY: Mark Ford
TARGET: B4P, Section 4

DESCRIPTION: Constellations 1 and 2 allow the task definition to reference
variables from the enclosing BPEL process. In some cases, these variables
can be used within a part of the task that is not immediately evaluated when
the people activity executes.

For example:

<ht:escalation ...>
   <ht:condition>ht:getActualOwner() = 'Bob' and $v1 = 100</ht:condition>
   ...
</ht:escalation>

Will a change to v1 after the people activity executes but prior to its
escalation having been reached be visible to the task? The spec should say
something about this since it could be an interop issue.

COMMENTARY: I don't have a good use case for this issue since I would expect
the task to make its decisions based on its input data and not the state of
the enclosing process.



                Based on discussions of the TC:

PROPOSAL:

                In section 4.3.1 under properties, after “In addition, XPath expressions used in enclosed elements may refer to process variables,” add:

Enclosed elements MUST use the current value of the process variable.  Changes to process variables MUST NOT directly cause changes in the execution of the enclosed elements, but only provide more current values when the enclosed elements choose to re-evaluate the expressions.

 

                Same in section 4.4.1 under properties, after “In addition, XPath expressions used in enclosed elements may refer to process variables.”

 

REASON:

This provides the maximum ability for consumers of the standard to migrate a b4p:peopleActivity from constellation 1 to 2 to 3 to 4 (any direction) with the actual functioning of the human task to be as consistent across them as possible.

 

Thanks,

Phil Allen

 

 



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