Re: [wsbpel] Outgoing link from a fault handler

[Yuzo Fujishima <fujishima@bc.jp.nec.com>]

Chris,

Thank you for the response.

Chris Keller wrote:
> Hi Yuzo,
> 
> E to H should be allowed. If the text is not clear we should change the
> sentence to:
> 
>   In addition, a link that crosses a fault-handler boundary 
>   MUST be outbound, that is, it MUST have its source activity 
>   within the fault handler and its target activity MUST not be 
>   to an activity within the fault handlers associated scope.

Now I think I get it. I like the above line much better
than the original line in the current draft.

Please help me verify my understanding.
My understanding is as follows:
  The "main part" of a scope will never be executed after
  the scope has faulted. Hence, a link from within the fault
  handler to inside the main part is senseless, and mostly
  harmful because such link is a likely source of "dead-lock",
  i.e., an activity waiting for the link status determination
  that will never be determined.

Do I get it right?

Yuzo

> 
> If you think the text is not clear then it probably isn't so I think you
> should open an issue?
> 
> - Chris
>  
> -----Original Message-----
> From: Yuzo Fujishima [mailto:fujishima@bc.jp.nec.com] 
> Sent: Thursday, October 20, 2005 1:56 AM
> To: ws bpel tc
> Subject: [wsbpel] Outgoing link from a fault handler
> 
> Hi,
> 
> Could someone help me understand a restriction on link?
> 
> In the last paragraph of "12.6 Flow", 
>   In addition, a link that crosses a fault-handler boundary 
>   MUST be outbound, that is, it MUST have its source activity 
>   within the fault handler and its target activity within a 
>   scope that encloses the scope associated with the fault handler. 
> 
> What is the rationale behind the line above?
> 
> Example:
> 
> Scope A
>   Sequence
>     Scope B
>       Sequence
>         Scope C
>           FaultHandler D
>             Activity E
>         Activity F
>     Scope G
>       Activity H
>     Activity I
> 
> As far as I understand the specification,
> Link from E:
>   to F: Allowed, because B encloses C
>   to H: Disallowed, because G does not enclose C.
>   to I: Allowed, because A encloses C, although indirectly via B.
> 
> I don't quite understand why we should allow/disallow linking as above.
> 
> Some may claim:
>   E to H is allowed, because it is within A, because it is WITHIN A,
>   although indirectly.
> 
> One of the problems is, IMHO, the definition of "WITHIN A SCOPE" is unclear.
> 
> Yuzo Fujishima
> NEC Corporation
> 
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail.  You may a link to this group and all your TCs in OASIS
> at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php 
> 
> 
> 
> 
> 
> 
>