Re: [wsbpel] Issue - 125 - Proposal for Vote

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

Hi Andrew,

Thank you for your comment.
Please see my reply in-line.

andrew.francis@mail.mcgill.ca wrote:
> Hi Yuzo:
> 
> 
>>INCORRECT
>> <variable name="intTypeVar" type="xsd:int"/>
>> ...
>> <assign>
>>   <copy>
>>     <from>0</from>   <!-- INCORRECT -->
>>     <to variable="intTypeVar"/>
>>   </copy>
>> </assign>
> 
> 
>>CORRECT
>> <variable name="intTypeVar" type="xsd:int"/>
>> ...
>> <assign>
>>   <copy>
>>     <from><p:intVal xsi:type="xsd:int">0</p:intVal></from>
>>     <to variable="intTypeVar"/>
>>   </copy>
>> </assign>
> 
> 
> Why is it so difficult for a BPEL parser to do the right
> thing in this example? The <from> has an attribute that
> specifies its "category" (i.e., variable). Absence impliesa literal. The compiler looks at the category and type of
> the <to>, and sees if <from> is type compatible (or
> can be coerced into a compatible type). Perhaps if the
> parser is not sure what to do, it generates a warning.

It is not difficulty but ambiguity that I intended to adress.
A reader of the current specification will probably be unable to
tell which of the above is correct (or both are acceptable).
I tried to define an unambiguous rule that is applicable to both
XML Schema type and element data.

As for knowing the type of the <to>:
It is generally difficult to tell the type because
the <to> clause can contain a query and an expression.


Yuzo Fujishima
NEC Corporation

> 
> Autoboxing springs to mind when I look at these
> two examples.
> 
> Cheers,
> Andrew
> 
> 
> 
> 
> To unsubscribe from this mailing list (and be removed from the roster of the OASIS TC), go to http://www.oasis-open.org/apps/org/workgroup/wsbpel/members/leave_workgroup.php.
> 
> 
>