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: Fw: [wsbpel] Issue - 157 - conf call brief recap ... and carryforward ...



Hi Rania,

During the discussion of the "F2F-Table", I did sense that we are treating AII and TII virtually the same.

That means: if people think we don't lose any clarity through this simplification, we can potentially collapse the rows and columns of AII and TII. That is:

From:
Src\Dest
EII
AII
TII
EII with
Complex-Content
RE
F
F
EII with
Simple-Content
RE
RC
RC
AII
RC
RC
RC
TII
RC
RC
RC


To:
Src\Dest
EII
AII/TII
EII with
Complex-Content
RE
F
EII with
Simple-Content
RE
RC
AII/TII
RC
RC


[** On the other hand:
  • Your first table is NOT a simpification of the "F2F-Table". In the "F2F-Table", copy from-TII-to-EII is defined as Replace-Content, NOT Fault.  
  • I am thinking whether collapsing those rows/columns might make using XSLT to define those entries a bit more complicated ... I guess we can worry about that when we actually arrive that step.
**]

Next 2 questions are:
  • Whether we want to differentiate Complex-Content from Simple-Content of EII: My current preference is: yes, we should differentiate. Because, our typing story will be a bit too weak, if we allow the following happen at runtime:


    <!-- 
         $nameVar is initialized to:
               <foo:personName>
                   <foo:firstName>John</foo:firstName>
                   <foo:firstName>Dole</foo:firstName>
               </foo:personName>
    -->

    <assign>
       <copy>
           <from> $nameVar
    </from>
           <to> $var/name/text() </to>
       </copy>
    </assign>
    <!--
    if we apply XPath 1.0 String() function / <xslt:value-of> automatically to nameVar, the result would be "JohnDole".
    Two strings will be concatenated together without any delimiter. I still find this default XPath 1.0 conversion kind of funky, except used in during string-value equal comparison.
    -->

    If people want to assign "JohnDole" to that name text field, I would say people would apply String() function to the from-spec. E.g.:

    <assign>
       <copy>
           <from> string($nameVar)
    </from>
           <to> $var/name/text() </to>
       </copy>
    </assign>


  • Whether and how to support XSD substitutionGroup pattern in <bpel:copy>.


Thanks!


Regards,
Alex Yiu



Rania Khalaf wrote:
Hi Alex,

So it looks like the table you have reduces to the diagonal,and the rest are just short-hand of using 'text()' (in the case of XPath. Otherwise whatever the equivalent thing is .. ). AII doesn't need to show up on the 'basic' table because it's really just selecting the AII and then doing'text()'.

Src\Dest

EII

TII

EII

Replace (except name)

F

TII

F

Replace


  Makes your original table reduce to pointing at the above cells (in italics, we just show the usage of 'text()' to undo the shortcut ):

Src\Dest

EII

AII

TII

EII with
Complex-Content

EII/EII

F

F

EII with
Simple-Content

EII/EII

TII/TII

From
EII/text()
To
AII/text()

TII/TII

From

EII/text()

AII

TII/TII

From AII/text()
To
EII/text()

I/TII

From
AII/text()
To
AII/text()

TII/TII

From

AII/text()
 

TII

TII/TII

To
EII/text()

TII/TII

To
AII/text()

Replace


If this is the case, do we really need all these entries ? Are they helping that much or just adding clutter ?

Note that I haven't addressed substitution groups here .. just trying to look closely at your table as you asked :)

Regards,
Rania





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