Re: ISSUE-5: Open Enumerated Type (NEW ISSUE?)

Hi Paul,

----- Original Message From: <paul.downey@bt.com>

>> This may have been discussed, but from what I have seen, xs:unions in
>> general are badly supported by binding tools.  Hence this issue refers
>> to a particular case of the more general xs:union issue.

> It might be worth my raising a separate issue for xs:union.


I think that would be appropriate.  Maybe something like:

* Title:          Support for xs:union
* Description:    see below
* Target:         "Basic"?
* Proposal:       see below

Description: A number of binding tools do not support the xs:union construct 
well.  They often map a attribute/element xs:union value to a simple string 
irrespective of whether the type contains integers or other more specific 
types.  (It would be hoped that binding tools will improve in this area in 
future.)

Proposal:  Developers should be made aware that binding tools don't always 
handle xs:union well, and as such should avoid the construct if possible. 
However, there are occasions when xs:union can not be avoided.  In this case 
developers should be aware that additional application level processing may 
be required.

How does that sound?

Pete.
--
=============================================
Pete Cordell
Tech-Know-Ware Ltd
                         for XML to C++ data binding visit
                         http://www.tech-know-ware.com/lmx
                         (or http://www.xml2cpp.com)
=============================================

Received on Wednesday, 11 January 2006 11:06:52 UTC