Hi Richard, Kiran and Gareth,
A few things:
1. I think Kiran's problem is very significant. If we're to expose an
interface returning XMLCh *, then it should preserve XML tags, namespace
decls etc.; I think this could be provided for relatively easily by using
the DOMWriter.
2. The problem I have with only exposing DOM nodes is that it's very
DOM-centric; any SAX-based application that needs access to annotation
components gets left out in the cold. So my take is that we need both: a
way of getting at DOM node representations to satisfy DOM-based apps, and a
way of serializing those nodes to that their contents can be piped to a
SAXParser and consumed by SAX-based applications.
3. I'm not sure the approach (or implementation) currently under
discussion takes into account the fact that section 3.13.2 of the schema
structures spec [1] wants annotation components to contain more than just
appInfo or userData: it also wants non-schema attributes from the parent
component to be included. To whit, among the properties of an annotation
component is
[[
{attributes}
A sequence of attribute information items, namely those allowed by the
attribute wildcard in the type definition for the
<annotation>
item itself or for the enclosing items which correspond to the component
within which the annotation component is located.
]]
So I think the problem at hand is a bit more complex than has so far
appeared...
Cheers,
Neil
[1]: http://www.w3.org/TR/xmlschema-1/#declare-annotation
Neil Graham
XML Parser Development
IBM Toronto Lab
Phone: 905-413-3519, T/L 969-3519
E-mail: [EMAIL PROTECTED]
|---------+---------------------------->
| | Sir Woody |
| | Hackswell |
| | <[EMAIL PROTECTED]|
| | com> |
| | |
| | 04/02/2003 01:59 |
| | PM |
| | Please respond to|
| | xerces-c-dev |
| | |
|---------+---------------------------->
>---------------------------------------------------------------------------------------------------------------------------------------------|
|
|
| To: "'[EMAIL PROTECTED]'" <[EMAIL PROTECTED]>
|
| cc:
|
| Subject: RE: RFC: Schema annotation support
|
|
|
|
|
>---------------------------------------------------------------------------------------------------------------------------------------------|
On Wed, 2 Apr 2003, Bagepalli, Kiran wrote:
> A small glitch when I started using the code.
>
> If I have XML embedded in the documentation
> for eg.
> <annotation>
>
<documentation><store><ID>12</ID><NAME>S1</NAME></store></documentation>
> </annotation>
>
> Ideally speaking if I get the documentation I would be expecting a text
of
> <store><ID>12</ID><NAME>S1</NAME></store>.
> However the traverseSchema DOM eats away this as a Node and only return
> 12S1. This forces me escape all this content by
> > < which is very inconvenient.
>
> Kiran
This is exactly the reason I made documentation and appinfo domnodes. ;)
You don't lose anything. And I clone the nodes, so that when the domtree
is
killed, you still have something to traverse. But this was a quick and
easy
hack. I'm sure there's something better... or is there?
-Richard Balint
-----
Hark!
I am yclept ps, son of Bourne Shell,
daughter of Emacs, son of Dash-shell,
great-great-grandson of the ancient and holy INIT!
[EMAIL PROTECTED] http://sir.woody.hackswell.com
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
