W3C home > Mailing lists > Public > www-dom@w3.org > April to June 2001

Re: DOM 3 Core: Overlap with JAXP, Document order of attributes, getC hildElementsByTagName, Exceptions

From: Arnaud Le Hors <lehors@us.ibm.com>
Date: Tue, 15 May 2001 17:01:37 -0700
Message-ID: <3B01C361.DE8B150D@us.ibm.com>
To: "Arnold, Curt" <Curt.Arnold@hyprotech.com>
CC: "'www-dom@w3.org'" <www-dom@w3.org>
"Arnold, Curt" wrote:
> 
> Section 1.2: DOM Level 3 Java Binding Extension
> 
> JAXP seems established and useful and covers this requirement,
> why does it need to be replicated within the DOM spec.
> I could see it if it was trying to replicate the same
> architecture for the other bindings, but then there should
> be a generic discussion in a core and discussion of the Java
> and ECMAScript implementations in the bindings section.

That's actually what we plan to have. In addition, the code you can see
in the public draft is out of date. We now have something much more
powerful which allows to fetch an implementation based on a set of
requested features.

> Enumeration DocumentOrder and TreePosition:
> 
> What would be the result of comparing the order of two attributes
> of the same element, two attributes of a different element, an
> attribute and a child of the same element.

All of these cases give "unordered".

> getChildElementsByTagName:
> 
> Newbies are typically suprised to find out that getElementsByTagName()
> returns all descendant elements with the specified tag name.  Would it
> be useful to add a method that supports getting only the immediate
> children of the target object?

I agree the name is misleading. It should really be called something
like getDescendantsByTagName(). Now, do we need another method for
immediate children? I personally think people are really lazy, the cost
of looking for this themselves by walking the list of children is minor
and there is little hope the implementation can do better. But users
always want more... So, I'll bring it up to the Working Group.

> Exceptions:
> 
> Trying to determine the nature of an exception is difficult to do in
> a implementation portable way on non-Java platforms.  The HRESULT's
> and messages used in IE or the Adobe SVG Viewer don't have a ready
> correspondance to each other or to the DOMException.code's.  Is there
> any mechanism that DOM 3 could add to map these platform specific
> exceptions back to the defined DOMException codes, maybe a
> domImplementation.getExceptionCode(Object exception) method?

That's an interesting idea. I'll forward that one to the WG too.
Thanks for your input!
-- 
Arnaud  Le Hors - IBM Cupertino, XML Strategy Group
Received on Tuesday, 15 May 2001 20:00:30 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 22 June 2012 06:13:48 GMT