Re: section ordering and HTMLTableElement31 (Re: DOM Level 2 HTML Tests results)

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Tuesday 26 November 2002 23:51, Philippe Le Hegaret wrote:
> On Tue, 2002-11-26 at 16:29, David Faure wrote:
> > Most of the failing table tests are due to the section reordering issue. I sent
> > a mail to this list about it, but it remained unanswered. The subject line was
> > "Order of table sections". I would be glad if this could trigger some discussion :)
> 
> Looking at [1], it appeared that this issue was closed at:
> http://lists.w3.org/Archives/Public/www-dom/2002OctDec/0044.html

This sounds like a different issue to me.
The above is only about the rowIndex property, whereas the concern I'm
raising here (and in [1]) is the fact that the testsuite uses constructs like
document.getElementsByTagName("tr").item(4).

> It seemed like Mozilla and IE are following the same behavior here.
> (In case this issue needs a follow-up, I set the reply-to [email protected]
> given that this is an issue on the draft, and not the test suite).

I believe it's an issue with the test suite only.

> > I feel that relying on the browser's internal ordering of the table sections 
> > in the DOM tree is wrong.
... and more precisely, that rows should be found by ID, or by actually using rowIndex,
but NOT using document.getElementsByTagName("tr").item(4),
which is rather fragile IMHO.

- -- 
David FAURE, [email protected], [email protected]
http://people.mandrakesoft.com/~david/
Contributing to: http://www.konqueror.org/, http://www.koffice.org/
Get the latest KOffice - http://download.kde.org/stable/koffice-1.2/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE94/2x72KcVAmwbhARAohJAKCVr/WHN73izaVeCdNfFk1AWvXA4wCaAoNU
RCMw++NMAeNZI4ibXZOesvk=
=QSPV
-----END PGP SIGNATURE-----