W3C home > Mailing lists > Public > www-dom@w3.org > July to September 1998

Re: DOM Level 1 doesn't do everything

From: Lauren Wood <lauren@sqwest.bc.ca>
Date: Tue, 28 Jul 1998 13:59:33 -0700
Message-Id: <199807282100.OAA23879@sqwest.bc.ca>
To: www-dom@w3.org
At 28/07/98 01:49 PM , keshlam@us.ibm.com wrote:
>DOM1 not doing everything: Makes sense to me.
>
>But the more I look at it, the more I'd argue that live index-based
>NodeLists belong in that second category, if they're needed at all. Live
>isn't bad. Array-like isn't bad. The combination of both with a "gather"
>call is murderous.
>
>Is there any chance we could talk you into changing the NodeList.item()
>call back into a tree-walking first()/next()/previous()/last() API? Among
>other justifications, that's going to be cheaper to process if it turns out
>that we only needed the first few responses from ...ByTagName().

The are some on the committee who agree with you 100% on this one; and
others who equally can't see why we want next/previous at all, because they
think that index-based lookups and navigation are much more natural to the
user... so we ended up with both methods in effect. (all those
next/previous-walking methods on Node).

Yes, the getElementsByTagName is an exception, but when we do the full
filter/query system (planned for Level 2) we'll also probably have
iterators to hold the results of those, so getElementsByTagName might
become completely superfluous.


Lauren
Received on Tuesday, 28 July 1998 16:59:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 22 June 2012 06:13:45 GMT