W3C home > Mailing lists > Public > www-dom@w3.org > October to December 1998

Re: An observation about "live" NodeLists

From: <keshlam@us.ibm.com>
Date: Tue, 20 Oct 1998 16:40:27 -0400
To: www-dom@w3.org
Message-ID: <852566A3.0070AEC9.00@us.ibm.com>
1.  Our implementation directly implements children as lists, so this is
not an issue when processing child lists.

Child lists are relatively easy. My implementation's tree-based, but Node
implements NodeList for the getChildren case and it works well enough.

2.  getElementsByTagName is not something we call much on the server

I don't think I can make that assumption for my users. But I also think
that there are going to be _relatively_ few cases of intensively
interleaved find-and-edit similar to Steve's remove-all case. I've set up
non-editing access to be reasonably fast, and editing access to be
not-intolerably-slow (I hope); I'm not delighted with it and am considering
alternatives, but it's functional and fairly compact in both codesize and
memory use.

4.  If we ever did forsee a use case that might stress test it, I would
quickly build indexes.

Miles' trick may overcome the problems I had with that approach.
______________________________________
Joe Kesselman  / IBM Research
Unless stated otherwise, all opinions are solely those of the author.
Received on Tuesday, 20 October 1998 15:55:24 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 22 June 2012 06:13:46 GMT