- From: Stephen R. Savitzky <steve@crc.ricoh.com>
- Date: 28 Jul 1998 15:28:24 -0700
- To: Lauren Wood <lauren@sqwest.bc.ca>
- Cc: www-dom@w3.org
Lauren Wood <lauren@sqwest.bc.ca> writes: > At 28/07/98 02:31 PM , keshlam@us.ibm.com wrote: > > >Of course there's always the other approach to resolving the conflict: Is > >it Really Necessary that NodeList be "live"? How often is this going to > >matter, and is it in fact the behavior you want (as opposed to requesting a > >new NodeList when you know that changes have been made that you need to > >respond to)? > > The main reason for making NodeList live is so that scripts can start > accessing the document while it's still being loaded, and not have to poll > the NodeList every so often to see if there are now five H3's in the > document where only three had turned up before. There seems to be some differences of opinion about whether all NodeLists are live. I think this is a matter which can usefully be left unresolved. > It's not meant to be the solution to multi-threading problems; in fact > Level 1 assumes that only one thread at a time can access the document so > that you are in charge of everything that changes (apart from the process > loading in the document which can't delete anything or move it around). If this is really the intent, it should say so. This merely specifies that the NodeList returned by getElementsByTagName should be computed lazily in a context where the document may not be entirely present in memory, and does _not_ require that it be "live" across all possible modifications to the document. The latter is a much stronger requirement, and much harder to implement. The former can be ignored completely by an application that parses the entire document before processing it. -- Stephen R. Savitzky Chief Software Scientist, Ricoh Silicon Valley, Inc., <steve@rsv.ricoh.com> California Research Center voice: 650.496.5710 fax: 650.854.8740 URL: http://rsv.ricoh.com/~steve/ home: <steve@starport.com> URL: http://www.starport.com/people/steve/
Received on Tuesday, 28 July 1998 18:23:37 UTC