- From: Steve Schafer <steve@fenestra.com>
- Date: Sun, 04 Jan 2004 23:10:55 -0500
- To: www-dom@w3.org
On Sat, 03 Jan 2004 21:05:55 -0600, Curt Arnold <carnold@houston.rr.com> wrote: >When populating a DOM tree by loading a document using DOM L&S, the >application has no hook to control the type of nodes created during parse. Aha. I understand your point now. My mistaken assumption was that you were talking about an integrated parser/application, but you're actually thinking of a standalone, off-the-shelf parser, and you're looking for a mechanism to selectively turn on/off its "smart" parsing capabilities. In that case, I agree that your proposal makes sense. But I have to ask: Won't _any_ parser that's capable of "smart" parsing necessarily contain a heavy dose of implementation-dependent features? It seems to me that an application-specific DOM tree that doesn't actually do anything isn't much more use than a generic DOM tree. Steve Schafer Fenestra Technologies Corp http://www.fenestra.com/
Received on Sunday, 4 January 2004 23:12:08 UTC