- From: Norman Walsh <Norman.Walsh@Sun.COM>
- Date: Thu, 22 Mar 2007 10:52:21 -0400
- To: public-xml-processing-model-wg@w3.org
- Message-ID: <87y7lpz5y2.fsf@nwalsh.com>
/ Alex Milowski <alex@milowski.org> was heard to say:
| 1.8 Serialize / 1.9 Parse
|>
|> Do we really need these?
|
| Yes, yes, yes.
No, no, no.
And if we must, then surely the semantics has to be that the *entire*
document is escaped and returned in a new wrapper:
<doc><foo/></doc>
becomes
<wrapper><doc><foo/></doc></wrapper>
| 2.2 RELAX NG Validate
|>
|> I wonder if we need an option to enable annotations and data typing?
|
| If there are options, let's enumerate them. I didn't put any on it because
| I'm
| not a Relax NG expert.
Yeah, I need to think about that one a bit.
| 2.3 XInclude From Sequence
|>
|> Barf. I want to solve this problem in a more general way
|
| OK. Let's discuss that.
On the list.
| 3.3 Label Elements
|>
|> How does the component garauntee that it doesn't introduce duplicate
|> IDs?
|
| Well... it sees the whole document, right ?
So it doesn't stream, it buffers the whole thing. Fine, but it has to
say that. Or it should say that it streams and uses a particular
pattern for generating IDs and if you get dups, the component fails.
Be seeing you,
norm
--
Norman Walsh
XML Standards Architect
Sun Microsystems, Inc.
Received on Thursday, 22 March 2007 14:52:31 UTC