- From: Ian Hickson <ian@hixie.ch>
- Date: Sat, 11 Mar 2006 01:21:41 +0000 (UTC)
On Sat, 25 Feb 2006, Henri Sivonen wrote: > > On Feb 25, 2006, at 02:02, Ian Hickson wrote: > > > On Sat, 23 Jul 2005, Henri Sivonen wrote: > > > > > > Which characters should a text/html HTML5 conformance checker consider > > > forbidden? The same characters that are forbidden in XML 1.0 (\0, FF, > > > etc.)? Or some other set? > > > > In what context? > > In the pre-parse Unicode character stream on one hand and in the > post-parse (that is NCRs expanded) character data and attribute values > on the other. IIRC, in XML 1.0 (but not 1.1) the restrictions are the > same in both cases. Well, the spec says to drop U+0000, and do something with U+000D such that U+000D never appears in the parse stream; the post-parse is just the DOM. Does that answer your question? -- Ian Hickson U+1047E )\._.,--....,'``. fL http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Received on Friday, 10 March 2006 17:21:41 UTC