On Jan 27, 2006, at 01:17, Ian Hickson wrote:
On Wed, 25 Jan 2006, Henri Sivonen wrote:
These events don't go to the DOM builder / content sink directly.
Instead, there's a filter layer that takes care of tag inference and
emits a well-formed event stream (TagInferenceFilter.java and
EmptyElementFilter.java in my parser). Additionally, there's a filter
(not present in my parser, which is designed for conformance
checking;
this may need to be integrated into the tag inference filter) that
performs the "residual style" fixups.
That wouldn't work. You can't know whether something is well-formed
or not
til you get to the end of it.
I was not suggesting anything that needs observation to the end of
the document. The whole point was that decisions could be made on a
per parse event basis using the current parse event and two stacks.
(Mozilla does a "pre-parse" with what it has, sort of like what you
are
suggesting,
That wasn't at all what I was suggesting.
Actually, I think what you have in the spec now is very similar to
what I suggesting except I described the algorithm in terms of SAX
events before the tree construction and not in terms of DOM nodes.
--
Henri Sivonen
[EMAIL PROTECTED]
http://hsivonen.iki.fi/