- From: Boris Zbarsky <bzbarsky@MIT.EDU>
- Date: Sun, 03 Jan 2010 23:54:01 -0500
- To: James Hopkins <james@idreamincode.co.uk>
- CC: www-style list <www-style@w3.org>
On 1/3/10 8:24 PM, James Hopkins wrote:
> If "bar" is split over two adjacent text nodes (e.g "b ar"), it could be
> matched by ::text(b ar){}.
I'm not sure I follow. ::text(b ar) would match the string "b ar" (with
a space in it), no?
> I personally can't envisage a use case where crossing textnodes (or
> element boundaries, for that matter) in order to match a single word,
> would be beneficial.
Really? Textnode boundaries can come in arbitrary places in text (more
on this in my upcoming reply to Brad in this thread).
Element boundaries happen in the middle of words all the time right now;
whether that's because sites are trying to apply particular styling to
parts of words or whatever, the fact of the matter is that it's pretty
common at the moment. As I recall, Gecko ends up having to do things
like whitespace processing across element boundaries, ligature
processing across element boundaries (as in, ligatures which consist of
characters with different parent elements), substring matching across
element boundaries for the browser's "find" functionality, and so forth.
-Boris
Received on Monday, 4 January 2010 04:54:37 UTC