- From: Ian Hickson <ian@hixie.ch>
- Date: Sat, 6 Jan 2007 08:20:46 +0000 (UTC)
- To: Cyril Concolato <cyril.concolato@enst.fr>
- Cc: "WAF WG (public)" <public-appformats@w3.org>
On Wed, 6 Dec 2006, Cyril Concolato wrote: > > In Section 2.6 "The inherited element" and Section 2.7 "The xbl:inherits > attribute" use the same root 'inherit' while they are about different > behaviors: - "inherited" is related to the extends attribute and to > inheritance - "inherits" is related to attribute forwarding. Please > rename the second to "forwards" or something less confusing. I spoke with David Hyatt about this and we agree that xbl:inherits clashing with inherited/extends is a bit confusing. We don't really like the idea of calling it 'xbl:forwards', though, because that is easier to typo than 'inherits' for some reason. I'd rather have xbl:inherits and the resulting minor confusion than xbl:forwards and the resulting typos. Do you have any better suggestions for an alternative name? > Could you explain why the WG prefered specifying the attribute > 'inherits' with a specific syntax (difficult) to parse and not an > element with sub-elements like the following ? > > <forwards> > <attribute name="[bound element's attribute]"/> > <attribute name="xbl:text">some text</attribute> > </forwards> Elements are harder to process (you have to handle erroneous tree structures, whitespace nodes, unexpected other nodes, etc), and significantly harder to author. Also, in the case of the shadow tree we would have to somehow hide the <forwards> elements from the final flattened tree, the creation of which is already quite a confusing operation. Overall, the parsing of the attribute is less difficult than the processing of elements in this situation. -- Ian Hickson U+1047E )\._.,--....,'``. fL http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Received on Saturday, 6 January 2007 08:21:04 UTC