- From: Ian Hickson <ian@hixie.ch>
- Date: Sat, 6 Jan 2007 08:03:41 +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: > > "Each node that is to be distributed (each explicit child node) must be > assigned to a content element as follows: > > 1. If the node is already assigned to a content element, and the content > element is locked, then that is the content element to which the node must be > assigned, stop here." > The formulation is difficult to understand: why would I try to assign a node > to content element if it is already assigned? The nodes are redistributed for all kinds of reasons, as described in the very next section. > Please give an example in the specification. Done. > Then, if it is not locked, should I remove the previous assignment ? The algorithm reassigns all the children ("Each node that is to be distributed (each explicit child node) must be assigned to a content element as follows"). > Could you please clarify the following points in the spec: > - is there a one-to-one mapping between a child of the bound element and a > content element ? No, there's a many-to-many mapping. Each explicit child can be assigned to multiple <content> elements (one per shadow scope), and each <content> element can be assigned multiple explicit children. (An explicit child rarely has more than one <content> element assigned, but it can happen when an element in a shadow tree that contains a <content> element is itself bound to a shadow tree with <content> elements, as noted in the spec.) > - Can two content elements match the same child of the bound element ? Yes, in the rare case described above. > - if multiple children match the selector, are they all assigned to the > content element ? Yes, this seems clear to me in the spec; could you elaborate on what makes you think otherwise? > - what happens if a selector matches a grandchild of the bound element ? is it > valid ? The grandchild is never examined for the purposes of <content> processing, only the explicit children are examined. > I suggest adding a sentence like: "Only explicit child nodes are dispatched. > It is not permitted for the includes attribute to specify other nodes than the > explicit children of the bound element." This is already there; the spec always says that it is only the explicit children that are distributed, not any random node. > In section 5.4, there is a typo in step 3, remove 'and' 'If T contains a > correct and content element ' Fixed. Cheers, -- Ian Hickson U+1047E )\._.,--....,'``. fL http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Received on Saturday, 6 January 2007 08:03:55 UTC