- From: <bugzilla@jessica.w3.org>
- Date: Thu, 01 Mar 2012 12:54:00 +0000
- To: public-webapps-bugzilla@w3.org
https://www.w3.org/Bugs/Public/show_bug.cgi?id=16176
Summary: [Shadow]: What should we do if an event happens on
light child which is distributed to a insertion point.
Product: WebAppsWG
Version: unspecified
Platform: PC
OS/Version: All
Status: NEW
Severity: normal
Priority: P2
Component: Component Model
AssignedTo: dglazkov@chromium.org
ReportedBy: hayato@chromium.org
QAContact: public-webapps-bugzilla@w3.org
CC: dglazkov@chromium.org, dominicc@chromium.org,
morrita@google.com, rolandsteiner@chromium.org
Blocks: 14978
Let me explain using the following example.
Case A)
DOM Tree is here:
<div id='a'>
<div id='b'>
</div>
</div>
An element #a has a ShadowRoot, which has a <content> element as a child.
<shadow-root>
<content>
A result of node distribution algorithm is here (it's the same to original DOM
tree except for there is a shadow boundary between #a and #b):
<div id='a'>
-- <shadow boundary> --
<div id='b'>
</div>
</div>
Case B)
DOM Tree is here:
<div id='a'>
<div id='b'>
</div>
</div>
There is no ShadowRoot.
Suppose a user move mouse cursor from '#b' to '#a'.
In Case A)
- 'mouseout' event should happen only at '#b' and stop here since there is a
shadow boundary between '#a' and '#b'. Mouse cursor moves within a shadowHost
'#a', so a shadowHost '#a' should not receive any mouseout events.
In Case B)
- 'mouseout' event should bubble up usually. An 'mouseout' event
(currentTarget='#a', target='#b') happens also at '#a' as well as at '#b'.
I am afraid that users might be upset in case A) because it behaves differently
than normal DOM tree.
I've found this issue when I was implementing new event dispatching logic in
WebKit and apply a new event dispatching logic to <details> and <summary>:
<details>
<summary>summary<summary>
details
</details>
If mouse moves form <summary> to <details>, 'mouseout' event should happen at
<summary> and <details> (at BUBBLE).
But WebKit uses ShadowRoot internally, 'mouseout' event does not happen at
<details> if I apply the new event dispatching algorithm.
I'd like to treat these cases uniformly.
Which should I fix? I think this is worth discussing because we might want to
modify the spec.
Is there any ideas?
--
Configure bugmail: https://www.w3.org/Bugs/Public/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.
Received on Thursday, 1 March 2012 12:54:07 UTC