- From: Doug Schepers <schepers@w3.org>
- Date: Wed, 11 May 2011 14:52:58 -0400
- To: "www-dom@w3.org" <www-dom@w3.org>
http://www.w3.org/2011/05/11-webapps-minutes.html
Web Applications Working Group Teleconference
11 May 2011
See also: [2]IRC log
[2] http://www.w3.org/2011/05/11-webapps-irc
Attendees
Present
Doug_Schepers, [Microsoft], Olli_Pettay
Regrets
Chair
SV_MEETING_CHAIR
Scribe
jrossi
Contents
* [3]Topics
1. [4]ISSUE-137
2. [5]legacy key events
3. [6]ISSUE-142?
4. [7]ISSUE-142
5. [8]onclick after removing element from document during
onmouseup
6. [9]WebApps-ISSUE-178?
7. [10]http://dev.w3.org/2006/webapi/DOM-Level-3-Events/html/D
OM3-Events.html#event-type-DOMActivate
8. [11]strange DOMACtivate wording
9. [12]Next Steps
10. [13]DOM Core
* [14]Summary of Action Items
_________________________________________________________
<trackbot> Date: 11 May 2011
<smaug> let's see if the connection is good enough
<scribe> scribenick: jrossi
ISSUE-137
smaug: it *may* fire for IME
jrossi: yes, seems implementation specific. IE9 doesn't seem to fire
at all.
shepazu: let's add spec text stating that
resolution: add text to keypress that states this event may not fire
for IME input, depending on platform/implementation
legacy key events
<shepazu>
[15]http://dev.w3.org/2006/webapi/DOM-Level-3-Events/html/DOM3-Event
s.html#KeyboardEvent-supplemental-interface
[15]
http://dev.w3.org/2006/webapi/DOM-Level-3-Events/html/DOM3-Events.html#KeyboardEvent-supplemental-interface
jrossi: yes, I'm fine with this with the new spec text
shepazu: I also felt it OK to use WebIDL here since it's
non-normative
jrossi: agree
smaug: yes, that's fine
resultion: keep it as is
ISSUE-142?
ISSUE-142
ISSUE-142?
<trackbot> ISSUE-142 -- one keydown might fire multiple
keypress/textInput events -- raised
<trackbot> [16]http://www.w3.org/2008/webapps/track/issues/142
[16] http://www.w3.org/2008/webapps/track/issues/142
shepazu: maybe test this with a macro
jrossi: testing with CTRL+V, we get one textinput
... are there dead keys that are macros of multiple characters?
shepazu: don't see many results from a search for these
... I think it should result in only a single textinput event
jrossi: I'm fine with Hallvord's suggested spec text with the caveat
that IE may be broken in some unknown cases. But we probably
wouldn't value those scenarios highly and would treat them as bugs.
smaug: I'm fine with this. We should ask him for clarification
though.
shepazu: I'll follow up with Hallvord to confirm.
resolution: shepazu will discuss the effect of keypress for multiple
characters, but won't change the spec pending some evidence from
hallvord
onclick after removing element from document during onmouseup
<shepazu>
[17]http://lists.w3.org/Archives/Public/www-dom/2011JanMar/0052.html
[17] http://lists.w3.org/Archives/Public/www-dom/2011JanMar/0052.html
<shepazu>
[18]http://lists.w3.org/Archives/Public/www-dom/2011AprJun/thread.ht
ml#msg58
[18]
http://lists.w3.org/Archives/Public/www-dom/2011AprJun/thread.html#msg58
issue-180?
<trackbot> ISSUE-180 -- onclick after removing element from document
during onmouseup -- raised
<trackbot> [19]http://www.w3.org/2008/webapps/track/issues/180
[19] http://www.w3.org/2008/webapps/track/issues/180
<shepazu> [20]https://bugs.webkit.org/show_bug.cgi?id=60600
[20] https://bugs.webkit.org/show_bug.cgi?id=60600
jrossi: IE9 matches Gecko in that we don't fire click events or
perform the activation when the element is removed from the document
during mouseup
... I don't believe it's desirable (or potentially safe) to activate
links which have been removed from the document
smaug: yes, I agree
[21]http://dev.w3.org/2006/webapi/DOM-Level-3-Events/html/DOM3-Event
s.html#events-mouseevent-event-order
[21]
http://dev.w3.org/2006/webapi/DOM-Level-3-Events/html/DOM3-Events.html#events-mouseevent-event-order
smaug: it should probably go in 5.2.3.2 Mouse Event Order
jrossi: agrees
<a href="[22]http://www.google.com" id="test">test</a>
[22] http://www.google.com/
<script>
var elem = document.getElementById("test");
elem.addEventListener("mousedown", function(e) {
alert('mousedown');
elem.parentNode.removeChild(elem);
}, false);
elem.addEventListener("mouseup",function(e){
alert('mouseup');},false);
elem.addEventListener("click",function(e){ alert('click');},false);
</script>
jrossi: this doesn't apply specifically to click, if you remove the
element at any point in the sequence then the rest of the sequence
is aborted
<shepazu> "If an element is removed from the DOM between mousedown
and mouseup, no events will be dispatched for mouseup, click, or
subsequent activation events."
<shepazu> "A mouseup event may still be dispatched on a different
element, depending upon the speed of a 'click' event in the
environment."
<shepazu> "If an element is removed from the DOM between mousedown
and mouseup, no events will be dispatched for mouseup, click, or
subsequent activation events for that element. A mouseup event will
still be dispatched on a different element."
"If an element is removed from the DOM during the dispatch of
mousedown and mouseup, no events will be dispatched for mouseup,
click, or subsequent activation events for that element. A mouseup
event will still be dispatched on a different element."
"If an element is removed from the DOM between mousedown or mouseup,
no events will be dispatched for mouseup, click, or subsequent
activation events for that element. A mouseup event will still be
dispatched on a different element."
"If an element is removed from the DOM dispatch mousedown or
mouseup, no events will be dispatched for mouseup, click, or
subsequent activation events for that element. A mouseup event will
still be dispatched on a different element."
<shepazu> "If an element is removed from the DOM during the dispatch
of a mousedown event, no events for that element will be dispatched
for mouseup, click, or subsequent activation events. A mouseup event
will still be dispatched on a different element."
"If an element is removed from the DOM during the mouse events
sequence, the remaining events of the sequence will not be fired at
that element."
resolution: add something to the effect of what jrossi typed
<smaug>
[23]http://lists.w3.org/Archives/Public/www-dom/2011AprJun/0025.html
[23] http://lists.w3.org/Archives/Public/www-dom/2011AprJun/0025.html
WebApps-ISSUE-178?
<shepazu> WebApps-ISSUE-178?
issue-178?
<trackbot> ISSUE-178 -- Implementations and DOM Core allow empty
string and null event types -- pending review
<trackbot> [24]http://www.w3.org/2008/webapps/track/issues/178
[24] http://www.w3.org/2008/webapps/track/issues/178
jrossi: my recent change removed the requirement that initializing
an event with a type of empty string or null should throw an
exception
... no implementation did that, and there's not really a great
reason to
<smaug> WDC
[25]http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html#dom-eve
nttarget-dispatchevent
[25]
http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html#dom-eventtarget-dispatchevent
jrossi: jonas proposed that empty string not be allowed and instead
be the indicator that an event object has yet to be initialized
... I replied saying that I didn't think there was any reason type
was the correct place to expose such an API. And at any rate, my
change lays the ground work for another spec (such as DOM Core) to
spec how such an API (to detect whether an object has been
initialized) would work.
... but that it seems like feature creep to add that to D3E at this
point
smaug: it's not the cleanest way to expose such an API
shepazu: not a very intuitive way of exposing it, seems like a hack
jrossi: agree
resolution: spec is fine as is
<scribe> ACTION: shepazu to clean out "issues" marked up in spec
text [recorded in
[26]http://www.w3.org/2011/05/11-webapps-minutes.html#action01]
<trackbot> Created ACTION-624 - Clean out "issues" marked up in spec
text [on Doug Schepers - due 2011-05-18].
[27]http://dev.w3.org/2006/webapi/DOM-Level-3-Events/html/DOM3-Events.h
tml#event-type-DOMActivate
[27]
http://dev.w3.org/2006/webapi/DOM-Level-3-Events/html/DOM3-Events.html#event-type-DOMActivate
strange DOMACtivate wording
<shepazu>
[28]http://dev.w3.org/2006/webapi/DOM-Level-3-Events/html/DOM3-Event
s.html#event-type-DOMActivate
[28]
http://dev.w3.org/2006/webapi/DOM-Level-3-Events/html/DOM3-Events.html#event-type-DOMActivate
<shepazu> [29]http://www.w3.org/TR/DOM-Level-2-Events/events.html
[29] http://www.w3.org/TR/DOM-Level-2-Events/events.html
<a href="[30]http://www.google.com" id="test">test</a>
[30] http://www.google.com/
<script>
var elem = document.getElementById("test");
elem.addEventListener("DOMActivate", function(e) {
console.log(e.detail);
e.preventDefault();
}, false);
</script>
jrossi: don't see a detail property in FF for DOMActivate anyway
Next Steps
shepazu: I can close out these issues we discussed today
jrossi: then will you send mail to Art and the WG list to ask for a
CfC?
shepazu: sure
DOM Core
shepazu: are there any outstanding descrpancies between DOM Core and
D3E?
jrossi: not that I'm aware of
... we covered those last week (and then rehashed one issue today)
shepazu: smaug, you checked this?
smaug: yes, we fixed these issues (dispatch through non DOM trees,
emptystring/null event types, keeping EventException). I'm not aware
of any others.
shepazu: then I'll withdraw my objection to publication of DOM4 Core
<shepazu> trackbot, end telcon?
<trackbot> Sorry, shepazu, I don't understand 'trackbot, end
telcon?'. Please refer to [31]http://www.w3.org/2005/06/tracker/irc
for help
[31] http://www.w3.org/2005/06/tracker/irc
<shepazu> tracbot, end telcon
<shepazu> trackbot, end telcon
Summary of Action Items
[NEW] ACTION: shepazu to clean out "issues" marked up in spec text
[recorded in
[32]http://www.w3.org/2011/05/11-webapps-minutes.html#action01]
[End of minutes]
Received on Wednesday, 11 May 2011 18:55:01 UTC