WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED INVALID
106040
gesture event scrolling of iframe in overflow div scrolls wrong target
https://bugs.webkit.org/show_bug.cgi?id=106040
Summary
gesture event scrolling of iframe in overflow div scrolls wrong target
Robert Kroeger
Reported
2013-01-03 14:23:47 PST
When a scrollable iframe is placed in a fixed size overflow-scroll div, the (synthetic) wheel events used to implement the gesture scroll are dispatched on the overflow div rather than the iframe. As a result the contents of the iframe cannot be scrolled with gesture events.
Attachments
Patch
(7.71 KB, patch)
2013-01-03 14:36 PST
,
Robert Kroeger
no flags
Details
Formatted Diff
Diff
Show Obsolete
(1)
View All
Add attachment
proposed patch, testcase, etc.
Robert Kroeger
Comment 1
2013-01-03 14:36:01 PST
Created
attachment 181222
[details]
Patch
Robert Kroeger
Comment 2
2013-01-03 14:39:21 PST
jamesr@: Please take a look?
James Robinson
Comment 3
2013-01-03 16:04:44 PST
Comment on
attachment 181222
[details]
Patch I'm a little confused by the naming in here. Why would wheel events ever be latched?
Robert Kroeger
Comment 4
2013-01-04 08:21:07 PST
(In reply to
comment #3
)
> (From update of
attachment 181222
[details]
) > I'm a little confused by the naming in here. Why would wheel events ever be latched?
AFAIK: the "latching" here arranges for all wheel events comprising a single touchscreen scroll gesture to be delivered to the same original node. Aside: I am not convinced that I like the pre-existing code here -- I think that
https://bugs.webkit.org/show_bug.cgi?id=103952
is going to make it better -- but I wanted to fix
http://crbug.com/167482
in a timely fashion.
James Robinson
Comment 5
2013-01-04 10:25:37 PST
(In reply to
comment #4
)
> (In reply to
comment #3
) > > (From update of
attachment 181222
[details]
[details]) > > I'm a little confused by the naming in here. Why would wheel events ever be latched? > > AFAIK: the "latching" here arranges for all wheel events comprising a single touchscreen scroll gesture to be delivered to the same original node.
That doesn't make any sense - touchscreen interactions should not generate wheel events at all and touchpad interactions (which do generate wheels) should deliver events at a point, not at a node.
> > Aside: I am not convinced that I like the pre-existing code here -- I think that
https://bugs.webkit.org/show_bug.cgi?id=103952
is going to make it better -- but I wanted to fix
http://crbug.com/167482
in a timely fashion.
Robert Kroeger
Comment 6
2013-01-04 12:58:25 PST
(In reply to
comment #5
)
> (In reply to
comment #4
) > > (In reply to
comment #3
) > > > (From update of
attachment 181222
[details]
[details] [details]) > > > I'm a little confused by the naming in here. Why would wheel events ever be latched? > > > > AFAIK: the "latching" here arranges for all wheel events comprising a single touchscreen scroll gesture to be delivered to the same original node. > > That doesn't make any sense - touchscreen interactions should not generate wheel events at all and touchpad interactions (which do generate wheels) should deliver events at a point, not at a node.
You're right.
http://crbug.com/134520
tracks in-progress work to fix. I had intended this patch to fix
http://crbug.com/167482
in m25 but if this particular patch is pointless, perhaps we could discuss alternatives there?
> > > > > Aside: I am not convinced that I like the pre-existing code here -- I think that
https://bugs.webkit.org/show_bug.cgi?id=103952
is going to make it better -- but I wanted to fix
http://crbug.com/167482
in a timely fashion.
Terry Anderson
Comment 7
2013-01-08 13:05:49 PST
(In reply to
comment #1
)
> Created an attachment (id=181222) [details] > Patch
Note: I have included this change in my latest WIP patch for
https://bugs.webkit.org/show_bug.cgi?id=103952
Terry Anderson
Comment 8
2013-01-08 16:52:43 PST
Comment on
attachment 181222
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=181222&action=review
> LayoutTests/fast/events/touch/gesture/resources/scroll-inside-editable-iframe.html:9 > + margin: 0px;a
should that 'a' be there?
> LayoutTests/fast/events/touch/gesture/resources/scroll-inside-editable-iframe.html:35 > +<body>
</body>
Terry Anderson
Comment 9
2013-01-16 12:43:35 PST
Comment on
attachment 181222
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=181222&action=review
> LayoutTests/fast/events/touch/gesture/touch-gesture-scroll-iframe-editable.html:75 > + touchtarget.contentDocument.body.addEventListener("mousewheel", recordWheel);
Why do you add the event listener to contentDocument for a scroll, but to contentDocument.body for a mousewheel? Shouldn't these both be set on contentDocument.body?
Robert Kroeger
Comment 10
2013-02-04 12:50:36 PST
Was fixed properly via other changes.
Eric Seidel (no email)
Comment 11
2013-03-01 02:49:15 PST
Comment on
attachment 181222
[details]
Patch Cleared review? from
attachment 181222
[details]
so that this bug does not appear in
http://webkit.org/pending-review
. If you would like this patch reviewed, please attach it to a new bug (or re-open this bug before marking it for review again).
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug