Now that stuff is a bit more complex and that we're adding media lazy
loading, factor the observer management down to Element.
Make UnbindContext more parallel to BindContext while at it. This
allows the new calls to work regardless of whether they get called
before or after the base class's Bind/UnbindFromTree, which is nice.
Differential Revision: https://phabricator.services.mozilla.com/D282723
nsCSSPropertyID only enumerates non custom properties, but the name suggests
otherwise. (eCSSPropertyExtra_variable only signals a custom property and does
not carry the custom name.) This mismatch has been confusing and also blocks
renaming AnimatedPropertyID to CSSPropertyId.
This patch renames nsCSSPropertyID to NonCustomCSSPropertyId to make the non
custom scope explicit. It also normalizes identifier naming (aPropID ->
aPropId, mID -> mId, etc.) for consistency.
No behavior changes intended.
Potential follow-ups:
- Move the enum to the mozilla namespace.
- Convert the enum to an enum class.
- Rename nsCSSPropertyIDSet.
Differential Revision: https://phabricator.services.mozilla.com/D271443
The found frame may be covered by another frame. However, current
implementation may use the covered frame as the target if it's the
nearest frame from the click/touch point.
A frame is covered by another frame means that web apps try to prevent
the users to click/tap the frame. Therefore, current behavior may make
it possible to click/tap an element which is tried to hidden. Thus,
current behavior may be not expected by both the web app developers and
the users.
This patch uses a part of intersection observer to get clipped border
box of a candidate frame. So, if the frame is clipped by the
containers, the hidden region becomes not clickable/touchable.
NOTE: This backouts D67521 which is the patch for the test. If the
resolution is not 1.0, it's fine to skip the midpoint tests on Android
because it's tested on the desktop platforms within the decimal world.
Differential Revision: https://phabricator.services.mozilla.com/D261223
The found frame may be covered by another frame. However, current
implementation may use the covered frame as the target if it's the
nearest frame from the click/touch point.
A frame is covered by another frame means that web apps try to prevent
the users to click/tap the frame. Therefore, current behavior may make
it possible to click/tap an element which is tried to hidden. Thus,
current behavior may be not expected by both the web app developers and
the users.
This patch uses a part of intersection observer to get clipped border
box of a candidate frame. So, if the frame is clipped by the
containers, the hidden region becomes not clickable/touchable.
Differential Revision: https://phabricator.services.mozilla.com/D261223
aRootBounds is meaningless in that case, since it's the top level
viewport bounds, not the iframe's viewport bounds (which might be a lot
taller).
Differential Revision: https://phabricator.services.mozilla.com/D262575
Most changes are automated. Relevant bit is dom/.clang-format. I had to
manually disable clang-format in a few places where importing external
files and so on that aren't self-contained, and fix a couple includes /
forward declarations.
Differential Revision: https://phabricator.services.mozilla.com/D256800
What goes on here is that there's a couple of unfortunate style change
sequences which end up making us not do the EffectsInfo dance correctly.
Twitter uses (maybe didn't use to, which would explain the regression) a
visibility: hidden, out-of-flow iframe for a bit (which we correctly
throttle). But then they switch to an in-flow, visible, zero-height
iframe. That we should _not_ throttle. However, we end up not getting to
the display list code at all, because nsBlockFrame decides that we don't
need to descend into an empty line[1].
It seems less error prone to re-use the IntersectionObserver timing and
computation to determine whether the iframe is visible. That completely
matches in-process iframes, too.
Removing the empty frame border and putting them on an empty line in
dom/base/test/test_bug1639328.html is enough to reproduce the issue
without this patch.
[1]: https://searchfox.org/mozilla-central/rev/fe2743c6c5c708061c7f6504b26958fcc815bb4a/layout/generic/nsBlockFrame.cpp#7569-7579
Differential Revision: https://phabricator.services.mozilla.com/D207479
What goes on here is that there's a couple of unfortunate style change
sequences which end up making us not do the EffectsInfo dance correctly.
Twitter uses (maybe didn't use to, which would explain the regression) a
visibility: hidden, out-of-flow iframe for a bit (which we correctly
throttle). But then they switch to an in-flow, visible, zero-height
iframe. That we should _not_ throttle. However, we end up not getting to
the display list code at all, because nsBlockFrame decides that we don't
need to descend into an empty line[1].
It seems less error prone to re-use the IntersectionObserver timing and
computation to determine whether the iframe is visible. That completely
matches in-process iframes, too.
Removing the empty frame border and putting them on an empty line in
dom/base/test/test_bug1639328.html is enough to reproduce the issue
without this patch.
[1]: https://searchfox.org/mozilla-central/rev/fe2743c6c5c708061c7f6504b26958fcc815bb4a/layout/generic/nsBlockFrame.cpp#7569-7579
Differential Revision: https://phabricator.services.mozilla.com/D207479
This fixes src loads to be consistent with srcset/picture loads, modulo
the special synchronous case in the spec
(https://html.spec.whatwg.org/#update-the-image-data step 7), which
requires src loads to be sync if the image is available.
We now avoid triggering the load from the parser consistently for src /
srcset / picture, and unify the codepath with BindToTree. That should
avoid some useless task allocations.
Only the sync load code-path needs a script runner (mostly to deal with
anonymous content like the video poster <img> and such, but it also
helps not trigger sync loads at unexpected times like on adoption).
About the HTMLImageElement::Complete() getter change, we need to also
return false if there's an existing load task. That is the proposal in
https://github.com/whatwg/html/issues/4884, and prevents some failures
in the-img-element/{update-src-complete,img.complete}.html WPTs. It
technically changes our behavior on .srcset changes, but it makes it
consistent with .src changes and other browsers, so seems fine.
There are a couple regressions in CSP tests and the networkEvent stubs,
but these are really a pre-existing issue. What happens is that, since
the loads are now async, CSP can't figure out the script that triggered
the load anymore. I need to look if there's an easy way to propagate
that information in the image load tasks, but this is trivially
reproducible by changing these tests to use srcset rather than src.
The rest of the test changes are as expected: either new passes, or
expected test changes from this.
Differential Revision: https://phabricator.services.mozilla.com/D215519
Chrome uses bimodal layout viewport for IntersectionObserver;
a) While the dynamic toolbar is visible, Chrome uses the layout viewport
(which doesn't include the dynamic toolbar area)
b) While the dynamic toolbar is completely hidden, Chrome uses the layout
viewport + the height that the dynamic toolbar was originally occupied.
In this change we align the behavior with Chrome's behavior.
Differential Revision: https://phabricator.services.mozilla.com/D212920
UpdateIntersectionObservation takes some time (see profile in blocked
bug). It it needs to do multiple hashmap lookups unnecessarily.
We can store the observation state in the IntersectionObserver object
and remove the bespoke intersectionobserverlist property, using
BindObject / UnbindObject to keep stuff alive, like ResizeObserver and
MutationObserver do.
Differential Revision: https://phabricator.services.mozilla.com/D215370
This also makes `PhysicalAxis::{Horizontal,Vertical}` more useful when we
previous needed `PhysicalAxes::{Horizontal,Vertical}`.
Differential Revision: https://phabricator.services.mozilla.com/D211941
Also, in EventStateManager, convert `ComputeScrollTarget()` and
`ComputeScrollTargetAndMayAdjustWheelEvent()` to return `ScrollContainerFrame*`
so that we can drop more unnecessarily `do_QueryFrame`.
Due to removing `nsIScrollableFrame` forward declaration in DOM headers, we have
to add `nsIScrollableFrame` forward declaration temporarily in
nsMenuPopupFrame.h to make this patch compile, but we'll remove it once we
remove `nsIScrollableFrame` under layout/.
Differential Revision: https://phabricator.services.mozilla.com/D211496
`PresShell::GetRootScrollFrameAsScrollable()` is equivalent to
`PresShell::GetRootScrollContainerFrame()`.
In ScrollContainerFrame.h, `DecideScrollableLayer()` has two versions, one has
four parameters, and the other has five parameters with the fifth parameter
`aDirtyRectHasBeenOverriden` having a default value `nullptr`. When we switch
the caller from `nsIScrollableFrame` to `ScrollContainerFrame`, we need to
remove the default value for the fifth parameter to avoid ambiguity.
Differential Revision: https://phabricator.services.mozilla.com/D211494
In theory, changing the return type from `nsIFrame*` to `ScrollContainerFrame*`
exposes `ScrollContainerFrame` to the callers who might not needed, but almost
all of the callers in cpp files are already exposed to `nsIScrollableFrame`, as
demonstrated in this patch via replacing the #include from
"nsIScrollableFrame.h" to "ScrollContainerFrame.h", so this is OK.
Some callers can be simplified since we no longer need `do_QueryFrame` to
`nsIScrollableFrame`.
Differential Revision: https://phabricator.services.mozilla.com/D211488
This also makes `PhysicalAxis::{Horizontal,Vertical}` more useful when we
previous needed `PhysicalAxes::{Horizontal,Vertical}`.
Differential Revision: https://phabricator.services.mozilla.com/D211941
This also makes `PhysicalAxis::{Horizontal,Vertical}` more useful when we
previous needed `PhysicalAxes::{Horizontal,Vertical}`.
Differential Revision: https://phabricator.services.mozilla.com/D211941
Also, in EventStateManager, convert `ComputeScrollTarget()` and
`ComputeScrollTargetAndMayAdjustWheelEvent()` to return `ScrollContainerFrame*`
so that we can drop more unnecessarily `do_QueryFrame`.
Due to removing `nsIScrollableFrame` forward declaration in DOM headers, we have
to add `nsIScrollableFrame` forward declaration temporarily in
nsMenuPopupFrame.h to make this patch compile, but we'll remove it once we
remove `nsIScrollableFrame` under layout/.
Differential Revision: https://phabricator.services.mozilla.com/D211496
`PresShell::GetRootScrollFrameAsScrollable()` is equivalent to
`PresShell::GetRootScrollContainerFrame()`.
In ScrollContainerFrame.h, `DecideScrollableLayer()` has two versions, one has
four parameters, and the other has five parameters with the fifth parameter
`aDirtyRectHasBeenOverriden` having a default value `nullptr`. When we switch
the caller from `nsIScrollableFrame` to `ScrollContainerFrame`, we need to
remove the default value for the fifth parameter to avoid ambiguity.
Differential Revision: https://phabricator.services.mozilla.com/D211494
In theory, changing the return type from `nsIFrame*` to `ScrollContainerFrame*`
exposes `ScrollContainerFrame` to the callers who might not needed, but almost
all of the callers in cpp files are already exposed to `nsIScrollableFrame`, as
demonstrated in this patch via replacing the #include from
"nsIScrollableFrame.h" to "ScrollContainerFrame.h", so this is OK.
Some callers can be simplified since we no longer need `do_QueryFrame` to
`nsIScrollableFrame`.
Differential Revision: https://phabricator.services.mozilla.com/D211488
Daniel Holbert