Similar to the optional aContainingBlockSize parameter, both border and
padding should use logical coordinates in ReflowInput::mFrame's writing
mode.
Table frames that need to override border and padding can be simplified a bit.
However, DR_init_constraints_cookie and DR_init_offsets_cookie become more
complex, but they're only for debugging. I'm not planning to update their
internal APIs.
Differential Revision: https://phabricator.services.mozilla.com/D95369
This should make the optimization landed earlier in this bug apply for
some of the NotifyThemeChanged() calls in nsWindow.cpp which are causing
all the extra invalidations.
If we know that system colors/fonts didn't change, we can avoid doing a
bunch of reflow work and the patch from earlier in the bug can avoid
re-rasterizing images too.
Differential Revision: https://phabricator.services.mozilla.com/D94425
This is mostly a revert of the patch in bug 1425686 that removed the old
probe, but rebased to new code locations and clang-formatted. The histogram
entry is also updated with new bug numbers and fields.
The next patch will refine some of these telemetry recording points; the patch
is split into two for easier reviewing as this part is basically what landed
originally.
Differential Revision: https://phabricator.services.mozilla.com/D92995
when there is no composition, `IMEStateManager::HandleSelectionEvent()` calls
`TextComposition::HandleSelectionEvent()` with `BrowserParent` which is
retrieved from `aEventTargetContent` which is focused content in the main
process. This means that `eSetSelection` event is handled in root document
of the focused tab when there is no composition. However, following composition
events will go to focused `BrowserParent`, i.e., if an OOP iframe has focus
in the tab, `eSetSelection` event won't be handled in it. Therefore, IME
always fails to replace existing text with new composition.
This patch makes both `IMEStateManager::HandleSelectionEvent()` and
`PresShell::EventHandler::DispatchEventToDOM()` use
`IMEStateManager::GetActiveBrowserParent()` for sending both `eSetSelection`
and composition events to same process. Additionally, this patch make
`IMEStateManager::GetActiveBrowserParent()` returns
`IMEStateManager::sFocusedIMEBrowserParent` when it's set to non-nullptr
because native IME tries to modify the editor which enabled IME context.
On the other hand, for making the behavior safer, we should make
`eCompositionStart` event have replacing range optionally. I filed
bug 1669907 to change the design, but it requires several non-tiny patches.
Once we fix it, we can write automated tests for this simply, but without
the fix, I need to write too many lines with low level API, and it becomes
unnecessary after fixing the bug. Therefore, I give up to write a test
for this for now.
Differential Revision: https://phabricator.services.mozilla.com/D93053
This patch tries to mark root callers of `nsINode::GetSelectionRootContent()`
which calls `nsINode::GetAnonymousRootElementOfTextEditor()` as far as possible
(and reasonable).
It's used by `ContentEventHandler` so that a lot of methods of
`EventStateManager`, `ContentEventHandler`, `IMEContentObserver` which are main
users of it are also marked as `MOZ_CAN_RUN_SCRIPT`. I think that this is
reasonable.
On the other hand, it might not be reasonable to mark `IMEStateManager` methods
as `MOZ_CAN_RUN_SCRIPT` for initializing `IMEContentObserver` because
`IMEStateManager` may be able to initialize `IMEContentObserver` asynchronously
and its root callers are in XUL layout code. Therefore, this patch uses
`MOZ_CAN_RUN_SCRIPT_BOUNDARY` for `IMEStateManager` at least for now.
Differential Revision: https://phabricator.services.mozilla.com/D92730
This patch tries to mark root callers of `nsINode::GetSelectionRootContent()`
which calls `nsINode::GetAnonymousRootElementOfTextEditor()` as far as possible
(and reasonable).
It's used by `ContentEventHandler` so that a lot of methods of
`EventStateManager`, `ContentEventHandler`, `IMEContentObserver` which are main
users of it are also marked as `MOZ_CAN_RUN_SCRIPT`. I think that this is
reasonable.
On the other hand, it might not be reasonable to mark `IMEStateManager` methods
as `MOZ_CAN_RUN_SCRIPT` for initializing `IMEContentObserver` because
`IMEStateManager` may be able to initialize `IMEContentObserver` asynchronously
and its root callers are in XUL layout code. Therefore, this patch uses
`MOZ_CAN_RUN_SCRIPT_BOUNDARY` for `IMEStateManager` at least for now.
Differential Revision: https://phabricator.services.mozilla.com/D92730
The only caller passed in DisplayRelativeTo::ScrollFrame.
Removing this makes it easier to consolidate display port options
in a subsequent patch.
Differential Revision: https://phabricator.services.mozilla.com/D92014
The frame we pass to PageMove is used to extend the selection, we can't extend a selection outside of the document. The scrolling that takes place can still scroll scroll frames in parent documents afterwards.
Differential Revision: https://phabricator.services.mozilla.com/D92176
The current formulation is inconsistent. If you call it with ScrollableDirection::Either it goes into nsLayoutUtils::GetNearestScrollableFrame which will return the first scroll frame with a non-hidden overflow. If you call it with any other ScrollableDirection it calls nsLayoutUtils::GetNearestScrollableFrameForDirection which returns the first scrollframe it finds that has non-hidden overflow AND has at least one dev pixel of scroll range.
So remove that function and call nsLayoutUtils::GetNearestScrollableFrameForDirection directly. This is a slight change of behaviour but it seems desirable for all callers.
Differential Revision: https://phabricator.services.mozilla.com/D91236
The problems that it causes that are referred to in the patch is that we use the vvoffset and GetScrollRangeForUserInputEvents to compute the scrollbar cur, min, and max attributes. If we have a non-zero vvoffset when we are zooomed backed out, GetScrollRangeForUserInputEvents will be a zero range and so the cur attribute (non-zero) will be larger than the max attribute (0).
This then causes nsSliderFrame::AttributeChanged to call ScrollByWhole at
https://searchfox.org/mozilla-central/rev/30e70f2fe80c97bfbfcd975e68538cefd7f58b2a/layout/xul/nsSliderFrame.cpp#204
Which we definitely do not want.
Differential Revision: https://phabricator.services.mozilla.com/D90372
Otherwise the only time it gets set is when the ZoomConstraintsClient specifically sets it in response to one of the things that can change it. But if the scroll frame gets reconstructed and one of those things don't happen it will be wrong.
This shows up with desktop zooming scrollbars because we check mZoomableByAPZ (or WantAsyncScroll()) before we use the apz scroll path in ScrollFrameHelper::ScrollBy().
Differential Revision: https://phabricator.services.mozilla.com/D90384
Otherwise the only time it gets set is when the ZoomConstraintsClient specifically sets it in response to one of the things that can change it. But if the scroll frame gets reconstructed and one of those things don't happen it will be wrong.
This shows up with desktop zooming scrollbars because we check mZoomableByAPZ (or WantAsyncScroll()) before we use the apz scroll path in ScrollFrameHelper::ScrollBy().
Differential Revision: https://phabricator.services.mozilla.com/D90384
eTopLevel is reused in content process to indicates that the mouse leaves
the puppet widget rendering area, now we add a separated type, ePuppet, for it.
Differential Revision: https://phabricator.services.mozilla.com/D84748
eTopLevel is reused in content process to indicates that the mouse leaves
the puppet widget rendering area, now we add a separated type, ePuppet, for it.
Differential Revision: https://phabricator.services.mozilla.com/D84748
The abstract observer base classes are moved to a separate header file
nsRefreshObservers.h and the includes are adjusted accordingly.
Some method implementations are moved to the corresponding implementation files
to avoid the need to include the nsRefreshDriver.h file in the header.
Differential Revision: https://phabricator.services.mozilla.com/D85764
nsPreContext::mDynamicToolbarHeight is representing the state of the dynamic
toolbar in transition or is being collapsed. So if we leave the value as it is,
we misrecognize the state, the visual viewport width/height are significantly
affected by the misrecognition, it returns 0 because in the collapsed or the
transition state, it returns the special value [1] which is never properly set
without calling nsPresContext::UpdateDynamicToolbarOffset.
This mismatching state doesn't much matter in the real sites since we don't
often change the max height of the dynamic toolbar but in automated tests it's
a big problem. In fact a test case which will be introduced in the next commit
will not work properly without this fix.
[1] https://searchfox.org/mozilla-central/rev/1b95a0179507a4dc7d4b0c94c2df420dc1a72885/dom/base/VisualViewport.cpp#78
Differential Revision: https://phabricator.services.mozilla.com/D83902
This is a straightforward conversion except that
`NS_SUBTREE_DIRTY(this)` can be written terser as `IsSubtreeDirty()`.
Differential Revision: https://phabricator.services.mozilla.com/D82811
There's two code changes in this patch:
- The update to the visual viewport that was happening just before positioning
the fixed items gets moved to happen after determining the scrollbars for
the root scrollframe. This moves it a little bit earlier, to basically the
earliest point at which the visual viewport can actually be computed, since
it depends on the presence of the root scrollframe's scrollbars.
More importantly, this change sets the visual viewport without checking to
see if one was already set, as the old code did. This means every reflow
of the root scrollframe on a presShell with an MVM will now have a visual
viewport set. Previously the visual viewport would only get set for the first
time when the MVM got a load or first-paint event, and then would get updated
for subsequent reflows. The net effect here is that the visual viewport is
set earlier, and this can sometimes eliminate extra reflows from after the
load event, because everything is already in a consistent state.
- The NotifyResizeReflow call to MVM is replaced by a NotifyReflow call that
runs before every reflow, instead of just on resizes. Note that the
NotifyReflow also doesn't update the visual viewport like NotifyResizeReflow
used to do, because that is taken care of by the above-mentioned code change
to set the visual viewport.
This is desirable because there are things that run during reflow that attempt
to read the display size from the MVM, and they were getting a zero size
for reflows that happened before the first resize or load/first-paint events.
Now they get a valid display size on every reflow, and so again this allows
fewer overall reflows as the code converges to a stable state faster.
Together these changes ensure that every reflow has access to up-to-date
properties (display size, mobile viewport size, visual viewport size) from the
MVM. This eliminates unnecessary reflows because of out-of-order computations
based on stale values and such. Therefore the number of reflows goes down,
which is reflected by the changes to the crashtest assertion counts.
Differential Revision: https://phabricator.services.mozilla.com/D81375
While it makes sense to do this when the visual viewport offset changes, it
doesn't make as much sense when the visual viewport size changes. The size can
change whenever non-overlay scrollbars are added or removed to the root
scrollframe, which is an operation that seems like it shouldn't interfere with
scroll anchoring at all.
Differential Revision: https://phabricator.services.mozilla.com/D81071
After the above, I don't think this is needed anymore, because we
shouldn't be looking at the widget size from layout.
It also shouldn't cause more reflows on desktop at least, because of the
early out in ResizeReflowIgnoreOverride before calling
SimpleResizeReflow().
Differential Revision: https://phabricator.services.mozilla.com/D80731
We did not pass FLAG_HIGH_QUALITY_SCALING down to the document used to
rasterize the SVG to a surface, resulting in embedded raster images to
not use high quality downscaling.
Differential Revision: https://phabricator.services.mozilla.com/D56422
There's a lot of documents that get created (about:blank and friends) and this
makes it easier to figure out which MVM we actually care about.
Differential Revision: https://phabricator.services.mozilla.com/D80085
This avoids arbitrary precision loss when computing REM units and so on,
which is particularly important if we ever change the base of our app
units (but useful regardless).
Differential Revision: https://phabricator.services.mozilla.com/D79928
There's a lot of documents that get created (about:blank and friends) and this
makes it easier to figure out which MVM we actually care about.
Differential Revision: https://phabricator.services.mozilla.com/D80085
There's no need to put this in nsLayoutUtils since it's so closely bound
to presShell anyway. So we move it from being a static nsLayoutUtils function
that takes a PresShell pointer, to being a method on the PresShell itself.
Two functional changes here:
1. Don't run the MaybeReflowForInflationScreenSizeChange code unless the
visual viewport actually changes
2. Run the MaybeReflowForInflationScreenSizeChange if the visual viewport
is reset.
These functional changes can be logically thought of as "group the font-
inflation reflow stuff together with the other visual-viewport-triggered
reflow stuff".
Depends on D80040
Differential Revision: https://phabricator.services.mozilla.com/D80041
There are two parts here. One is the "backstop" in the scrollframe's
ReflowFinished callback, that recomputes the visual viewport size if layout
scrollbars are being used in the root scrollframe. This ensures that the
visual viewport gets resized properly after a reflow, possibly at the expense
of a second reflow to reposition fixed-pos items.
There is also an update to the visual viewport during the reflow itself, after
we have reflowed the in-flow contents (including the root scrollframe) but
before we reflow the fixed-pos items. This allows us to avoid the second reflow
by using the new visual viewport for positioning the fixed-pos items correctly.
This early reflow also skips marking things for a second reflow.
This patch fixes a problem described in bug 1644271 comment 2, among other
things. Specifically, it ensures that when the scrollbar properties (e.g.
maxpos, minpos) are computed, they are computed using an up-to-date visual
viewport size. The up-to-date visual viewport size correctly excludes the space
taken up by non-overlay scrollbars, which wasn't happening without this patch.
Depends on D80039
Differential Revision: https://phabricator.services.mozilla.com/D80040
The existing comment in BrowserChild::RecvUpdateDimensions may have been
accurate at some point in the past, but I'm seeing cases where setting
the content viewer size itself triggers a reflow. Since the widget size
hasn't been updated yet, the reflow uses some stale values and produces
incorrect outcomes. This patch ensures both the content viewer and widget
get their sizes updated first, and only then do we do the reflow.
Differential Revision: https://phabricator.services.mozilla.com/D79885
This avoids arbitrary precision loss when computing REM units and so on,
which is particularly important if we ever change the base of our app
units (but useful regardless).
Differential Revision: https://phabricator.services.mozilla.com/D79928
For regular elements, whenever -moz-appearance is used, the CSS background is
ignored. Root elements were behaving specially, and the background color also
needed to be adjusted.
For example, for Windows 7, we have the following CSS rule;
```
:root {
background-color: transparent;
-moz-appearance: -moz-win-borderless-glass;
}
```
This change makes the root element more consistent with other elements, if non-glass
-moz-appearance values are used. For example, for platforms where `-moz-appearance: dialog`
has (partially) transparent rendering, the extra `background-color: transparent`
declaration is now no longer necessary.
This patch preserves the behavior for Windows 7 glass because there are several other
things that would need to be adjusted for the glass case (see bug 1601183 and bug 1599366).
Maybe we can just wait until we drop Windows 7 entirely, to clean this up.
This change does not let content documents opt out of forced opaqueness:
Root content documents still get an opaque background color from an existing
check further down in this method.
Differential Revision: https://phabricator.services.mozilla.com/D51459
This is important because if the apz.allow_zooming pref is enabled (and
eventually even without that pref), VisualViewportOnly-type MVMs do set
the visual viewport size based on the display size and zoom, and so need
to be updated when those change.
It's also important to keep the mMobileViewportSize updated and in sync
with the display size, because it may get used via ComputeIntrinsicResolution
when fullscreening.
Differential Revision: https://phabricator.services.mozilla.com/D79594
It does not make sense to ask for layout coordinates relative to
a chrome document. If the chrome document has a zoomed (RCD)
descendant in the same process, this means we do not apply the
visual-to-layout transform when later entering the RCD.
Depends on D79588
Differential Revision: https://phabricator.services.mozilla.com/D79589
This is important because if the apz.allow_zooming pref is enabled (and
eventually even without that pref), VisualViewportOnly-type MVMs do set
the visual viewport size based on the display size and zoom, and so need
to be updated when those change.
Differential Revision: https://phabricator.services.mozilla.com/D79594
Allowing the MVM to control the reflow means that the requested reflow size
is ignored, and instead the existing CSS/layout viewport is used. This is
undesirable for calls to SizeToContent(), where the intent is to do a reflow
to figure out the smallest amount of space the content fits in.
In general though unless we are using mobile viewport sizing we shouldn't be
needing the MVM to drive reflows.
Differential Revision: https://phabricator.services.mozilla.com/D79225
The MVM is needed for both handling of meta-viewport tags and APZ zooming.
However, the set of functionality needed in the two modes are not the same.
This patch adds a mechanism to create an MVM with a flag that lets it know
which mode it is operating in. Eventually we may want to split this into two
or more classes but for now this seems like a reasonable way forward.
The flag is currently set on the MVM on creation based on whether or not the
meta-viewport support is needed. There's no code that meaningfully *uses* the
flag yet, so this patch should have no functional change. The bulk of the
patch is ensuring that we appropriately destroy and re-create the MVM if the
flag required changes.
Differential Revision: https://phabricator.services.mozilla.com/D79224
This patch is a fairly mechanical conversion. The old `nullptr` gets converted
to ScrollOrigin::NotSpecified, and all the other possible values get corresponding
values in the new ScrollOrigin enum. A few switch statements are introduced to
clean up big if statements, but other than that, additional cleanups will happen
in later patches.
Differential Revision: https://phabricator.services.mozilla.com/D78438
This adjusts the position at which the drag images appear when doing drag
actions, so that they appear where you would expect when APZ zoom is applied.
There doesn't seem to be a good way to test this, but I did a bunch of manual
testing, with all the possible expansions of this sentence:
Dragging {a small image,a large image,some text} in {an iframe,the root
content document}, with {,no }zooming applied.
In all cases, the drag image/text should appear such that the part under the
cursor is the same as what was under the cursor on the original rendering of
the page.
Differential Revision: https://phabricator.services.mozilla.com/D77436
When rasterizing the drag image, we pick up the resolution from ancestor
presShells and ensure that the drag image is rasterized at that resolution,
with appropriate limits for memory usage.
Differential Revision: https://phabricator.services.mozilla.com/D77435
Make it show the contents of style sheets (as it used to before Stylo)
and make it work in --disable-debug --enable-layout-debugger builds.
Differential Revision: https://phabricator.services.mozilla.com/D76640
In favor of the NativeAnonymous versions which they forward to.
Done automatically with:
rg -l 'IsInAnonymousSubtree' | xargs sed -i 's/IsInAnonymousSubtree/IsInNativeAnonymousSubtree/g'
And removing the function definitions afterwards.
Differential Revision: https://phabricator.services.mozilla.com/D76681
This patch is generated by using my editor's rename functionality.
In the next patch, `nsIFrame::` prefix is going to be removed manually
from all the ChildLists() calls.
Differential Revision: https://phabricator.services.mozilla.com/D75893
This is the "core" change of the patch series, which causes most
existing layout codepaths to correctly factor in the visual to
layout transform (or its inverse), as long as the callers correctly
propagate it in the correct ViewportType.
Differential Revision: https://phabricator.services.mozilla.com/D68920
This "upgrades" various nsLayoutUtils functions which take as inputs
a set of coordinates and a frame that the coordinates are relative to,
to accept a RelativeTo object instead of a frame.
Most of the patch is just dumb propagation, but the few places where
we use an explicit ViewportType::Visual are important. There are
probably a few other places I've overlooked, but this seems to cover
the important ones that come up commonly.
There are undoubtedly other functions into which we can propagate
RelativeTo, in this patch I've propagated it as far as necessary
for my needs in this bug (mainly GetTransformToAncestor() and
GetEventCoordinatesRelativeTo()).
Differential Revision: https://phabricator.services.mozilla.com/D68919
Prior to this bug, it was necessary to handle non-e10s specially, because the
resolution was being unapplied at the process boundary, and in non-e10s there
was no process boundary.
The remaining patches in this bug move the resolution unapplication away from
the process boundary in all cases, making special handling for non-e10s
unnecessary.
Differential Revision: https://phabricator.services.mozilla.com/D68273
Even in comm-central and BlueGriffon, `nsISelectionController::*ForDelete()`
are not used. Therefore, we can remove them safely.
Differential Revision: https://phabricator.services.mozilla.com/D72296
This is the "core" change of the patch series, which causes most
existing layout codepaths to correctly factor in the visual to
layout transform (or its inverse), as long as the callers correctly
propagate it in the correct ViewportType.
Differential Revision: https://phabricator.services.mozilla.com/D68920
This "upgrades" various nsLayoutUtils functions which take as inputs
a set of coordinates and a frame that the coordinates are relative to,
to accept a RelativeTo object instead of a frame.
Most of the patch is just dumb propagation, but the few places where
we use an explicit ViewportType::Visual are important. There are
probably a few other places I've overlooked, but this seems to cover
the important ones that come up commonly.
There are undoubtedly other functions into which we can propagate
RelativeTo, in this patch I've propagated it as far as necessary
for my needs in this bug (mainly GetTransformToAncestor() and
GetEventCoordinatesRelativeTo()).
Differential Revision: https://phabricator.services.mozilla.com/D68919
Prior to this bug, it was necessary to handle non-e10s specially, because the
resolution was being unapplied at the process boundary, and in non-e10s there
was no process boundary.
The remaining patches in this bug move the resolution unapplication away from
the process boundary in all cases, making special handling for non-e10s
unnecessary.
Differential Revision: https://phabricator.services.mozilla.com/D68273
The existing infrastructure which stored cached BrowsingContexts on the
BrowsingContextGroup was added before WindowContexts were added, and can cause
racing issues with partially discarded trees during process switches.
Differential Revision: https://phabricator.services.mozilla.com/D71238
The existing infrastructure which stored cached BrowsingContexts on the
BrowsingContextGroup was added before WindowContexts were added, and can cause
racing issues with partially discarded trees during process switches.
Differential Revision: https://phabricator.services.mozilla.com/D71238
The existing infrastructure which stored cached BrowsingContexts on the
BrowsingContextGroup was added before WindowContexts were added, and can cause
racing issues with partially discarded trees during process switches.
Differential Revision: https://phabricator.services.mozilla.com/D71238
The existing infrastructure which stored cached BrowsingContexts on the
BrowsingContextGroup was added before WindowContexts were added, and can cause
racing issues with partially discarded trees during process switches.
Differential Revision: https://phabricator.services.mozilla.com/D71238
The existing infrastructure which stored cached BrowsingContexts on the
BrowsingContextGroup was added before WindowContexts were added, and can cause
racing issues with partially discarded trees during process switches.
Differential Revision: https://phabricator.services.mozilla.com/D71238
Users have much better, easier alternatives, like
DOMWindowUtils.{loadSheetUsingURIString,removeSheet}, which we use to
replace the only caller that exists in mozilla-central (the editor
element, which loads EditorOverride.css).
This allows to clean up the style system and editor. There are other
callers in comm-central, but it seems they can switch to DOMWindowUtils
trivially, as the DOMWindowUtils APIs also use the system principal and
thus they can load any URL.
I'll make sure to give them some time with the migration and/or help
out of course.
Differential Revision: https://phabricator.services.mozilla.com/D71263
We should always do this, otherwise stuff may not end up being visible which is
not acceptable for focus navigation.
Differential Revision: https://phabricator.services.mozilla.com/D70541
--HG--
extra : moz-landing-system : lando
Note that we intentionally don't move the SetDisplaySelection stuff to the
runnables. It would probably be safe enough, but it's not required and it makes
reasoning about this code harder.
Differential Revision: https://phabricator.services.mozilla.com/D70183
--HG--
extra : moz-landing-system : lando
This avoids a bunch of ugly casts and void pointers, without much overhead
(unlike std::function or such).
Differential Revision: https://phabricator.services.mozilla.com/D68182
--HG--
extra : moz-landing-system : lando
Converts dom.w3c_pointer_events.implicit_capture to a static pref, removes the initializer in PointerEventHandler.
Differential Revision: https://phabricator.services.mozilla.com/D67204
--HG--
extra : moz-landing-system : lando
This rejiggers a bit the way selection focus is handled so that focusing a
disabled form control with the mouse handles selection properly, and hides the
document selection and so on.
This matches the behavior of other browsers as far as I can tell.
Given now readonly and disabled editors behave the same, we can simplify a bit
the surrounding editor code.
Differential Revision: https://phabricator.services.mozilla.com/D66464
--HG--
extra : moz-landing-system : lando
Converts dom.w3c_pointer_events.implicit_capture to a static pref, removes the initializer in PointerEventHandler.
Differential Revision: https://phabricator.services.mozilla.com/D67204
--HG--
extra : moz-landing-system : lando
