| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
|
|
|
|
| |
BUG=None
TEST=None
Review URL: http://codereview.chromium.org/7906015
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@102011 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is for Layers that participate in Draw() ordering, transforming and clipping, but do not themselves actually render anything.
Also changes LayerTest to use RunAllPending() rather than using the weird QuitOnComposited test type I had.
BUG=none
TEST=see unittest
Review URL: http://codereview.chromium.org/7941016
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@101829 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When the View is hidden, we destroy the associated Layer, but the external texture should persist.
Similarly, if a View is made visible, we create a Layer and the Layer should have that external texture.
The solution is to push the external texture logic and state up into the LayerHelper, which persists.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/7904024
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@101388 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
This is a nearly-mechanical translation of wstring to string16.
Removes a *ton* of string conversions.
Review URL: http://codereview.chromium.org/7886023
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@101329 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
We don't want to SetCanvas (or prepare a canvas to be set), if the Layer is being updated externally.
BUG=none
TEST=TOUCH_UI build displays google.com
Review URL: http://codereview.chromium.org/7889035
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@101080 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
LayerDelegate.
http://crbug.com/93947
TEST=none
Review URL: http://codereview.chromium.org/7845033
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@100961 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
Nothing (much) renders. WIP.
http://crbug.com/93947
TEST=none
Review URL: http://codereview.chromium.org/7849004
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@100773 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
about:rotate?right. The about handler fires orientation events that the TouchBrowserFrameView is listening for. When notified, the TBFV either rotates itself, or the views desktop (if it is active). The animation code lives in views/animation and knows nothing about sensors.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/7273073
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@100148 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
R=ben@chromium.org
TBR=ben@chromium.org
BUG=89439
TEST=none
Review URL: http://codereview.chromium.org/7710003
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@97777 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fixes TouchFrame's NonClientHitTest to ignore VirtualKeyboard area.
Move mouse capture logic to Window Manager so that nested
mouse capture with nested NWVs works correctly.
Note1: This is a tentative WM that allows us to move/resize window in views desktop until we have real window manager based on aura/layer API.
Note2: There is an compositor related issue and doesn't work well when compositor is enabled. I'll look into it after this CL.
Review URL: http://codereview.chromium.org/7530017
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@97492 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
ClearNativeFocus in View::OnFocus is triggering 'this' to be
deleted. I couldn't readily find any views that do anything
interesting when the native focus changes, but this should help us
find out if that's true.
BUG=89439
TEST=none
R=ben@chromium.org
Review URL: http://codereview.chromium.org/7670077
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@97412 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
an external texture.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/7605024
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96177 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A simple 3D CSS benchmark shows a 20% improvement in framerate on T25 hardware: an increase from about 30 fps to 36 fps.
This is a first pass to getting rid of double painting by introducing a hole in layers Currently a layer which wants its parent not to paint under it will call SetParentLayerHoley
There are a number of questions in terms of implementation:
1) We are doing this with layers. We are currently setting all layers to be holey if the touch_ui flag is enabled. Another option is to merge the holey functionality into regular layers. (Get rid of the holey layer subclass)
2) We are always generating vertices matrix, even in the ui::Layer case, is this ok?
3) We are wasting video memory. We are still uploading a full bitmap. I wasn't able to find a function call which would be able to efficiently slice the bigger bitmap, notably changing the origin of where the painting starts (texture upload)
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@91848 0039d316-1c4b-4281-b951-d872f2087c98
BUG=
TEST=
Review URL: http://codereview.chromium.org/7330025
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@94677 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
Fix TestTexture and ViewLayerTest compile.
BUG=90548
TEST=none
Review URL: http://codereview.chromium.org/7470030
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@94241 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
Even Chrome OS builds of chromium.
BUG=none
TEST=build with GYP_DEFINES="chromeos=1 touchui=1 use_ibus=0" on the desktop
Review URL: http://codereview.chromium.org/7396028
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@93001 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/7374007
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@92711 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
| |
BUG=72040
TEST=none
Review URL: http://codereview.chromium.org/7349021
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@92705 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
We use context sharing in CompositorGL to avoid compiling the same shaders for different compositors. Previously, we kept the shaders around until the program exited. With this change, we keep the shaders around until there are no more compositors.
BUG=89239
TEST=ui_tests on linux_views pass
Review URL: http://codereview.chromium.org/7367005
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@92683 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Also, rename it to GetViewsInGroup().
BUG=72040
TEST=views_unittests --gtest_filter=ViewTest.GetViewByID
R=ben@chromium.org,sky@chromium.org
Review URL: http://codereview.chromium.org/7328031
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@92430 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
BUG=89239
TEST=linux_view trybot cycles green
Review URL: http://codereview.chromium.org/7362002
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@92422 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
BUG=none
TEST=none
TBR=sky
Review URL: http://codereview.chromium.org/7352001
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@92305 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
BUG=
TEST=ui_unittest
Review URL: http://codereview.chromium.org/7044062
Patch from Ian Vollick <vollick@chromium.org>.
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@91855 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Steps for Repro. From a touch build
1) Goto google.com
2) Type into text field
3) Typed text does not show up
BUG=87840
TEST=views_unittest (ToggleVisibilityWithLayer)
Review URL: http://codereview.chromium.org/7273067
Patch from Peter Kotwicz <pkotwicz@chromium.org>.
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@91563 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
BUG=none
TEST=covered by unit test.
R=wjmaclean@chromium.org
Review URL: http://codereview.chromium.org/7210048
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@91317 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
BUG=none
TEST=none
R=ben@chromium.org
Review URL: http://codereview.chromium.org/7280002
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@91312 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/7276053
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@90952 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
Also, make sure the touch_pressed_handler_ is the view being removed before resetting it to NULL.
BUG=none
TEST=things compile and existing tests continue to pass
Review URL: http://codereview.chromium.org/7276002
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@90762 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
the layer we force the complete bounds to be painted. This way the
Compositors don't have to worry about partial update of a size they've
never seen. This matches old logic I removed thinking it was no longer
needed.
BUG=none
TEST=none
R=ben@chromium.org
Review URL: http://codereview.chromium.org/7272013
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@90711 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
BUG=none
TEST=none
R=ben@chromium.org
Review URL: http://codereview.chromium.org/7242025
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@90589 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
This will be used to display texture produced by the GPU process for the RWHVV via the browser compositor.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/7231012
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@90362 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
in a couple of cases.
BUG=none
TEST=none
R=ben@chromium.org
Review URL: http://codereview.chromium.org/7247020
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@90333 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
BUG=none
TEST=none
R=ben@chromium.org,wjmaclean@chromium.org
Review URL: http://codereview.chromium.org/7242005
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@90254 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
BUG=72040
TEST=None
R=ben@chromium.org,sky@chromium.org
Review URL: http://codereview.chromium.org/7238006
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@90224 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
BUG=72040
TEST=None
R=ben@chromium.org
Review URL: http://codereview.chromium.org/7202015
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@89921 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
BUG=none
TEST=none
R=ben@chromium.org,wjmaclean@chromium.org
Review URL: http://codereview.chromium.org/7189067
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@89880 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
views.
BUG=none
TEST=the close button of the rotated window highlights on hover.
Review URL: http://codereview.chromium.org/7193025
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@89501 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
BUG=None
TEST=views_unittests
R=sky@chromium.org
Review URL: http://codereview.chromium.org/7191028
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@89422 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
BUG=86343
TEST=none
Review URL: http://codereview.chromium.org/7184024
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@89403 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The z-ordering of child views is currently maintained (implicitly?) by the order
they are added to the parent view. It may often be necessary to change the
z-order of views, and View::ReorderChildView can be used to do that.
I changed views_desktop to have the two pure-views windows to have overlapping
regions, and View::ReorderChildView is used to change the z-ordering of the two
windows. So it is possible to click a window to activate it and bring it on top
of the other. It could have been done without adding Widget::MoveToTop, but
given Widget::MoveAbove and MoveAboveWidget, it sounded like a good idea to have
Widget::MoveToTop (MoveToFront?). I do now know what the implementation for
windows should be, though. So left it as NOTIMPLEMENTED.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/7185005
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@89363 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Move private NativeWidget methods to new internal interface
NativeWidgetPrivate.
This should make it harder to abuse accidentally.
BUG=72040
TEST=none
Review URL: http://codereview.chromium.org/7170019
Review URL: http://codereview.chromium.org/7189012
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@89357 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
NativeWidgetPrivate.
This should make it harder to abuse accidentally.
BUG=72040
TEST=none
Review URL: http://codereview.chromium.org/7170019
TBR=ben@chromium.org
Review URL: http://codereview.chromium.org/7097016
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@89232 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
This should make it harder to abuse accidentally.
BUG=72040
TEST=none
Review URL: http://codereview.chromium.org/7170019
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@89216 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
ui::TouchStatus.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/7147005
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@89079 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
View::parent() function.
While I'm here:
- Rewrite some while loops into for loops, since that is what they are doing anyway.
BUG=72040
TEST=None
R=sky@chromium.org
Review URL: http://codereview.chromium.org/7046102
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@89000 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
accessibility tree again, which makes it possible to debug Chrome
accessibility on Windows using tools like AccExplorer32 or AccProbe.
This change refactors the code that retrieves the Windows IAccessible
interface from a View that wraps a native HWND. The previous code was
brittle, using ViewProps and hardcoded View classname tests. This
change moves that logic directly to the classes that wrap native
views.
BUG=54220
BUG=85673
TEST=Updated AccessibilityWinBrowserTest.ContainsRendererAccessibilityTree
Review URL: http://codereview.chromium.org/6995126
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@88913 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
| |
BUG=72040
TEST=none
Review URL: http://codereview.chromium.org/7108047
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@88905 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
understand why we can't create tooltips for some folks, but this
should at least keep us from crashing.
BUG=82193
TEST=none
R=ben@chromium.org
Review URL: http://codereview.chromium.org/6995157
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@88897 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
BUG=Browser compositor is on by default, not wanted at this time.
TEST=None.
Review URL: http://codereview.chromium.org/7046108
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@88831 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
No other boolean variable in the View class has the "is_" prefix and this is
making me nervous for a while now :)
BUG=72040
TEST=None
R=ben@chromium.org
Review URL: http://codereview.chromium.org/7104098
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@88748 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
BUG=none
TEST=compiles, run manually
Review URL: http://codereview.chromium.org/6975051
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@88671 0039d316-1c4b-4281-b951-d872f2087c98
|