| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
| |
The test discovered a serious bug where we for alien widgets on mac
returned a zero sized paintdevice. The reason is that we used to
ask the nsview backing the widget for its size, but with alien, there
where no such view. Very strange that this was not seen visually for
any of the many applications tested. This patch does the same
implementation strategy as found in QWidget::metric in qwidget_x11.cpp
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Moving the cursor to the corner of the desktop (0, 0) will typically
trigger things to happend in the OS, at least on Mac (spaces, synergy etc).
So I changed those tests to move it to (1, 1) instead. The reason to move
it to (0, 0) was just to get the mouse away from the window anyway.
And second, it takes some time after using QCursor::setPos on Mac
before the events from the move actually arrive. And many tests
that use this function relies on this. So I introduces a little
delay. This alone will fix a couble of tests.
|
| |\ |
|
| | |\
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/qt-water-staging:
Cocoa: non-native toolbar shows in window when hidden
Cocoa: disable scroll optimization for certain cases
Cocoa: fix scroll optimization bug
Cocoa: add scroll optimization when scrolling both hor and ver
Partly revert b1c715f8214233f5b573ed58fc89c9dd70beabb4
Cocoa: remove compiler warning
Add support for disabling touch on to enhance scrolling in Cocoa
Cocoa/Alien: replace depricated API
Cocoa/Alien: remove unneded code for scrolling with alien
Cocoa/Alien: keep the scrolling within correct bounds
Cocoa/Alien: bugfix scrolling, dirty region issue
Cocoa/Alien: let QWidget::scroll_sys handle overlapping widgets
Cocoa: enable more autotests for qwidget as a result of Alien
Cocoa/Alien: fix qwidget autotest (setClearAndResizeMask)
Cocoa: fix qwidget autotest (optimizedResizeMove) for raster engine
Alien implementation for the Cocoa port
|
| | | |\
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/qt-water-staging:
Cocoa: non-native toolbar shows in window when hidden
Cocoa: disable scroll optimization for certain cases
Cocoa: fix scroll optimization bug
Cocoa: add scroll optimization when scrolling both hor and ver
Partly revert b1c715f8214233f5b573ed58fc89c9dd70beabb4
Cocoa: remove compiler warning
Add support for disabling touch on to enhance scrolling in Cocoa
Cocoa/Alien: replace depricated API
Cocoa/Alien: remove unneded code for scrolling with alien
Cocoa/Alien: keep the scrolling within correct bounds
Cocoa/Alien: bugfix scrolling, dirty region issue
Cocoa/Alien: let QWidget::scroll_sys handle overlapping widgets
Cocoa: enable more autotests for qwidget as a result of Alien
Cocoa/Alien: fix qwidget autotest (setClearAndResizeMask)
Cocoa: fix qwidget autotest (optimizedResizeMove) for raster engine
Alien implementation for the Cocoa port
|
| | | | |\
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/qt-water-staging:
Cocoa: non-native toolbar shows in window when hidden
Cocoa: disable scroll optimization for certain cases
Cocoa: fix scroll optimization bug
Cocoa: add scroll optimization when scrolling both hor and ver
Partly revert b1c715f8214233f5b573ed58fc89c9dd70beabb4
Cocoa: remove compiler warning
Add support for disabling touch on to enhance scrolling in Cocoa
Cocoa/Alien: replace depricated API
Cocoa/Alien: remove unneded code for scrolling with alien
Cocoa/Alien: keep the scrolling within correct bounds
Cocoa/Alien: bugfix scrolling, dirty region issue
Cocoa/Alien: let QWidget::scroll_sys handle overlapping widgets
Cocoa: enable more autotests for qwidget as a result of Alien
Cocoa/Alien: fix qwidget autotest (setClearAndResizeMask)
Cocoa: fix qwidget autotest (optimizedResizeMove) for raster engine
Alien implementation for the Cocoa port
|
| | |\ \ \ \
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
into master-integration
* 'staging-master' of scm.dev.nokia.troll.no:qt/qt-lighthouse: (69 commits)
Lighthouse: License headers to new files in testlite
Make sure we blit the fbo on flush in QGLWindowSurface
Lighthouse: Wayland. Make the wayland integration closer to Lighthosue
Lighthouse: Wayland, only make one fbo for the WaylandPaintDevice
Making clearer separation between responsibility of different classes
Make it possible to vertically mirror gl painting
Lighthouse:Wayland Moving some logic into files
Fix Wayland plugin to work with Wayland after some interfaces changed
Make QGLContext::fromPlatformGLContext show correct sharing
Remove Lighthouse specific code from QGLWindowSurface
Fix X11 clipboard bug.
Fix for uninitialized member in QWaylandCursor
wayland: use pkgconfig for libdrm in wayland.pro
wayland: remove non-public header from config.tests
wayland: fix SOURCES to point to wayland.cpp in config.tests
Wayland: request rbg and premultiplied argb visuals as needed
Wayland: clamp window resizes to screen size
Wayland: split GL code into separate files
Wayland: use correct viewport for swapBuffers and correct coords
Wayland: fix geometry of swapBuffers
...
|
| | | |\ \ \ \ |
|
| | | | | | | | |
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Reviewed-by: sroedal
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Today there is a connection between QWidget and QPlatformScreen. So I
added a accessor in QWaylandScreen to get the corresponding
QWaylandDisplay. So now it should be possible to have different
QWaylandDisplays. But also its possible to use the internal Lighthouse
api closer in the wayland plugin if that should be of interest ;)
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Actually the fbo belongs to the QWaylandPaintDevice, which also holds n
QWaylandDrmBuffers. Then the QWaylandDrmWindowSurface gets the current
drmbuffer used on flush and attaches it to the window and calls
wl_surface_damage
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Ie. Moving code into different files and refactoring some of the
classes. The Drmbuffer was most affected.
It has a depth and stencil buffer.
QWaylandGLContext contain the context. The drm buffer class has to
be bound to the context before used.
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
The GLPaintDevice gets a flip property. If this is set, the pvr matrix
is altered to paint "upside down"
Reviewed-by: sroedal
|
| | | | | | | | |
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
such as
wl_display_create -> wl_display_connect
wl_surface_attach (attributes)
wl_surface_map -> wl_surface_map_toplevel
wl_shell_listener
wl_output_listener
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
information. Prior to this, if a context was created with QGLWidget it
would setup the sharing correct, but all sharing actually created on a
QPlatformGLContext level was ignored by the QGLContext, thus indicating
that it wasnt sharing with any other context.
|
| | | | | | | | |
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Truth be told, the x11 clipboard integration in testlite was done rather
quickly, thus leaving some cornercases open
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Without this patch Qt applications will crash the moment the mouse pointer
enters the window.
Signed-off-by: Rolf Offermanns <roffermanns@sysgo.com>
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
On this system the IncludePath for drm.h was missing
So maybe pkgconfig should be additionally also used for
libxkbcommon, EGL and GLESv2 to prevent such problems.
but i dont know whether its common to use pkg-config in qmake
|
| | | | | | | | |
|
| | | | | | | | |
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
When creating DRM buffers we should try to match the visual type
requested with the buffer format.
|
| | | | | | | | |
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
For clarity and to prevent merge conflicts etc. in future.
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
We're drawing into the parent, so we need to use the parent's geometry
for the GL viewport, or we may not be able to draw it. We also need to
use the parent width & height when checking whether to normalize to 1.0
or -1.0 in the coord transform calculations.
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
If the geometry isn't mapped 1:1 we need to divide the bottom by height,
not width, to calculate the texture rectangle.
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Pull the swap fbo and rbo into the context and make sure to destroy them
when the context goes away. Add unfortunate lazy allocation of the fbo
and rbo because we need them to be part of the right context.
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Make sure we copy from the private surface into the parent surface at
swapBuffers time.
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
We'll need this to handle GL widgets.
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
This will let us source from offscreen GL widget surfaces and copy into
the parent surface.
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Even GL widgets will allocate window surfaces, so share everything
there. This still leaves a hole when we mix GL widgets and the raster
back end (the created platform GL context won't have a DRM buffer to use
for its objects & context), but that won't work right now anyway...
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
GL widgets need a platform GL context. Add one to get hellogl_es2
limping along.
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
This will set mUseOpenGL when -platform waylandgl is passed to clients.
|
| | | | | | | | |
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Works with analogclock, draws upside down, fails to resize, and doesn't
show GL widget with hello_es2.
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Unused variables, debug output.
|
| | | | | | | | |
|
| | | | | | | | |
|
| | | | | | | | |
|
| | | | | | | | |
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
The surface is destroyed when the window is hidden, so we can't use that
for looking up the widget to send events to.
|
| | | | | | | | |
|
| | | | | | | | |
|
| | | | | | | | |
|
| | | | | | | | |
|
| | | | | | | | |
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
(configure has trouble separating host and target platforms.)
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Remove code which was wrong
|
