| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We did this in 4.8 too, to avoid swamping the event queue with stale
motion events. If you're interested in fine grained motion events you
just need to have a responsive main loop. Most applications are only
interested in the latest mouse position in any case.
Fixes performance issues in QML where an onMouseXChanged or
onMouseYChanged in a mouse area does directly or indirectly runs some
heavy computations, slowing rendering to a halt since the main loop is
constantly busy.
Change-Id: I169c96458db4d57b689d6c2c915765b11c35e123
Reviewed-by: Jens Bache-Wiig <jens.bache-wiig@digia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Threaded rendering had issues due to bugs in xcb before
xcb_poll_for_queued_event() was added. Now xcb_poll_for_queued_event()
is resolved at runtime and if present xcb events are received on a
separate thread.
Change-Id: I05420c8c9f9a20f41ce0f86ed255bc8b295e7fe5
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
windowTitle, windowModality, windowIcon and so on are named that way
to be similar to the ones in QWidget. However QQuickWindow inherits
all of the declared properties, and we would like to have shorter
property names in QML. If you are working with a Window then it's
obvious the title property is the window title. Unfortunately,
there must be patches in many other modules which depend on this one.
In order to avoid the need to merge them all at the same time,
there is also patch https://codereview.qt-project.org/#change,39001
which temporarily adds backwards-compatible accessors, which can be
removed after the other modules are able to build without them.
We should not rename windowState to state, because in QML, state
usually drives the state machine for animation transitions etc.
(although QWindow is not an Item, a user might get confused about it).
Related patches are
https://codereview.qt-project.org/#change,39001
https://codereview.qt-project.org/#change,37764
https://codereview.qt-project.org/#change,37765
https://codereview.qt-project.org/#change,37766
https://codereview.qt-project.org/#change,37762
Change-Id: Ie4424ec15fbdef6b29b137f90a2ae33f173edd21
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Some of the xcb- libraries we depend upon are not (yet) common across
distributions. This is problematic for binaries that should be working
on different distributions. The patch mitigates this by:
Adding the files from
libxcb-proto (version 0.1.6), compiled with libxcb-1.5
xcb-util (version 0.3.9)
xcb-util-image (version 0.3.9)
xcb-util-keysyms (version 0.3.9)
xcb-util-renderutil (version 0.3.8)
xcb-util-wm (version 0.3.9)
from xcb.freedesktop.org/dist to src/3rdparty/xcb.
Adding a configure option '-qt-xcb' to use the sources instead of
linking to the respective runtime libraries.
Task-number: QTBUG-27803
Change-Id: I6ea87daa382871b2b9072a601511523fa0b9f44b
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
| |
By not making this a compile time decision we ensure forward
compatibility for older xcb versions if the xcb plugin is built against
a newer xcb.
Change-Id: I744777d53bf7b8deb6eff372494f4403d19d364c
Reviewed-by: Kai Koehne <kai.koehne@digia.com>
|
|
|
|
|
|
|
|
| |
Bring back code from keymapper_x11.cpp.
Task-number: QTBUG-27648
Change-Id: I2a189a0537bcd2ae0eb093576c84814b2f09933b
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
| |
Change-Id: Ib9a0f96007e6dc5e125bfcd39bd1ef09baebae3b
Signed-off-by: Uli Schlachter <psychon@znc.in>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
| |
The function createDummyWindow() allocated a colormap, but never freed
it. Freeing it is safe after the window is created.
Change-Id: I4c876568572c9e1e9dde7047850a51917ef3702f
Signed-off-by: Uli Schlachter <psychon@znc.in>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When the XCB connection breaks, all following XCB function calls will fail and
function calls that are currently in progress will return. This means that
xcb_wait_for_event() will return NULL and thus the QXcbEventReader thread will
exit.
This patch uses the above behavior to make sure that processXcbEvents() will be
called. The error handling should always be done before normal event processing,
because all XCB calls will fail once the connection is in an error state. This
is especially unexpected for xcb_get_setup() which suddenly returns a NULL
pointer.
Task-number: QTBUG-27686
Change-Id: Ie3e4058f9d92bcbfc45934a8b36d9a7254e2b4bb
Signed-off-by: Uli Schlachter <psychon@znc.in>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Several platform plugins, like eglfs, kms, etc don't support multiple
windows as there's no system compositor, they're rendering directly to
a single back buffer. By adding a platform capability we'll be able to
provide better error reporting when an application tries to create
multiple QWindows on a single-window platform. Also, QML apps can use
this capability to figure out whether they should create a QWindow for
dialogs / popups / menus, or whether to just create items in the same
scene, that are shown on top of the rest of the content.
Change-Id: I15b8d21ee2bc4568e9d705dbf32f872c2c25742b
Reviewed-by: Andy Nichols <andy.nichols@digia.com>
|
|
|
|
|
|
|
|
|
| |
Modal dialogs that do not have a QWidget parent may be hidden
by their parent unless they actually have a transient parent.
Task-number: QTBUG-27786
Change-Id: I7847df3517e5ba6e8d77a2a18c905e908a3cd2f4
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
| |
it now defines the DESTDIR and creates an INSTALLS rule.
Change-Id: I15a462ccad9acbe3521c352fa98327825dc27c05
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
xcb_get_setup might return 0, but we already have it in QXcbConnection.
Task-number: QTBUG-27686
Change-Id: I58418aedd6bc121ae2b9605389beb3f6612d7fb7
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
Reviewed-by: Gatis Paeglis <gatis.paeglis@digia.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
This change enables receiving the native GLXContext object that is used
by a QOpenGLContext in case of GLX.
This clearly is non-public api that is only meant to be used as a
last resort for cases where it is really necessary to get hold of
a native context object.
Change-Id: I7f1f974f18063ed334b8034a0c0192c875c10cec
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Reproducible with:
QBackingStore store( &window );
store.beginPaint(QRect(0,0,500,500));
All other methods already have null pointer checks.
Change-Id: Ie278a263760900b58cf4a2ef286deb7f35d50cd0
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
| |
Change-Id: If311211076c05ed8d0fb62a9ff1129b93a813850
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
These were used to test QtWayland, but there are other options available
for that now. The DRI2 code hasn't been maintained and doesn't even
compile at the moment, so let's just remove it to de-clutter the source.
Change-Id: I7db0f4db82348497b9f4d6c2dcf2e13f3ab14a76
Reviewed-by: Uli Schlachter <psychon@znc.in>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
| |
Prefetch all the extensions, to avoid having to do blocking calls later.
Change-Id: I1527dbf03d76372ec88bc0d5d9f7af18a4cc2a26
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
|
|
|
|
|
| |
Change-Id: I9285d7524586ff404206c088019ece33335137d9
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
|
|
|
|
|
|
|
|
| |
- Move dnd docs and examples out of QtDoc module to gui library in QtBase
- Remove info related to Motif dnd since Qt5 doesn't implement it
Change-Id: Id7eb4eb422f4294a36dd92709ce3007903371f03
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The current implementation requests the platform window to set
the window state if it can, and return the actual window state
back.
The problem with this approach is that the platform window is created
as late as possible, so a call to QWindow::setWindowState would in
many (most?) cases never be forwarded to the platform window (instead,
the platform window is responsible to check the current window state
upon creation). As such, the window state might be left unsynched with
the platform window.
This patch suggests removing the return value from
QPlatformWindow::setWindowState. This will at least be consistent, so
that setting/getting state would produce the same result independent of
delayed window creation. If needed, we can later add new API to
QPlatformIntegration or QPlatformWindow for querying supported/actual
window state.
Change-Id: Ie43f56169656854a765ce88b47a808f8f3d51bb4
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Thomas McGuire <thomas.mcguire@kdab.com>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
|
|
|
|
|
| |
Change-Id: I415416e82d342b232607fe7087350a9dad615f7a
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The current implementation requests the platform window to set
as many of the flags it can, and return the same flags with the
unsupported flags removed.
The problem with this approach is that the platform window is created
as late as possible, so a call to QWindow::setWindowFlags would in
many (most?) cases never be forwarded to the platform window (instead,
the platform window is responsible to check the current window flags
upon creation). As such, the filtering would never be done.
Looking at the current set of plugins, most of them also seems to
ignore this protocol, returning the flags unfiltered.
This patch suggests removing the return value from
QPlatformWindow::setWindowFlags. This will at least be consistent, so
that setting/getting flags would produce the same result independent of
delayed window creation. If needed, we can later add new API to
QPlatformIntegration or QPlatformWindow for querying supported window
flags.
Change-Id: I9c759b5f9fab5ebed764a982f77fe19881118875
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch makes transaction mechanism less scattered around and
conforms to the xdnd specification:
Don't block and keep a history of previous data. This can be very difficult to implement,
but it is clearly the ideal behavior from the user's perspective because it allows him to
drop something and then continue working with the assurance that the target will get the
data regardless of how slow the network connections are.
When the source receives XdndFinished, it can remove the item from its history, thereby keeping
it from getting too large. The source must also be prepared to throw out extremely old data
in case a target malfunctions.
I assume that 10min for drag-and-drop operation can be considered 'extremely' old data.
Change-Id: I73dcd21aee3ad188d2260e49d80824da6ba040ab
Task-numer: QTBUG-14493
Reviewed-by: David Faure (fixes for KDE) <faure@kde.org>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Convention from icccm: Clients attempting to acquire a selection must set
the time value of the xcb_set_selection_owner request to the timestamp of
the event triggering the acquisition attempt, not to XCB_CURRENT_TIME. In
some cases it happened that timestamp was set to XCB_CURRENT_TIME.
A zero-length append to a property is a way to obtain a timestamp for this
purpose; the timestamp is in the corresponding XCB_PROPERTY_NOTIFY event.
We used to have this mechanism in 4.8, it was achieved by XWindowEvent.
AFAIK there isn't an equivalent for XWindowEvent in XCB. Therefore i had to
introduce a new mechanism in QXcbConnection - getTimestamp. This function
blocks until it receives the requested event.
Change-Id: Ide46a4fdd44cf026fdd17a79d3c4b17741d1b7d4
Task-number: QTBUG-26783
Reviewed-by: Uli Schlachter <psychon@znc.in>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
| |
Now it's propagated to QKeyEvent::nativeVirtualKey() as it should.
Task-number: QTBUG-27353
Change-Id: I40921b99662951decc83d2278e0cd337266dd530
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
|
|
|
|
|
|
|
| |
Fixes compile failure introduced by 394249616cbb4c0861a032d33f846f.
Change-Id: I5794d22ee87eb18c08ea1de227b943f2e675356c
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The function wglCreateContextAttribsARB will fail if we request a
context version higher than is supported. We therefore upper-bound the
requested version by the version of the static context. This results
in context creation succeeding and having the closest possible match
to the requested format.
The xcb qpa plugin is modified to operate similarly to the windows
plugin in that it now creates a "static" context which is used to
limit the versions of contexts requested by the user.
Change-Id: I277ad7cc82edfdf7b9d8502ad921c8175feb1a4a
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
xcb uses 32bits for xcb_keysym_t, but Xlib uses 64bit longs on 64bits
systems for KeySym and all other XIDs on the client side. Passing
an xcb_keysym_t* to XLookupString, expecting a KeySym*, would overwrite
the next 32bits in memory and possibly cause a crash.
This patch makes sure that a KeySym* is passed to XLookupString, and use
the signature declared in Xutil.h to make sure the types are right.
Encapsulate it in qxlibconvenience.cpp since including Xutil.h inside
qxcbkeyboard.cpp causes macro expansion problems.
Change-Id: I68451a24cb44a43dfa4382b5dce1ea7845f14e26
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Reviewed-by: Laszlo Papp <lpapp@kde.org>
|
|
|
|
|
|
|
| |
Task-number: QTBUG-27406
Change-Id: If14c108060ad6de70991658f0e0249eb544b87a0
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Occasionally drag-and-drop was returning empty mimedata due to
using wrong value for timestamp. Accoring to the xdnd specification
we must use timestamp from XdndPosition or XdndDroptime for retrieving
the data.
Task-number: QTBUG-27367
Change-Id: Iadb6b6989cfc4e8ab241e2fd0ded59355108f22f
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch fixes broken code in handleSelectionRequest(). There are 2 cases
when this function is called:
1) When XdndDrop has arrived (on mouse release)
2) When drop has not occurred yet (while mouse are still pressed down)
Second case was not working due to this bug.
Task-number: QTBUG-27405
Change-Id: I913cfd332128a28861e2fcc027b406eb821d2597
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Bug introduced when porting from Xlib to xcb. Xcb returns
actual number of bytes read while Xlib returns number
of 8-bit, 16-bit, or 32-bit items stored in the returned data.
Task-number: QTBUG-27403
Change-Id: Ia64da1953ec7be8d32cc591565cd1b41ce8d7dbb
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
| |
Fixes bug introduced in SHA e247e2810 (QTBUG-26709).
Change-Id: I9aa80239675b2d564d08a019ca267a9aa814074e
Reviewed-by: David Faure <faure@kde.org>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
For consistency, this behavior has been kept across Qt versions... Just
get rid of it.
Also fixes native child widgets not being notified of the change of
window handle (winId) when being reparented.
Updated auto-test.
Change-Id: I3616dc0f1c32a519d78a4846297d6d4a6e926fbf
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
| |
This seems to fix the ability to run a Qt app across a forwarded X
connection (e.g. ssh -X).
Change-Id: I2b7a26985cf11107e69b303337a5fbb369e38c9e
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Reviewed-by: Uli Schlachter <psychon@znc.in>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This commit fixes the xcb qpa plugin such that it now correctly reports
the version and profile of the created OpenGL context in the
QOpenGLSurfaceFormat. To do this we have to create a temporary X window
so that we can make our new context current.
We also handle the buggy nVidia drivers which incorrectly report 0 for
the GL_CONTEXT_PROFILE_MASK query.
The reduced format is also copied back from qglx_findVisualInfo.
Change-Id: I6f34fe1c6130aebbb6b40c36df4acc216069d2b1
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
|
| |
Marc Mutz already reported problems with ssh -X fowarding. Now all
such errors are treated the same: if we can't get screen output
attributes, just assume there is only one.
Change-Id: I96802fc90072c623de3370ed2898893daf58198a
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A new QScreen is created when an output is activated (monitor or
projector is added, for example), and destroyed when the output is
turned off. Ensures that screens and siblings are always in
the right order: primary comes first.
Logical DPI is derived from virtual geom / virtual size,
which will be different than output geom / physical size
if X was started with --dpi override. This is a good thing:
when X gets wrong EDID info for physical size and you need to
override it to get reasonable font sizes, Qt will heed the
logical DPI for font sizing.
Change-Id: I5e3de34013c1b6b21067243de56f3f1eb72787fa
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
| |
Change copyrights and license headers from Nokia to Digia
Change-Id: If1cc974286d29fd01ec6c19dd4719a67f4c3f00e
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Sergio Ahumada <sergio.ahumada@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
QScreen has notifiers for its properties, but they were not being
emitted when one changes the resolution or arrangement of individual
outputs, e.g. via xrandr. Also there should be one QScreen per
"output", e.g. laptop LCD + external monitor means 2 QScreens
which will be siblings, rather than just 1 QScreen to represent
the whole desktop.
Change-Id: Ia61bbc5e6a3506f813ab11f87c03d14cf7f4ce85
Reviewed-by: Uli Schlachter <psychon@znc.in>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
|
|
|
|
|
|
|
|
| |
Thanks to Janne Kulmala for noticing this and informing about the fix.
Task-number: QTBUG-27123
Change-Id: Idd3cfd74fb7be277b6d805446aea5784b8fdf2f6
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
It is necessary to use the n9 device file for now in scratchbox or/and on the
community open build service because the maemo platform mkspecs file assumes
that a cross-toolchain is used all the time. If no platform file is used, then
for instance certain plugins may not be built in general. There is currently an
ongoing issue with the meego plugin for context management in the Harmattan
components project. That is currently not built due to this issue, so no
orientation works in those applications.
The nice solution would be to make the maemo platform file work with cross and
native toolchains as well, but that requires a decent amount of investigation
and work. Thereby, the scope is extended this way for now.
Change-Id: I172c7d152bdbb2db279526d9fd1ca5648d0cd0a9
Reviewed-by: Simon Hausmann <simon.hausmann@nokia.com>
|
|
|
|
|
|
|
| |
Do not include a header more than once
Change-Id: Ia2e5d66e72988ad833cf5177a3f8aa988bf510e9
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
|
| |
The TRUE and FALSE macros are obsolete and should be replaced with
true and false (all lower case) respectively.
Change-Id: Iee352e8173500683e6319be0abbf5bacf29016e0
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
|
| |
The variable ptrXcursorLibraryLoadCursor is only used in code
wrapped in XCB_USE_XLIB.
Change-Id: If0b4997ac6e80a3709263d9793d1ca0dfd5907f3
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
|
|
|
|
|
|
| |
Task-number: QTBUG-26954
Change-Id: I6981338d4bbc9cf1440c9d67e9d034e0553dfeae
Reviewed-by: Jason Barron <jason.barron@nokia.com>
|
|
|
|
|
| |
Change-Id: I5a08d6067272575aa56074aaebe308c3d49299bb
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Handle 0 WId parameter as meaning "desktop window"/whole screen.
Also, re-add the default values for the grab area, both for
convenience and compatibility with QPixmap::grabWindow() in Qt4.
Update the screenshot example so it doesn't comlain about usage of
deprecated QPixmap::grabWindow().
Change-Id: I2ad229113ddb8ded0388f2ebc0e8c703c6657f1f
Reviewed-by: Gunnar Sletta <gunnar.sletta@nokia.com>
|
|
|
|
|
|
|
|
| |
This is a plugin that bridges the QAccessible world
to AT-SPI 2 on Linux.
Change-Id: I7af22621ee6a3cefc723b137b7f227a611cf6641
Reviewed-by: Jan-Arve Sæther <jan-arve.saether@nokia.com>
|