A stretch equal to 0 is since 5.8 defined as "accept the stretch of the font",
and this needs to be accounted for in the font engines.
Task-number: QTBUG-57491
Change-Id: Idabbe44677c4b92cbd8ad8278b054de53e9cc7f9
Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
Reviewed-by: Alessandro Portale <alessandro.portale@qt.io>
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
Check if it's an X11 window before calling XSetTransientForHint().
No transient parent will be set for GTK+ dialogs on Wayland. That
has to be implemented separately.
Task-number: QTBUG-55583
Change-Id: Iabc2a72681c8157bb2f2fe500892853aa397106b
Reviewed-by: Dmitry Shachnev <mitya57@gmail.com>
Reviewed-by: Shawn Rutledge <shawn.rutledge@qt.io>
The C++ standard says in [support.initlist.access]/1:
constexpr const E* begin() const noexcept;
Returns: A pointer to the beginning of the array. If size() == 0 the
values of begin() and end() are unspecified but they shall be
identical.
So we can't assume it's non-null. I didn't want to remove the Q_ASSERT,
so passing a non-null pointer to append() remains required. This patch
simply won't call append() if the initializer list is empty.
This was already tested, but the failure is with a compiler that is not
part of the Qt CI.
Task-number: QTBUG-57277
Change-Id: Iaeecaffe26af4535b416fffd1489806872b412ee
Reviewed-by: Olivier Goffart (Woboq GmbH) <ogoffart@woboq.com>
Previously, we would activate the application during
QCocoaIntegration construction, which means at QApplication
creation time. This now seems to interfere with application
startup on macOS Sierra, where the application window
ends up in an unfocused state.
Move application activation to applicationDidFinishLaunching,
at which point the Cocoa runtime should be completely
initialized. Do this for 10.12+ only to avoid regressions/
test failures on previous versions.
Change-Id: Ic5f150d53f06a302b53a3ba86a4a9b18bb2a1783
Task-number: QTBUG-57044
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Commit 2bc7a40048 taught the CoreText font database to populate the
families lazily, and in the process added a guard to ensure that we
didn't populate internal fonts (prefixed with a '.'), as these fonts
would then show up in font selection dialogs.
Commit 909d3f5c7 then added support for private fonts, by making it
possible to filter out any private fonts from font selection daialogs.
But the guard was not removed, so we were still not populating these
fonts. This guard has been removed, and the filtering function has
been updated to include the conditions of the guard.
Next, commit e5e93345c5 used [UIFont fontNamesForFamilyName:] to verify
that each family that we registered with the font database would also
have matching fonts when finally populated. This is not the right approach,
as [UIFont fontNamesForFamilyName:] does not handle internal fonts.
Instead we trust what CTFontDescriptorCreateMatchingFontDescriptors()
gives us, but make sure to register the resulting font descriptors
with the original/originating font family, instead of the one we pull
out of the font descriptor.
Finally, as of iOS 10, we can use CTFontManagerCopyAvailableFontFamilyNames
instead of [UIFont familyNames], which gives us all of the internal font
families like on macOS, instead of just the user-visible families. For
earlier iOS versions we manually add '.PhoneFallback', as we know it
will be available even if not listed in [UIFont familyNames].
The end result is that we register and populate families like '.PhoneFallback',
which is critical to supporting more esoteric writing systems.
The check in tst_QFont that styles for a given family is not empty has
been removed, as we can't guarantee that on all platforms, which is
also documented for QFontDatabase::styles().
Task-number: QTBUG-45746
Task-number: QTBUG-50624
Change-Id: I04674dcb2bb36b4cdf5646d540c35727ff3daaad
Reviewed-by: Jake Petroules <jake.petroules@qt.io>
Window managers typically grab the pointer after receiving
the _NET_WM_MOVERESIZE event. But they fail to do it for
touch sequences which have a receiver. So we should reject
the touch sequence before sending _NET_WM_MOVERESIZE event.
QSizeGrip calls startSystemResize() on MouseButtonPress event
which is synthesized by Qt on TouchBegin. We can find the id
of the touch point by comparing coordinates of the synthesized
MouseButtonPress event with coordinates of all TouchBegin events.
Then we use this id to reject the touch sequence (it's possible
only after receiving XI_TouchUpdate).
Change-Id: I26519840cd221e28b0be7854e4617c9aba4b0817
Task-number: QTBUG-51385
Reviewed-by: Shawn Rutledge <shawn.rutledge@qt.io>
The prolog and epilog did not force RGB32 to be converted to RGB64 with
alpha fully defined like the middle optimized part.
Change-Id: If7c4829f2239f9a3c524f78b9ce269e2b0b5b150
Reviewed-by: Eirik Aavitsland <eirik.aavitsland@qt.io>
this also removes the need for passing pre-processed options via
configure.cfg, so get rid of that.
a somewhat unfortunate side effect is that the android-style-assets
feature had to move back to the top level, as the licensing options
depend on it.
Started-by: Lars Knoll <lars.knoll@qt.io>
Change-Id: Id4d1e0ba18b3e3104400293b8f0c7f2f65e68dea
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
The function was incorrectly handling green and blue color channels
causing them to be dropped. This affects drawing non 32-bit images onto
10-bit per color channels formats such as RGB30.
Change-Id: I9211e253b1a9da0dada5c418d592a8f531265989
Reviewed-by: Eirik Aavitsland <eirik.aavitsland@qt.io>
Beside its usage in widgets, mouse grabs are required for QML menus to
work.
Task-number: QTBUG-57079
Change-Id: I306cb68624186da69725470e147bc7b979dac8e4
Reviewed-by: Oliver Wolff <oliver.wolff@qt.io>
Followup to 0484473: this is all new stuff for 5.8 and we don't need
to release it with pre-deprecated functions.
Task-number: QTBUG-54616
Change-Id: If17a4bec6fc36ca78d87517992374f101ae13b4f
Reviewed-by: Jan Arve Sæther <jan-arve.saether@qt.io>
This re-fixes commit d72ac3f35f, which
simply removed the #define but did so at the wrong place. Instead of
forcing the macro to be removed, let's simply not have it defined in the
first place.
Change-Id: Ie6dbad9bbbd9488887e8fffd148dd67d9a31b32e
Reviewed-by: Jake Petroules <jake.petroules@qt.io>
-[NSMenu itemWithTag:] clearly states that it'll return the first
item with that tag. Furthermore, when and item has been synced more
than once, it could be that more than one such item exists in the
same menu (e.g. lately changing the role of Edit->Copy).
Change-Id: I95a4f0a151659ae273ba03a3cab4a720b781fc3a
Task-number: QTBUG-57404
Reviewed-by: Jake Petroules <jake.petroules@qt.io>
This should not happen, but it's clearly not the user's fault.
So we should try to carry on as gracefully as possible instead
of letting Cocoa abort the application.
The patch also factors the repeated calls to QCocoaMenuItem::
nsItem() in QCocoaMenu::insertNative() and improves a warning
from QCocoaMenuIten::sync().
Change-Id: Id00135c219aaf40fb565b19a65cab68f6d9863b2
Task-number: QTBUG-57404
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
Adds platform and CPU features to the reported build type,
matching the format of the old configure.
Started-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
Change-Id: I6d93ec7416b38684da51af5238a5cf537810b21d
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QHostAddress allowed assignment from a QString, but the respective
constructor is explicit, and rightfully so. So it does not make
sense that the assignment operator is provided, because of the
asymmetry caused between
QHostAddress addr = funcReturningQString(); // ERROR
addr = funcReturningQString(); // OK (until now)
By the same token, since SpecialAddress is implicitly convertible
to QHostAddress, provide the missing assignment operator from that
enum.
Add tests, rewriting the _data() function to use the enum instead
of an int to pass SpecialAddress values, and to test !=, too.
Added setAddress(SpecialAddress), since a) it was missing and
b) to share code between the ctor and the assignment operator.
Change-Id: Ief64c493be13ada8c6968801d9ed083b267fa902
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This warning does not make sense. it seems to trigger when in code like
the following in template functions:
auto x = 1, y = 2;
3373: nonstandard use of "auto" to both deduce the type from an
initializer and to announce a trailing return type
Other reports on the Internet indicate that no one understands what
triggers this warning and have just worked around it. Additionally, the
same warning exists on other compilers with the same text, so it's
likely come from the EDG front-end. This has been reported to Intel.
Change-Id: I73fa1e59a4844c43a109fffd148d45065ab69eff
Intel-Issue-ID: 6000164202
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Several people agreed that the name was confusing and that this one
is better.
Task-number: QTBUG-54616
Change-Id: I31cf057f4bc818332b0551a27d1711599440207c
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Reviewed-by: Sune Vuorela <sune@vuorela.dk>
eliminating everying TARGET-related was a nice try, but in the real
world (e.g., qttranslations), extra compilers are activated by
PRE_TARGETDEPS, which of course doesn't work when TARGET is entirely
gone.
so instead, let it act as a phony target. this is consistent with the
unix generator.
supersedes 0810d48bc in amending af2847260.
Task-number: QTBUG-57423
Change-Id: I3d2ecc4ff42b37ffe5f71f5c20d17c06b31f4da2
Reviewed-by: Jake Petroules <jake.petroules@qt.io>
Reviewed-by: Oliver Wolff <oliver.wolff@qt.io>
Changes it to follow standard SIMD patterns so it can use
ALIGNMENT_PROLOGUE_16BYTES and SIMD_EPILOGUE helpers.
Should also improve performance by using aligned memory access.
Change-Id: I14a48b82e3f3de83bd7572aa82bed07f28ad944c
Reviewed-by: Erik Verbruggen <erik.verbruggen@qt.io>
Similar to the way datagrams are handled for udp sockets the worker now
takes care of tcp data. Thus we avoid race conditions which stopped data
processing. It could happen that data was read from the socket into the
buffer and before readyRead was emitted the buffer was completely read.
In this case readNotification is set to false and no new data is processed
afterwards.
Additionally the buffer was replaced by a vector of QByteArray. The buffer
kept growing and was never cleared (and there is no obvious way for
clearing the buffer), so that an overflow happened eventually.
pendingReadOperations (and its mutex) could be removed as well. There is
only one situation where they could clash and that's the initial read.
Having two members is preferred over having a list of operations and a
mutex.
Task-number: QTBUG-56438
Change-Id: Idbad58e47785996023748c310530892163f24594
Reviewed-by: Maurice Kalinowski <maurice.kalinowski@qt.io>
We have a 'channel' object connected to a socket with Qt::DirectConnection.
QHttpNetworkConnectionPrivate in its dtor (note, it's a private object destroyed
after its 'q' - QHttpNetworkConnection - was destroyed) calls socket->close()
and this can end up in socket setting an error and emitting (for example, in
QSslSocket::transmit). The slot (QHttpNetworkConnectionChannel::_q_error) will
access the now-dead/non-existing connection then. So disconnect the channel
from the socket early, before closing the socket.
Task-number: QTBUG-54167
Change-Id: I3ed4ba4b00650c3a39e5c1f33aa786e47bfbbc57
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Change 3370ab9119 introduced
warnings from MSVC:
tst_qsqlquery.cpp(4005): warning C4805: '==': unsafe mix of type 'const bool' and type 'int' in operation
tst_qsqlquery.cpp(4059): note: see reference to function template instantiation 'void runIntegralTypesMysqlTest<bool>(QSqlDatabase &,const QString &,const QString &,const bool,const T,const T)' being compiled
with [ T=bool ]
tst_qsqlquery.cpp(4006): warning C4805: '==': unsafe mix of type 'const bool' and type 'int' in operation
tst_qsqlquery.cpp(4006): warning C4804: '/': unsafe use of type 'bool' in operation
tst_qsqlquery.cpp(4026): warning C4804: '+=': unsafe use of type 'bool' in operation
Extract an overload taking a QVector of values and use that for the
bool case instead of looping over min/max to generate a sequence of values
for bool.
Change-Id: I72583774e788b8df899f22ed1a64278217e664f6
Reviewed-by: Milian Wolff <milian.wolff@kdab.com>
It doesn't like the access to the template instantiation in the ternary
operator.
error: operand types are incompatible ("FetchPixelFunc" and "<unnamed>")
Intel-Issue-ID: 6000164201
Change-Id: I73fa1e59a4844c43a109fffd148ca452796eebb1
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
That's what happens when you don't test and just rely on an the warning
listing. ICC has two warning numbers for deprecated warnings: one that
matches Q_DECL_DEPRECATED and one for Q_DECL_DEPRECATED_X.
Change-Id: I73fa1e59a4844c43a109fffd148ca7a05eda8f13
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Number of API changes in Qt 5.8 use these macros and QDoc needs to
ignore them the correctly match the documentation to the function
signatures.
Task-number: QTBUG-57424
Change-Id: I0c3a0eb4deb2d9b348f24800591bc6f6b47cf458
Reviewed-by: Martin Smith <martin.smith@qt.io>
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
The alpha channel of an RGB32 image was not properly ignored when doing
blending with partial opacity.
Now the alpha value is properly ignored, which is both more correct
and faster. This also makes SSE2 and AVX2 implementations match NEON
which was already doing the right thing (though had dead code for
doing it wrong).
Change-Id: I4613b8d70ed8c2e36ced10baaa7a4a55bd36a940
Reviewed-by: Eirik Aavitsland <eirik.aavitsland@qt.io>
Add a QT_QPA_EGLFS_KMS_PLANE_INDEX environment variable that applies
both to the GBM and EGLDevice backends. When set to a value between 0 and
the number of planes on the connector - 1, the chosen overlay plane will
be used for output, meaning there will be a drmModeSetPlane to configure,
and, in case of EGLDevice, the plane's corresponding EGL layer will get
chosen instead of the CRTC's.
Task-number: QTBUG-57386
Change-Id: I12c89472ea5730987052f39211fadc597d1302ef
Reviewed-by: Pasi Petäjäjärvi <pasi.petajajarvi@qt.io>
Add a QT_QPA_EGLFS_KMS_CONNECTOR_INDEX environment variable that applies
both to the GBM and EGLDevice backends. Instead of specifying all
uninteresting outputs as "off" in the config file in QT_QPA_EGLFS_KMS_CONFIG,
this variable provides a shortcut to force one single connector and ignore
all others in embedded systems with a fixed connector configuration. The index
must be between 0 and DRM connector count - 1.
Task-number: QTBUG-57386
Change-Id: I3f9562f48bf6b2ffaf9a0cc232e09a7e0c15645b
Reviewed-by: Pasi Petäjäjärvi <pasi.petajajarvi@qt.io>
Make it possible to identify from the logs that QT_QPA_EGLFS_LAYER_INDEX
was set, in order to help troubleshooting.
Change-Id: Ic22825e5df9f0eeb31f817f398b9f6c000c3a00f
Reviewed-by: Pasi Petäjäjärvi <pasi.petajajarvi@qt.io>
Parent patch introducing usage of this attribute did not add it to
the EGLStream support header. This is likely not fatal, but follow
the practice of defining the constants ourselves, in case they are
not present.
Change-Id: Ib16f9809f9c6a212570c49472bb840183232e68a
Reviewed-by: Pasi Petäjäjärvi <pasi.petajajarvi@qt.io>
If the StatusNotifierWatcher disappears and then appears again, we need
to register our tray icon again with it.
To do this, split the “register with watcher” part into a separate
method, and call it when m_dbusWatcher emits its serviceRegistered()
signal.
Change-Id: Id5fc8ac81b5038a61b678514dabd3eb9c8f1c106
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Shawn Rutledge <shawn.rutledge@qt.io>
- Declare as Q_MOVABLE_TYPE
- Prevent QList<QPointerUniqueId> from being instantiated
(use QVector instead)
- Add equality relational operators
- Add qHash() overload
- Replace non-default ctor with named ctor.
- Add Q_DECL_NOTHROW.
- Add Q_DECL_CONSTEXPR.
- Rename numeric() -> numericId().
- Update docs.
The extension vector for this class calls for additional
properties to be added later, but these are not user-
settable. It thus suffices to rely on the only data
member, a qint64, which can be reinterpreted to an index
into an array or hash with actual objects. This allows
to make the class a Trivial Type (ie. no overhead over
an int) while still supporting later extension. Cf.
QSslEllipticCurve as another example of such a class.
The extension has to maintain the following invariants,
encoded into user code by way of being used in inline
functions:
- m_numericId == -1 <=> !isValid()
This is trivial to support. An extension could not and
still cannot reinterpret the qint64 member as a d-pointer,
but a d-pointer is only necessary for user-settable
properties where updating a central private data structure
would cause too much contention.
Add a test.
Since this type is used in other modules, keep the existing
functions, but mark them as deprecated with the expectation
that these compat functions be removed before 5.8.0 final.
Task-number: QTBUG-54616
Change-Id: Ia3ede0ecaeeef4cd3ffa94a72b1050bd409713a5
Reviewed-by: Shawn Rutledge <shawn.rutledge@qt.io>