If we wanted to make the list exhaustive for buildCpuArchitecture, we'd
add alpha, avr32, bfin, s390(x), and sh. I don't want to make it
exhaustive because that's cluttering the documentation and we'd have to
keep in sync with archdetect.cpp. And we can't make
currentCpuArchitecture exhaustive, since it depends on the output from
uname(2).
You can argue that neither IA-64, POWER, nor SPARC architectures are
typical these days...
Change-Id: I0a5310770947263e1bafd9443ea59420813c51a8
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Jake Petroules <jake.petroules@petroules.com>
Reviewed-by: Richard J. Moore <rich@kde.org>
It's easier to read AVXState and AVX512State than 6 and 0xe6.
Also add a note that where we should have checked whether the SSE state
is being saved by the OS. However, we won't do it because it's just a
waste of CPU cycles: any OS Qt 5 runs on will enable the proper
state-saving.
Change-Id: Id87b59fe1388a6cab983c9412341e36a86dd15c5
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
It is documented to be, and the LanguageChange event is caught and processed.
However, retranslateStrings() uses QProgressDialog::setCancelButtonText(),
which unconditionally sets useDefaultCancelText=true, blocking any further
changes to the button text by subsequent LanguageChange events.
The fix is to use extracted QProgressDialogPrivate::setCancelButtonText()
which - quite intentionally - doesn't set useDefaultCancelText.
Task-number: QTBUG-40504
Change-Id: I6e701deda10c454cb088c0b0778ac2d6adff574a
Reviewed-by: David Faure <david.faure@kdab.com>
This is in preparation of a fix for the broken online-retranslatability of
the cancel button.
Change-Id: Ie62540766e50e1f1ec07d251cc56a2ee0745d434
Reviewed-by: David Faure <david.faure@kdab.com>
The associated test has unearthed that setBar() fails to make the new bar a child
of the progress dialog. This will be fixed in a separate commit.
Task-number: QTBUG-40502
Change-Id: I2d09ebb07ae6395449a4efe38a638df831eebdd7
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: David Faure <david.faure@kdab.com>
This is needed to make it possible to use the same interface for
retrieving the EGLDisplay on Windows and EGLFS, thus get rid of
cluttering ifdefs.
Change-Id: I37b848b1017eacbf8a29627cd157b74e22e5f40c
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
Reviewed-by: Laszlo Agocs <laszlo.agocs@digia.com>
The native interface implementation in QEGLPlatformIntegration
lower-cases the resource key strings, where as in the Windows
implementation we currently only check for camel-case resource
names to retriece the same resources.
Make it possible to use lower-case strings on Windows as well
by using the same key look-up mechanism as used in the eglfs
implementation.
Change-Id: Id2a594310df610cadbe420409c090f0abb316474
Reviewed-by: Laszlo Agocs <laszlo.agocs@digia.com>
On Linux for example Orca gets confused when showing a dialog that is a
child of another widget since it would show up twice in the hierarchy.
Task-number: QTBUG-39444
Change-Id: I84773ecc3d6774a652dbeb29ad201779f5b3191c
Reviewed-by: Jan Arve Sæther <jan-arve.saether@digia.com>
Seems to have been a cut'n'paste from 'minimum'.
Change-Id: Ifc3a4441809a9fc75ecac621cff59950235f6bc7
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: David Faure <david.faure@kdab.com>
This patch adds support for binding RGB30 images as textures, and as
internal format of framebuffer objects. Together with the
QOpenGLPaintDevice
this provides support for rendering to and from RGB30 in full precision.
[ChangeLog][QtGui][QOpenGLFramebufferObject] Support 10-bit per color
channels formats as the internal framebuffer format, making it possible
to render in that precision.
Change-Id: I06de2d12dfe1c1adc466d574fdffbc77f88f4f16
Reviewed-by: Laszlo Agocs <laszlo.agocs@digia.com>
The code path in QOpenGLVertexArrayObjectPrivate::create() that triggers
the creation of a QVertexArrayObjectHelper is guarded by runtime checks
for ES3 or the GL_OES_vertex_array_object extension, but the actual
function lookup was ifdef'ed, which broke on iOS where the SDK may
support ES3, but an older (supported) runtime target might not.
Change-Id: Id578667c1f5aebf53e197f3a79eb2f9273fea487
Reviewed-by: Laszlo Agocs <laszlo.agocs@digia.com>
It is useful in other places, for instance in QtQuick, to avoid
duplicating the same resolver logic.
Change-Id: I9748a420a0abeb07cc84f948965b1e0321a95ca2
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Reviewed-by: Laszlo Agocs <laszlo.agocs@digia.com>
This is just an optimization of the code: instead of opening and parsing
/etc/os-release every time QSysInfo::kernelType() is called on Linux, do
that only in QSysInfo::productType() and productVersion().
Change-Id: I201504934ecf7a51854cb49c790bd9d30eba644b
Reviewed-by: Kai Koehne <kai.koehne@digia.com>
Reviewed-by: Jake Petroules <jake.petroules@petroules.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
[ChangeLog][QtCore][QByteArray] Fixed a bug that would cause QByteArray
to stop converting toUpper or toLower at the first embedded null
character.
Change-Id: Ia369037206617813d86a8f1489589243c82aa51b
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
This patch splits the OS product from the kernel name and versions.
Previously, it was a little confusing.
Thus, osType is split in two: kernelType() always returns the kernel
type, which includes "linux" for Android, "darwin" for iOS and OS X,
"qnx" for BlackBerry, and "windows" for all the Windows OSes. The Linux
distribution name and the visible product names are found in
productType(), with the respective versions in productVersion().
For an update system, applications probably want to send all four
tokens:
- for Android, BlackBerry, iOS, OS X and Windows, the relevant
information is productType and productVersion
- for Linux distributions, in addition to productType and
productVersion, the kernelType ("linux") is useful as a fallback and
it's important for Debian/kFreeBSD and Debian/GNU
- for other Unix systems, the relevant information is kernelType and
kernelVersion
Change-Id: I66c36d5a4ddfad47babfc6744d752273705e7076
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
buildCpuArchitecture() returns the CPU Qt was built for, while this
function returns the CPU that Qt is running on -- if the OS was kind
enough to tell us.
Change-Id: Ib27937e3ff028cb500d263c4921ef00d3a567715
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
We build "Supporting Files" out of QMAKE_INTERNAL_INCLUDED_FILES, which
is really not supposed to be exposed to the user like that, but since
the variable will hold user-included pri files eg., the Xcode generator
piggy-backs on this variable to list the files.
To make the project view in Xcode a bit cleaner we explicitly exclude
any file living inside the Qt directory, meaning we won't show all the
pri and prf files from Qt's mkspecs directory anymore.
Change-Id: I828700aceac5fdf3ea2b27d9ba3885543c2ad137
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
Now that we rely on simd.prf for all SIMD sources (including NEON and SSE2),
we need to ensure that CONFIG has the right SIMD values to match simulator.
This worked before due to us checking QT_CPU_FEATURES.$$QT_ARCH and adding
directly to SOURCES.
Change-Id: I4ea7f559e83860eabff1948ad5d140bbb65454df
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Use the built-in hex, showbase manipulator to format the number in
hex. Also, apply nospace() only once.
Change-Id: Id4b3d5f082ad13f52c8711408d7ec609bec3a621
Reviewed-by: Alex Blasche <alexander.blasche@digia.com>
Commit d61e774307 got rid of the special handling of the 'default'
category in the qloggingregistry, so we don't need the private header
file anymore, nor does it gain us anything to 'normalize' all default
categories to one string.
Change-Id: I027b4b82ff915428287432f39f76b4e719cf7874
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Alex Blasche <alexander.blasche@digia.com>
Export the former qMessageFormatString() as qFormatLogMessage(). This
allows custom message handlers to format their messages just like the
default message handler, taking qSetMessagePattern() /
QT_MESSAGE_PATTERN into account.
The method should arguably not add the '\n' at the end, which a follow
up commit will fix.
Change-Id: Ib2a9cfda91473df079daf03bf3197e6ac63e013e
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Alex Blasche <alexander.blasche@digia.com>
when doing a module-by-module build, we need to also use includes and
libraries from the install tree, as it contains the current module's
dependencies. but a pre-existing installation of the current module must
not be found first, as it would cause trouble latest when it was somehow
incompatible.
but purely topological sorting of the dependencies could cause the
locations to be mixed up. therefore we give modules which are part of
the current build a priority boost.
Change-Id: I8fdbb46f0a2a630781c8a2177468039c1122151a
Reviewed-by: Giulio Camuffo <giulio.camuffo@jollamobile.com>
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
all else being equal, items with a higher numerical priority will appear
first in the result.
Change-Id: I4ee37ff404a53c4152a1e4fc2fc3c23ef525234d
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
the c'tor always determines the group itself anyway.
Change-Id: Ia8f1e747aaefdab164beae34851aa99cec9b790a
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
At the very least, include the files named in the sources, like
HEADERS. It was quite surprising to send a tarball that included the
.pro file and the .cpp sources, but none of the headers.
On the other hand, the .qmake.cache file need need not be sent either,
despite being include()d in qmake's processing.
Change-Id: I8f48ca3e8040f954f321f4643b01c0f36aafe2d7
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
When compiling the innerFunction() of the Q_GLOBAL_STATIC expansion, ICC
emits global symbols for the internal "holder" local static variable and
its guard. If there are two global statics of the same name in two
different .cpp files, the linker will incorrectly merge the two "holder"
variables.
This was noted between the "customTypes" global statics of qmetatype.cpp
and qdbusmetatype.cpp in a static build.
The C++ standard requires that local static variables declared in inline
functions must be the same, regardless of whether the body of the
function got inlined or not. The IA-64 C++ ABI does that by requiring
local static symbols for inline functions to be global and mergeable
("link once"). However, two functions in anonymous namespaces in
different files are not considered to be the same function, so their
local statics should not be merged. This is where ICC failed: the local
statics are global and mergeable, even though the function is in an
anonymous namespace. ICC correctly emits the function itself as a local
symbol.
Alternative solutions were:
1) add "static", but you can't use a static symbol in a template
parameter in C++98 mode
2) remove the "inline" keyword, but then GCC 4.8 will not inline
Intel issue ID: 6000058488
Task-number: QTBUG-40053
Change-Id: I307622222499682dde711b2771c8cf7557400799
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@digia.com>
It allows to create a QJsonObject instance in C++ by using
initializer list of pairs QString QJsonValue, for example:
QJsonObject o = {{"property1", 1}, {"property2", 2}};
[ChangeLog][QtCore][QtJson] QJsonObject now supports
C++11 initializer lists.
Task-number: QTBUG-26606
Change-Id: I67af881e175f427e563e685336c48a5f8466b476
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Halves the amount of paint events triggered by transient scrollbars.
Task-number: QTBUG-30316
Change-Id: Ifdf968d5c45013332758a6b751ce11d1ef2a2ca8
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Introduce function QMdiSubWindowPrivate::storeFocusWidget()
to store focus widget and call this when de-activating a
sub window. Change restoreFocus() to return a bool and call
it from QMdiSubWindowPrivate::setActive().
Task-number: QTBUG-38378
Change-Id: I18dbe66ce85213ca5b4907b5a09126544415351a
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Other platforms do not do this either. What is more, some of the
extension handling code will try to resolve functions that are not
necessiarly present, and this is not an error.
Change-Id: I39dad8f8d89fc45de3ea83f04727a8e38b6a3387
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
kCTFontWeightTrait returns a normalized value between -1.0 (lightest)
and 1.0 (heaviest), 0.0 being the regular font weight. The threshold
values used in this change have been estimated from the weight values
of fonts from the Helvetica Neue and Myriad Pro font families.
Change-Id: I49de8e8bd5894107de4842aeda7ace2e83f95be3
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
With MSVC it takes minutes to compile pbuilder_pbx.cpp.
So let's remove this generator that's never used.
Change-Id: I13038d551283d96dfb0baf0b8a8a68c6538193c8
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
For multi-pass RCC qmake generates broken VS project files, because
the RCC extra compiler directly calls the C++ compiler on a generated
source file. Adding this call to a VS project file will bypass any
project settings. Also, the VS project generator is not prepared to
add extra compilers that generate object files.
Task-number: QTBUG-39685
Change-Id: I1bcaad8936be8371d596f29ed8952888ba95f7b2
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
The fast scale methods qt_scale_image_16bit and qt_scale_image_32bit
were incorrectly assigning an unsigned value to a signed variable, which
broke their support of images with dimensions from 32384 to 65535.
Images with dimensions higher than that are already using another code-
path.
Task-number: QTBUG-40297
Change-Id: I8c971889e069381224cea2befbb5c66cd93ea5c2
Reviewed-by: Gunnar Sletta <gunnar.sletta@jollamobile.com>
Replace old virtual declarations by Q_DECL_OVERRIDE where
applicable.
Errors discovered: QWindowsScreen::primaryOrientation() did not
overwrite any method, changed to orientation().
Task-number: QTBUG-38858
Task-number: QTBUG-38993
Change-Id: Ie85d5235f3cd35c7a24cac3c1c7693c6508b1009
Reviewed-by: Oliver Wolff <oliver.wolff@digia.com>
Some Linux distributions disable EC by default which causes compile
errors on those platforms.
Task-number: QTBUG-40394
Change-Id: If5816d473bd1d64b1d4977860db662704a83310f
Reviewed-by: Richard J. Moore <rich@kde.org>
