When a peer closes the connection, the device remains opened for reading
purposes. However, we should disable writing on disconnected socket.
Otherwise, if the user issues a write() call, a new pipe writer object
will be created and the write call occurs with invalid handle value.
Pick-to: 5.15
Change-Id: Id136798c7663df1fce7ed0aa4e3c6f5c65218a11
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
In order to prevent too much voodoo in backends like D3D11, the input
layout is expected to specify the slice index for vecX that are part of
an unrolled matrix.
Also deoptimize the instancing manual test to exercise a matrix too
instead of just vectors.
Change-Id: If2dcbcbc483645ce2420b2f87dda765b95da6e80
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
Use a simpler constexpr to generate type name on gcc This
works around an ICE on gcc in release mode when compiling
with PCH enabled. As the type we're getting from Q_FUNC_INFO is
already in a somewhat normalized form, this requires significanlty
less processing and esp. not a recursive constexpr method which
I suspect triggers the ICE.
Fix integer type conversions to also properly normalize long long
values (to q(u)longlong. Make sure the mapping also works on
MSVC, where long long types get mapped to __int64. Also, normalize
unsigned short and unsigned char to ushort and uchar, respectively, to
follow the convention set by uint and ulong.
Add some test cases to verify the mappings.
Change-Id: I3dec5764450bf22ab6f066597803c3f46c2cd5ac
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
It's not possible to connect to microsoft.com with Schannel TLS 1.3 for
some reason (also tested with Internet Explorer), but other sites work
fine. Must be something they have to iron out for later.
In my experience this needs a preview release of Windows. One of my
machines is opted into the dev channel of Windows where they enabled TLS
1.3 by default, and it works well in my tests except for the part above.
On my other machine, after enabling TLS 1.3 through the registry, I fail
to complete the handshake with any site. So around March/April next year
is when this code would activate for most people.
MinGW apparently defines NTDDI_VERSION as the one for Windows Server
2003, so it currently doesn't build the new TLS 1.3 code. In Qt (as a
project) we could consider setting this higher, but that's out of scope
for this patch!
Fixes: QTBUG-81294
Change-Id: If329959c3a30ecbfbb8c0d335cc39ccb6d012890
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
tst_QApplication::sendEventsOnProcessEvents() and
tst_QItemDelegate::editorKeyPress()
There is some issue with the glib event dispatcher.
Task-number: QTBUG-87137
Pick-to: 5.15
Change-Id: I79a983192edef3c3560a4296cc9dea2dfc2ee1b0
Reviewed-by: Tony Sarajärvi <tony.sarajarvi@qt.io>
Use QScreen APIs instead.
Change-Id: Ie99af94fe4292223dbb165b3f5c1b74e8fe0498b
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
One can call addResult(value, index) twice and consequently set the
value twice by the same index. This seems rather strange and probably
should not be allowed. This commit rejects setting results when there's
already a valid result by that index. Consequently, this fixes memory
leaks caused by N-times-called addResult(..., index)
Fixes: QTBUG-86828
Change-Id: I77494f2cb73ce727ffad721cfcdcaa420899eb25
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
Using std::begin() and std::end() forces the user to have const begin()
and end() member functions being defined for the passed container. This
is because std::declval<T>() returns rvalue which forces the compiler
to select std::{begin, end}()(const Container &c) overloads and an test
for a presence of const {begin, end}() methods.
Change-Id: I9d96d9f73891ece53224f1741a1334500f7b35ad
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
This reverts commit 7544c242cb, which
reverted the first removal of the parameter under the assumption
that it caused flakiness in tests.
The flakiness was instead caused by changes to the wait functions
in QTest, so remove the parameter again.
Change-Id: I98154d5d7268375aebbcb09de757e75d9b765c5f
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
This commit restricts operator<<(QDebug lhs, QVariant rhs) to only work
if rhs is actually of type QVariant (instead of any type convertible to
QVariant). This is especially important as
a) we check in QMetaType whether (slightly simplified) QDebug{} <<
std::declval<T>() is valid, and if so, register a function which
simply uses the operator.
b) In QVariant, we ask the metatype system for the contained types
registered debug function and then use it.
If a type now does not have its own operator<< for QDebug, but is
implicitly convertible to QVariant containing itself, this would lead to
an infinite recursion, when trying to use qDebug with that type. The
registered function in a) would just convert the type to QVariant, and
then ask the QVariant to print itself.
Disallowing implicit conversions in qDebug in general was considered
(i.e. adding template<typename T> operator<<(T) = delete in QDebug ),
but discarded as it breaks too much code relying on conversions.
Fixes: QTBUG-87122
Change-Id: Ib709297670cbc6cc307efd0dfd8e5b0279df9414
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
If the difference between the families sizes is just 1 where one of them
is 0 in size then we can fallback to the family in that case.
Pick-to: 5.15
Fixes: QTBUG-87267
Change-Id: I62b25b06c88000b4d7defe91871c07873b1fc792
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@qt.io>
QtConcurrent::filtered test-cases for move-only containers were failing
to compile, because it is assumed that the passed container should have
value_type defined.
Change-Id: I3e9e5ebc07704cb98a15b125ae8bd5b5a84d497a
Reviewed-by: Andreas Buhr <andreas.buhr@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
QFuture::takeResult() currently returns std::vector instead of QList,
because QList does not support move-only types. Disable this method
until QList is fixed to work with move-only types in Qt 6.1.
Also did minor doc-fixes.
Change-Id: I87feaf75d9433a3b540edd00039c3e21d6994985
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
As propertyobservers can execute arbitrarily complex code, they can also
modify the obsever list in multiple ways. To protect against list
corruption resulting from this, we introduce a protection scheme which
makes the list resilient against modification.
A detailed description of the scheme can be found as a comment in
QPropertyObserverPointer::notify.
Task-number: QTBUG-87153
Change-Id: I9bb49e457165ddc1e4c8bbdf3d3c9fbf5ff27e94
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Implement the better rounding mechanism that was previously blocked
by requiring C++14 to be constexpr.
Change-Id: I4e5b179ce0703f5c0b41c3f0ea00d28dfe53740c
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
The original restriction to UniformBuffer was due to the GL backend
where there is no GL buffer object for QRhiBuffers with usage
UniformBuffer. However, we can still implement this for cases when
there is a true GL buffer object underneath. With other backends it
should all work as-is already.
This becomes useful when one has buffers with usage Vertex that need
full updates every frame. (f.ex. instance data)
Unfortunately this involves renaming the function. But while at it, add
an autotest case as well.
Change-Id: Iff59e4509a8bae06654cc92fe8428bd79eb012fb
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
The qsb files seem to be out of date for those two.
Change-Id: Id832e872667cac4d364e13c440011109a6dbdc7f
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
Draw 25000 cubes while doing a uniform buffer update for each.
Change-Id: I2216641c8bf7c6ea147fe3edd679317b472e1f04
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
Pending results were never cleared by result store. This led to memory
leaks when the results never transitioned to "visible" results
Change-Id: I674302eb51542ad5f4d918da68d616428c73ae9f
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
It was previously casting enum values to int, which produced unhelpful
answers when a test failed. Better to have them as enum values that
get printed as their names, which are actually informative.
Change-Id: I1bf2971b1426bdbbc321bb48f45ee3e5799e76ec
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
QDateTime's addDays(), addMonths() and addYears() neglected to check
for validity before doing their job, with the result that they could
produce "valid" (but wildly inappropriate) results if used on an
invalid date-time. Added tests for this case (and the boundary).
Change-Id: I7b0d638501cb5d875a678cde213547a83ed7529e
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QDateTime's range of possible values is wider than anyone generally
needs, but let's not do confusing things when someone does overflow
it.
Change-Id: Ifbaf7a0f02cd3afe7d3d13c829bf0887eba29f7f
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Previously, a QDate representing more than about 0.3 gigayears before
or after the epoch would overflow the millisecond count and produce a
"valid" date-time that didn't represent the date and time passed to
its constructor. Changed to detect such overflow and produce an
invalid date-time instead, if it happens.
Corrected some tests that wrongly expected to be able to represent
extreme date-time values with every time-spec. The (milli)seconds
since epoch are from UTC's epoch, so converting to another offset,
zone or local time may give a value outside the actual range. Added
some tests for the actual exact bounds.
Task-number: QTBUG-68855
Change-Id: I866a4974aeb54bba92dbe7eab0a440baf02124f0
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Also correct two comments which lied about the year being tested.
Change-Id: I5be491a2b2c1e0c3b49d6ff9a8de852e17321cff
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
As suggested by the message from QTestLib. This, indeed, fixed the
sadistic test. Also, make sure resources are not leaked.
Pick-to: 5.15
Fixes: QTBUG-87009
Change-Id: Id693a5a562ea5ebacc853e5fc0ab9654ba851e72
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Currently
A a;
QVariant::fromValue(&a).value<const A*>() == nullptr;
Still casting non const to const is safe, and worked in Qt5.
After this change
A a;
QVariant::fromValue(&a).value<const A*>() == &a;
Change-Id: I257049d084c712b00a338a2943d379aa478e0981
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Change types returned and accepted by capacity-related QArrayDataPointer
functions to qsizetype:
1) QArrayData (underlying d-ptr) works with qsizetype
2) QArrayDataPointer::size is of type qsizetype
3) All higher level classes that use QADP (e.g. containers)
cast capacity to qsizetype in their methods
Additionally, fixed newly appeared warnings through qtbase
Change-Id: I899408decfbf2ce9d527be7e8b7f6382875148fc
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This time based on grepping to also include documentation, tests and
examples previously missed by the automatic tool.
Change-Id: Ied1703f4bcc470fbc275f759ed5b7c588a5c4e9f
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
Unit tests are added to make sure QtConcurrent works on
move-only sequences.
Change-Id: I1d066f75ceab9cef98832e96c5827103cbfd72a8
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
Even before adding support for fractional hours, a fraction of a
minute might potentially have represented a whole number of seconds by
a fractional part that, due to rounding, was less than the whole
number of seconds by less than half a millisecond. Previously, the
parsing would have clipped the fractional part at 999 milliseconds, in
the preceding second, instead of correctly rounding it up to the whole
second.
For QTime::fromString(), which can't represent 24:00, and for
TextDate, which doesn't allow 24:00 as a synomym for the next day's
0:0, applying such rounding to 23:59:59.999999 would produce an
invalid result from a string that does represent a valid time, so use
the nearest representable time, as previously.
Added some tests and amended others.
[ChangeLog][QtCore][QDateTime] QDateTime and QTime, in fromString()
with format ISODate or TextDate, now allow a fractional part of the
hour, minute or seconds to round up to the next second (hence
potentially into the next minute, etc.) when this is the closest
representable value to the exact fractional part given. When rounding
up would turn a valid result into an invalid one, however, the old
behavior of clipping to 999 milliseconds is retained.
Change-Id: I8104848d246cdb4545a12819fb4b6755da2b1372
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Andreas Buhr <andreas.buhr@qt.io>
Previously we used 0:0 on the next day, which might fall in a
fall-back's gap.
[ChangeLog][QtCore][QDateTime] When fromString() reads 24:00 in ISO
format, it now uses the start of the next day, rather than 0:0 on the
next day. This only makes a difference if the next day's first hour is
skipped by a time-zone transition.
Change-Id: Ib81feca5dc09fa735321b6ab76d5d118d6db6fd2
Reviewed-by: Andreas Buhr <andreas.buhr@qt.io>
Reviewed-by: Paul Wicking <paul.wicking@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
We were already using the 'native' nomenclature when referring to these
kinds of APIs, e.g. when talking about native handles, or the existing
QPlatformNativeInterface on a QPA level. Using 'native' for the user
facing APIs also distinguishes them from the 'platform' backend layer
in QPA and elsewhere.
Change-Id: I0f3273265904f0f19c0b6d62471f8820d3c3232e
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Try to get rid of APIs that use raw 'const {char, QChar} *, length'
pairs. Instead, use QByteArrayView or QStringView.
As QStringConverter is a new class, simply change the API to what we'd like
to have. Also adjust hidden API in QStringBuilder and friends.
Change-Id: I897d47f63a7b965f5574a1e51da64147f9e981f6
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Introduce a bunch of helper functions to manipulate compiler
flags and linker flags for the
CMAKE_<LANG>_FLAGS_<CONFIG>
and
CMAKE_<LINK_TYPE>_LINKER_FLAGS_<CONFIG>
CMake variables.
These variables can be assigned and modified either in the cache
or for a specific subdirectory scope, which will apply the flags
only to targets in that scope.
Add qt_internal_add_optimize_full_flags() function which mimics
qmake's CONFIG += optimize_full behavior.
Calling it will force usage of the '-O3' optimization flag on supported
platforms (falling back '-O2' where not supported).
Use the function for the Core and Gui subdirectories, to enable full
optimization for the respective Qt modules as it is done in the qmake
projects.
To ensure that the global qmake-like compiler flags are assigned
eveywhere,
qt_internal_set_up_config_optimizations_like_in_qmake() needs
to be called after Qt global features like optimize_size and
optimize_full are available.
This means that qtbase and its standalone tests need some special
handling in regards to when to call that function.
Task-number: QTBUG-86866
Change-Id: Ic7ac23de0265561cb06a0ba55089b6c0d3347441
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
QUrl hostnames must be compliant with STD3, but we must somehow accept
file paths that begin with double slash but aren't valid hostnames.
Because the file URI spec requires us to start with "file://" anyway, we
can represent those with four slashes. Note that on Unix "//X/y" is a
valid but local file path. If given to QUrl::fromLocalFile(), if the
path at the root does parse as a hostname, we will still try to
normalize (the above becomes "file://x/y").
[ChangeLog][QtCore][QUrl] Changed QUrl::fromLocalFile() to accept
Windows UNC paths whose hostname component is not a valid Internet
hostname. This makes QUrl able to accept extended-length paths (\\?\),
device namespace (\\.\), WSL (\\wsl$), etc.
Pick-to: 5.15
Fixes: QTBUG-86277
Change-Id: I3eb349b832c14610895efffd1635759348214a3b
Reviewed-by: David Faure <david.faure@kdab.com>
No change in testing, juts changed names and order.
Pick-to: 5.15
Change-Id: I3eb349b832c14610895efffd16357498454bcd52
Reviewed-by: David Faure <david.faure@kdab.com>
Also add the very same operators to the QBasicUtf8StringView
class to overcome the compiler issues seen on gcc.
Fixes: QTBUG-86481
Change-Id: I12484455ebd3b7b38d4ad67c38977d76f9b3ddfa
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
In tst_QDateTimeEdit::dateEditCorrectSectionSize(), only the final
displayed date was tested; many of the tests checked day of week (by
name), which might well match despite actually selecting a different
date than the one intended. So test the date is as expected, too.
In the process, tidy up the code, give values more informative names,
turn a trivial static function into a lambda.
Change-Id: I0491159e9ee2f3cfdcf8a194c723be92c190c2a8
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Also adjust the QString constructor from QByteArray to ignore
\0 characters in the string (and not terminate conversion there).
[ChangeLog][QtCore][QString] Constructing a QString from a QByteArray
will not stop at intermediate '\0' (null) characters in the string as
in Qt 5, but will convert all characters in the byte array.
Change-Id: I1f6bfefe76dfa9072b165903fec7aa4af1abd882
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Fix remaining places that still exercised it.
Task-number: QTBUG-85700
Change-Id: I84562f53439197141343831c0b9f88983689e6bf
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
Previously the ISO time format would tolerate trailing cruft at the
end in various cases even though there might be an offset specifier
after the time, which should *not* be separated from it by anything
(not even the spaces we originally planned to still tolerate).
The RFC date format is forgiving about space, as is suitable for
parsing of RFC-822 headers, but the other formats should match the
handling in QDateTimeParser, which rejects any dangling cruft.
At the same time, since this required a re-write of
fromIsoTimeString() in any case, add support for the ISO format that
gives the hour a fractional part and skips minutes and
seconds. Previously we only had support for fractional minutes (with
no seconds). The hour without even a fractional part is also valid.
Reworked the documentation of Qt::DateFormat as it was wrong in
places, inconsistent in its formatting and incomplete. Adjusted some
tests to match the new behavior. A fraction separator with no
following digits should have been recognized as an error previously
and now is.
[ChangeLog][QtCore][QDateTime] The ISODate and ISODateWithMs formats
now reject trailing cruft (including spaces) at the end of a time
string. They also gain support for parsing hour-only formats,
including the hour-with-fractional-part format.
Task-number: QTBUG-86133
Change-Id: I38ad1479ae033407f7df97ffbeb7c4bcd463d04a
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Reviewed-by: Paul Wicking <paul.wicking@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The name toSecsSinceEpoch() gave no hint to the fact that the test
was, in fact, *also* testing round-tripping of the ISODateFormat.
Since there are other tests for string conversion, make this a simple
test of toSecsSinceEpoch().
It did the round-tripping via two methods with overly-terse names that
might just as well be local lambdas - now redundant, so removed.
Change-Id: I1e4fb1cc90224312c995596a8f3fe2bc5d9dfa15
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
During delivery of a TouchBegin event, if no widget accepts it,
we begin treating the first touchpoint as a synth-mouse, as before.
If a second touchpoint is pressed or released in any order, it's
irrelevant: the fake mouse button is released as soon as the first
touchpoint is released. This fixes the bug that such a scenario
caused the mouse release not to be sent, so that a widget could get
"stuck" in pressed state.
Done-with: Tang Haixiang <tanghaixiang@uniontech.com>
Fixes: QTBUG-86253
Pick-to: 5.15
Change-Id: I7fbbe120539d8ded8ef5e7cf712a27bd69391e02
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
QAbstractItemModel::canFetchMore and fetchMore implementations might only
add a few rows to the model each time they are called. The item views don't
generally expect that, and don't call fetchMore repeatedly, even if there
would be room in the itemview for more data. This problem cannot be
generally solved for all item views, as it would require in repeated expensive
laying out of items.
For nested indexes in a treeview however, we can try to fetch enough child
rows to populate the screen when the item is laid out by repeatedly
calling canFetchMore and fetchMore. To calculate how many items have space,
apply the same heuristics as in the scrollContentsBy implementation to
guess the number of items that can fit into the viewport.
Created test case for the fix.
Done-with: Doris Verria <doris.verria@qt.io>
Fixes: QTBUG-85366
Pick-to: 5.15
Change-Id: I54f95552993873dd4cba80b0f70f4db9d98ddc1d
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Reviewed-by: Andy Shaw <andy.shaw@qt.io>
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
Reports on the Web have it that there's nothing guaranteeing a
driver does any actual freeing of resources then doing
vkFreeCommandBuffer for a command buffer from a command pool that
does not have VK_COMMAND_POOL_CREATE_RESET_COMMAND_BUFFER_BIT, thus
leading to continuously growing resource usage with our current
allocate/free pattern. It could be that this is the source of out
of memory problems we are seeing on some Android devices.
Instead of just going straight for said command pool flag and doing
ResetCommandBuffer for the command buffers individually, take one
step further and use per-slot (slot being 0 or 1 if QVK_FRAMES_IN_FLIGHT
is 2) command pools. The current pool is reset in each
beginFrame/beginOffscreenFrame, moving all allocated command buffers
to the initial state (while other command buffers with the other pool
are not affected).
This may be (while impossible to tell from just guessing based on the
spec) our best approach to command buffer allocation since a Vulkan
implementation can likely just use some simple per pool allocator,
knowing that we never want to free or reset individual command buffers,
but we rather only reset the whole pool at once.
The option of importing an existing VkCommandPool when creating the
QRhi instance is now gone, but there was probably no point in offering
that in the first place.
When it comes to VK_COMMAND_POOL_RESET_RELEASE_RESOURCES_BIT it will
not be set unless releaseCachedResources() (in Qt Quick this is hooked
into QQuickWindow::releaseResources()) was called. What this does in
practice is unknown, but have an option to set it now and then if the
application really wants.
While we are at it, rename secondaryCbs to activeSecondaryCbStack to
indicate what it really is. (it's a stack as each call to
beginExternal() pushes a new one, while each endExternal() pops)
Change-Id: I2e5c1fad26d794e1f56c778e38f750998d706d84
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
