Compilers are catching up, so some #if 0 codepaths can now be
conditionally enabled.
Change-Id: Ia9e87a096bc2ae4789ab390a9170d9c1eb9690d6
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
Q_CC_GNU is defined on compilers that masquerade as GCC (Clang, ICC),
so using it to work around GCC-specific bugs is wrong. Introduce a
local define for _only_ GCC and use it in place of Q_CC_GNU.
Drive by: version-fence a test we now know it's been fixed upstream,
and correct the link to the corresponding bug report.
Pick-to: 5.15 6.0 6.1
Change-Id: I9059d6e6bf86157aca71590ac22afb1a1c114313
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
This function returns the metatype corresponding to the metaobject, or
an invalid metatype for namespaces.
This works as follows: First we increment the metaobject revision for
new metaobjects. Metaobjects with older revisions are handled by doing a
lookup by name. That fallback is also used for dynamic metaobjects (from
QtDBUS and those created by QMetaObjectBuilder).
For new metaobjects, we store the metatype in its metatype array, behind
the property metatypes. This avoids any changes to the property and
method metatype extraction logic: For properties, the metatype access
does not change, as the new metatype is after their metatypes. For
method metatypes, we already have an indirection layer (using offsets),
so by adjusting those offsets by one, the same logic keeps working.
To distinguish between namespaces and dynamic metaobjects, namespaces
store the metatypeinterface pointer for void in the metatype array,
whereas dynamic metaobjects store a nullptr.
One nice additional benefit is that this simplifies the generator logic
in moc, as the metatype array is now never empty.
Task-number: QTBUG-92077
Change-Id: Id3f920f28553f12032a71a1a87dad29e5374dbe7
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Fix operator checks for containers whose value_type equals themselves.
It does not make sense to recurse on value_type in that case. Thanks to
std::disjunction having short-circuiting semantics, we can avoid that
issue by checking first whether T is T::value_type.
As a drive-by, check for value_type typedef before checking for
begin/end in is_container. This works around an issue in gcc <= 8.1,
which fails to correctly SFINAE the case where begin and end are private
methods.
Pick-to: 6.0 6.1
Fixes: QTBUG-89456
Change-Id: I27305a7cfe050f13a279c07f00bc229c01daa25b
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
The code in qtdeclarative did not do anything at all anymore.
Change-Id: Idd97145cb74aeb4f43dfce2f282a765e90945073
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
There may be a race where e.g. thread 'B' is woken up by a queued invoke.
At the same time thread 'A' asks 'B' to quit, which will set various
atomics (some important ones are 'interrupt' in the dispatcher and
'exit' in the event loop), but it does _not_ try to send another wake
since there is already an unhandled wake triggered by 'B' itself.
Sadly 'B' reads the 'exit' atomic before 'A' updates it.
Then, slightly before, 'B' sets 'interrupt' back to 0, 'A' write 1 to
it, meaning 'A's interrupt is ignored. Then, since there is no
interrupt, 'B' goes back to waiting for events, leaving the thread alive
and running instead of quitting.
Maybe this has unforeseen consequences (one consequence is that it will
return and re-enter the event dispatcher once more, possible
unnecessarily)
Fixes: QTBUG-91539
Pick-to: 6.1 6.0 5.15
Change-Id: Ie6f861f42ffddf4817d5c8af2d764abe9d9103c2
Reviewed-by: Alex Trotsenko <alex1973tr@gmail.com>
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Based on the discussion in https://codereview.qt-project.org/c/qt/qtbase/+/336742
changing our types from RELOCATABLE to PRIMITVE is fine.
Change-Id: Ica867203aa813d19fdfd3753fc4ff36ef4332fc3
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Before this change, the QMetaType for T const* where T is derived from
QObject would not store the static QMetaObject of T. This commit changes
this. As a consequence, the metatype system can now convert between
const and non-const pointers to QObject. Note that this allows casting
const away; but so does C++ with const_cast.
In addition, a new flag, QMetaType::IsImmutable is introduced, and used
to tag the metatypes of pointer to const types. This allows code to
discern between pointers to mutable and const QObjects, which is
relevant for the QML engine.
Task-number: QTBUG-82354
Change-Id: I3e4e4f39f565bd99a65e161528ce5304df73d6d6
Reviewed-by: Shawn Rutledge <shawn.rutledge@qt.io>
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
A recent change ( a7ca8b1a28 )
led to failure of binding removal in setInterval().
This was fixed by introducing setterScope.
This patch add unit tests for this regression.
Change-Id: Ic8da1f2d82ad6c8ccd81c9b1eff72d42cf75f28a
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
The initial implementation and the commit
c00ab6f8ea was wrong:
* env->findClass() in fact returns a global reference, and in any
case we shouldn't be calling that, instead QJniObject would be
enough.
* The size param provided to env->RegisterNatives was wrong.
* A test for registerNativeMethods() is added to ensure such break
is not repeated again.
Task-number: QTBUG-89633
Pick-to: 6.1
Change-Id: I4d3a6a9270755f465c40add25521fb750dd4de0a
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Ville Voutilainen <ville.voutilainen@qt.io>
gcc 9 consumed enourmous amounts of memory building the test, regularly
dying on a VM with 4GB RAM. Splitting it up helps.
As a drive-by, use inline static variables, and rename the header used by
other tests to tst_qmetatype_common.h.
Change-Id: Ib716d8e3506aac6c87845e57b04cb1a4f6c68387
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Introduction of QObjectCompatProperty requires every write to
the property to be examined whether it is OK or should be replaced
by a setValueBypassingBindings/markDirty combination. The existence
of operator= make this difficult as it is easy to miss places where
it is written. By not having operator=, we can help developers
make sure they had a conscious decision about each write to the
property.
Change-Id: Ia61ea4722eb0bab26ce7684b85dd03d710cd1751
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
This is in line with QML where
import QtQuick 2.15
Rectangle {
width: 100
height: 100
color: "red"
Rectangle {
id: inner
x: 10
y: x
width: 50
height: 50
onYChanged: { console.log("hey"); inner.x = 10}
TapHandler {
onTapped: inner.x = 20
}
}
}
results in a binding loop warning when the tap handler triggers. While
the change handler would only run once, we cannot statically determine
if we need to loop once, twice, or if there actually is a diverging
loop. Thus we unconditionally warn about the binding loop and stop
executing the binding.
As a drive-by, verify in the related test that a change handler which
overwrites its properties binding itself removes the binding.
Change-Id: I5372019c2389ab724c49cd7489ecbd3ebced1c69
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
As part of Qt 6 restructring for the extras modules, this change exposes
the Jni APIs which are very important for Android platform. This patch
adds the APIs QJniObject, QJniEnvironment, QJniExceptionCleaner based
from private QtCore and QtAndroidExtras.
The Jni interface is cross-platform which justifies the name, but
currently, this API is used mainly for Android, and the naming comes
generic without Android keyword to avoid any future limitation on
supporting other platforms.
[ChangeLog][QtCore] Add new QJniObject, QJniEnvironment and
QJniExceptionCleaner APIs.
Task-number: QTBUG-89482
Fixes: QTBUG-89633
Change-Id: I4382dd53a225375759b9d042f6035a4a9810572b
Reviewed-by: Ville Voutilainen <ville.voutilainen@qt.io>
We missed takeBinding as a supported operation on Q(Untyped)Bindable.
To avoid adding version checks to code dealing with QBindableInterface,
we simply synthesize takeBinding as a combination of binding to retrieve
the binding and setBinding with a default-constructed
QUntypedPropertyBinding.
Change-Id: I43803a0dfe210353d0235f0373d2257f75ffe534
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
We have plenty of places where we add some squares and take a square
root; this may be done more accurately and faster by hypot().
Introduce QHypotHelper to handle hypot with more than 3 parameters,
and with 3 when the C++17 version is missing (which it never should
be). Include an overload taking arbitrarily many valus and ensure that
we can use qHypot() with qfloat16. Illustrate with some example uses,
add some tests.
[ChangeLog][QtCore][QMath] Header <QMath> now provides qHypot(), an
implementation of std::hypot() taking arbitrarily many numeric values,
including support for qfloat16, while avoiding the overflow and
underflow problems that arise when naively taking the square root of a
sum of squares.
Change-Id: Ia4e3913fe83fc27d17d8e7f1a52f03ad445c1fed
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
To optimize certain operations, it can be useful to know whether we are
currently evaluating a binding. For instance, we have properties whose
storage is only alloctaed on-demand when they are set. However, we would
also allocate them if they are used in a binding context, as we would
otherwise not properly track the dependency. Using
isAnyBindingEvaluating in the getter, we can detect this
situation, and avoid the allocation if it returns false.
This API is private for now, as it exposes some internals of the
property system and should be used with care. As it needs to access the
TLS variable, it also has a non-negligible cost.
Change-Id: I373aabee644fe7020b2ffba7d6a0ad9a1e1b4ec0
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
The timeout defaults to 0, give it 5s.
Pick-to: 6.0
Change-Id: I975810a1ecee8bb8b3a3f143f1379a9a09589a40
Reviewed-by: Alex Trotsenko <alex1973tr@gmail.com>
The test has been failing frequently, recently.
Pick-to: 6.0
Change-Id: I3ae00a64f67e4b6a0b5ade0c660805f4d12f8317
Reviewed-by: Oliver Wolff <oliver.wolff@qt.io>
Original QML-specific connection mechanism ignores the receiver argument
and uses sender as receiver. This causes uncontrollable memory growth
in certain cases as connections on receiver persist even after receiver
is destroyed
New connect() with receiver parameter uses underlying API correctly,
disconnect is provided for the symmetry (not sure it's really needed)
Task-number: QTBUG-86368
Pick-to: 5.15 6.0
Change-Id: I4580d75b617cb2c4dfb971a4dfb8e943e325572b
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
In Qt 5, QVariant::fromValue<T> would not compile unless
Q_DECLARE_METATYPE(T) was used, and Q_DECLARE_METATYPE(T) would lead to
a compile error if T were not copy constructible.
In Qt 6, we do not require Q_DECLARE_METATYPE before using fromValue,
and QMetaType itself works with non-copy constructible types just fine.
However, QVariant still requires it, thus we need to now enforce this in
fromValue itself.
Change-Id: Ib6964a438d8c46033dd3a037b9d871de2b42e175
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Those serve no purpose anymore, now that the .pro files are gone.
Task-number: QTBUG-88742
Change-Id: I39943327b8c9871785b58e9973e4e7602371793e
Reviewed-by: Cristian Adam <cristian.adam@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Kai Koehne <kai.koehne@qt.io>
If we create a QBindable from a const property, we should obtain a
read-only interface. Besides implementing this feature, this patch adds
a isReadOnly method to Q(Untyped)Bindable which can be used to check
whether one can modify the property via the bindable interface.
Task-number: QTBUG-89505
Task-number: QTBUG-89469
Change-Id: Ic36949a5b84c5119e0060ed0a1cf4ac94a66f341
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
Remove the qmake project files for most of Qt.
Leave the qmake project files for examples, because we still test those
in the CI to ensure qmake does not regress.
Also leave the qmake project files for utils and other minor parts that
lack CMake project files.
Task-number: QTBUG-88742
Change-Id: I6cdf059e6204816f617f9624f3ea9822703f73cc
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Kai Koehne <kai.koehne@qt.io>
As lists of QStrings and QByteArrays are sequentially iterable the base
types should really also be.
The only problem is that they don't have methods to remove items from
the back or the front, but that is well within what we can support with
QSequentialIterable.
Change-Id: I2ab551e7b11a092aba363fb4012d131bbc4b11b4
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Complete search and replace of QtTest and QtTest/QtTest with QTest, as
QtTest includes the whole module. Replace all such instances with
correct header includes. See Jira task for more discussion.
Fixes: QTBUG-88831
Change-Id: I981cfae18a1cabcabcabee376016b086d9d01f44
Pick-to: 6.0
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
It has been failing consistently, recently.
Pick-to: 6.0
Change-Id: I71b2e8857c3d5ce86ad17864c95aac7265ed9a8a
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
This adds functionality for marking properties (QProperty and related
classes) manually as dirty. This facilliates the integration of bindable
properties with non-binable properties and makes it possible for
bindable properties to change due to external events.
Fixes: QTBUG-89167
Change-Id: I256cf154d914149dacb6cadaba92b13c88c9d027
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
Remove a second argument to the setSocket(qintptr, bool) function as
it makes the API harder to understand.
Change-Id: Ib1852a4e9d96adde35bfbf0fe03b386d9ded395a
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Apparently some library definitions went overboard, link them directly.
Pick-to: 6.0
Change-Id: I009737f7e3edff5619241b700a627dc4e25e6018
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
Technically, having a single constructor limits the use-cases for this
class. We should take into account that:
- an opened socket descriptor must be available at the moment of
construction;
- the constructor unconditionally enables the notifier (the possible
solution
notifier = new QSocketNotifier(...);
notifier->setEnabled(false);
is suboptimal due to heavy operations inside the event dispatcher);
- for these reasons, QSocketNotifier most often cannot be a member of
another class (we have an extra allocation and indirect access).
This patch addresses this shortcoming by making it possible to set the
socket descriptor at a later point:
[ChangeLog][QtCore][QSocketNotifier] Added setSocket() and an additional
constructor which requires no socket.
Change-Id: I2eb2edf33ddafe99e528aac7d3774ade40795e7a
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
When an eager binding triggers a setBinding call, we end up with a
special kind of binding loop:
setBinding() -> evaluate -> notifyObserver
^ |
| /
----------------------------
We now catch set condition, and set the binding status to BindingLoop
(with a distinct description).
Task-number: QTBUG-87153
Task-number: QTBUG-87733
Pick-to: 6.0
Change-Id: I9f9915797d82eab820fc279baceaf89d7e5a3f4a
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
The intention was always that you can define properties that do
not require a changed signal. But having to explicitly pass
a nullptr as signal parameter into the macro is ugly, so
use the cool QT_OVERLOADED_MACRO to make it optional.
Pick-to: 6.0
Change-Id: I0ce366d043850f983c968d73c544d89933c48df9
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
If a QBindable is created from a computed property, it is not possible
to actually set a value or a binding. If we try to do it anyway, we'd
get a crash. Thus we now check whether the function pointer is null
before invoking it.
Pick-to: 6.0
Task-number: QTBUG-87153
Change-Id: I5bedb9080ccf79d9b8166b80d5733d095ed76f8d
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Add QMetaType::hasRegisteredDataStreamOperators() to complement
the method to check whether a data stream operator exists.
Fixes: QTBUG-82916
Change-Id: Ib2f841131b7c401d5a3ae76d49104e41697c4eac
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Avoid spurious bindings by resetting the binding state before calling
the setter of eager properties.
Fixes: QTBUG-88999
Pick-to: 6.0
Change-Id: I1e3b5662307d906598335a21d306be9c606529d4
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
QMetaObject::indexfOfMethod returns the method corresponding to a
specific signature. In QML, we however only want any of the methods with
a given name (and do overload resolution at a later point).
For this usecase this patch introduces the internal
QMetaObject::firstMethod function.
Change-Id: Ie3820354edffb273c4cbe1399201a955ebe79344
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
std::optional<int> is the wrong datatype to use for compare.
First and foremost, it can't be used in the idiomatic form of
auto r = a.compare(b);
if (r < 0) ~~~ // a is less than b
if (r > 0) ~~~ // a is greater than b
which we *already* feature in Qt (QString, QByteArray).
Also, std::optional<int> (explicitly) converts to bool, which is
a trap, because the result of the comparison can be accidentally
tested as a bool:
if (a.compare(b)) ~~~ // oops! does NOT mean a<b
Not to mention extending this to algorithms:
auto lessThan = [](QVariant a, QVariant b) { return a.compare(b); }; // oops!
std::ranges::sort(vectorOfVariants, lessThan);
which thankfully doesn't compile as is -- std::optional has
an *explicit* operator bool, and the Compare concept requires an
implicit conversion. However, the error the user is going to face
will be "cannot convert to bool because the operator is explicit",
which is deceiving because the fix is NOT supposed to be:
auto lessThan = [](QVariant a, QVariant b) { return (bool)a.compare(b); }; // big oops!
Instead: backport to Qt the required subset of C++20's <compare>
API, and use that. This commits just adds the necessary parts
for compare() (i.e. partial ordering), the rest of <compare>
(classes, functions, conversions) can be added to 6.1.
Change-Id: I2b5522da47854da39f79993e1207fad033786f00
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
(cherry picked from commit 3e59c97c3453926fc66479d9ceca03901df55f90)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
Q_MOVABLE_TYPE was conceived before C++ had move semantics. Now, with
move semantics, its name is misleading. Q_RELOCATABLE_TYPE was
introduced as a synonym to Q_MOVABLE_TYPE. Usage of Q_MOVABLE_TYPE
is discouraged now. This patch replaces all usages of Q_MOVABLE_TYPE
by Q_RELOCATABLE_TYPE in QtBase. As the two are synonymous, this
patch should have no impact on users.
Pick-to: 6.0
Change-Id: Ie653984363198c1aeb1f70f8e0fa189aae38eb5c
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Except for types marked as unresolved, we're doing it anyway - the only
difference is that now we skip looking up the metatype by typeid.
[ChangeLog][QMetaProperty][Important Behavior Change]
QMetaProperty::typeName returns now always the same name as name() of the
corresponding metatype. This can cause a change for enum properties
which were not fully-qualified.
Change-Id: I1f57743948b7262ac06095d3bbc838d620f6e481
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
