Traditionally when calling reserve it's because you expect to append
up to X amount of bytes. We should keep that behavior the same.
With another patch still in the works current behavior caused an issue
with QStringBuilder in QNAM, as mirrored in the testcase attached.
Change-Id: I9792a8f158fc9235e3de48ac8b06ac2c10e7f3dc
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Views / delegates absolutely *adore* hammering data(). A simple
QListView showing a couple of dozens entries can call data()
a hundred of times on the first show.
Back of the hand calculation,
* 2 times per visible item (sizeHint() + paint()),
* times 9 roles used by the default delegate,
* times 20 visible items
= 360 as a bare minimum, assuming the view doesn't redraw twice
accidentally. Move the mouse over the view, and that'll cause
a full update with certain styles: 360 calls to data() per update.
This has an overhead visible in profilers. The model's data()
has to re-fetch the index from its data structure and extract
the requested field every time.
Also, QVariant is used for the data interexchange,
meaning anything that won't fit in one is also a memory allocation.
This problem will likely be gone in Qt6Variant as that
will store sizeof(void*) * 3, meaning QImage/QPixmap and similar
polymorphic classes will fit in a QVariant now...
So I'm trying to to remove part of that overhead by allowing
views to request all the data they need in one go. For now,
one index a a time.
A view might also store the data returned. The idea is that
the same role on different indexes will _very likely_
return variants of the same type. So a model could move-assign
the data into the variant, avoiding the memory allocation
/deallocation for the variant's private.
This patch:
1) Introduces QModelRoleData as a holder for role+data.
2) Introduces QModelRoleDataSpan as a span over QModelRoleData.
The idea of a span type is twofold. First and foremost, we are
in no position to choose which kind of container a view should
use to store the QModelRoleData objects for a multiData() call;
a span abstracts any contiguous sequence, leaving the view free
to do whatever it wants (statically allocate, use a vector, etc.).
It also solves the problem of efficient passing the roles and
gathering the returned variants from multiData().
3) Add multiData(), which populates a span of roles for a given
model index. The main advantage here is that a model can fetch
all the needed information for a given index just once, then
iterate on the span and provide data for each requested role.
Cf. this with data(), where every call has to re-fetch
the information for the index.
A couple of models have been ported to multiData(), as well as
QStyledItemDelegate.
[ChangeLog][QtCore][QModelRoleData] New class.
[ChangeLog][QtCore][QModelRoleDataSpan] New class.
[ChangeLog][QtCore][QAbstractItemModel] Added the multiData()
function.
Change-Id: Icce0d108ad4e156c9fb05c83ce6df5f58f99f118
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
When the view was already sorted by the column and order given to
sortByColumn(), a resort was not triggered which is a regression since
d0f909f8db.
Therefore don't rely on sortIndicatorChanged() in this case and trigger
an explicit sort as it's done when no user-triggered sorting is enabled
Fixes: QTBUG-86268
Change-Id: I3ec30ab81f304d5a19cef6d67b8a81c21b4a112d
Pick-to: 5.15
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
So far QPropertyAlias was limited to working with QProperty<T>.
Change the implementation, so it can be constructed from any
property or even a QBindable<T>.
Change-Id: I175cffe94a9ef332367d39faa976eb065b0e6ffe
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Add a new BINDABLE declaration to the Q_PROPERTY() macro that tells moc
where to find the QBindable for the property.
Add a QUntypedBindable base class to QBindable<T> that gives access to
generic functionality and checks argument compatibility at runtime.
QBindable<T> will still do static checking at compile time.
Add QMetaProperty::isBindable() and QMetaProperty::bindable()
to be able to dynamically access the binding functionality.
Change-Id: Ic7b08ae2cde83fd43e627d813a886e1de01fa3dc
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Add a compatibility property class that makes porting to the new
property system as simple as possible.
Binding evaluation for those compat properties is eager, as we
do not control possible side effects of the code in the existing
setters.
Change-Id: Ic56347abb49e40631ec73e88c6d40d4bdb05ca29
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Add a QObjectComputedProperty. This class doesn't store the data
itself, instead relies on a getter method to compute it's value.
As the property is read-only, one can not bind to it, but it can
be used in other property bindings.
Change-Id: I0f6bffdd9f80f1d0829826f93a47257f2b3127af
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Add Q_OBJECT_BINDABLE_PROPERTY() macro that can be used to define
a bindable property inside QObject.
The macro and the class behind it creates storage for a property
that is bindable inside a QObject or QObjectPrivate. The property
only uses as much space as the data contained, ie. it has no
storage overhead, as long as no bindings are being used.
Bindings are being stored and looked up in the QBindingStorage
associated with the owning object.
Change-Id: I1dadd7bddbad6fbf10cfa791d6461574b9db82dd
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
QBindingStorage is a class that can store a set of binding objects
for the properties of a QObject. This will get used to reduce the
memory overhead of the property system when adding bindable properties
to QObject based classes.
The binding storage has a pointer to the TLS entry containing the
currently evaluating binding. Like that we avoid repeated TLS
lookups and reduce the overhead of the property system to one
pointer lookup and one compare for the case that properties
aren't being used.
Each QObject now owns one binding storage object, that can be used to
store binding data for properties that members of the QObject.
Change-Id: I27427c03c2ba281f072e074be96147bdbcaac246
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Add a private QBindableInterface and a public QBindable<T>
class, that will be the API interface for accessing bindings
for properties in QObject.
The QBindable class gives access to all aspects of
the property related to bindings. This includes setting
and retrieving bindings, installing observers and creating
a direct binding on this property.
Change-Id: Iaead54d2bd6947bd2cda5052142b2a47dd8bf7c4
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
These look rather weird, an explicit property.setBinding() call
is simply better in this case, and also more aligned with the API
we can offer in QObject.
Change-Id: Ifb00fd47a75e6b3bc94e34bf49e4f13249565bfe
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
And all related functionality. This is being replaced by
Q_BINDABLE_PROPERTY and Q_OBJECT_BINDABLE_PROPERTY in the
next few commits. The new infrastructure coming will play
nicer along with the existing property system.
Commented out some autotests, that will get reimplemented
with the updated infrastructure.
Change-Id: I50c30bd4d5c6c6b6471f8eb93870e27d86f5a009
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
Rename QPropertyBase to QPropertyBindingData, as it contains the
data related to bindings. The new name fits better, as the data
can now also live somewhere else than the data strored in the
property.
Change-Id: I489efb86ad2e0bad2740c9d1aa74506fe103d343
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
Since we will be storing property data differently in most cases,
having this special case would create too many additional complications.
Change-Id: I27042b0730559bb375d8e3c07324398403a9885d
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Enable the arrow operator for all types that could have members, so
that one can e.g. write myStringProperty->size() instead of having to
use the less convenient myStringProperty.value().size().
Also cleaned up the rvalue ref overloads to be
disabled for basic types. For those we now also
return by value, for more complex types we
return a const reference.
Change-Id: If6a75898dc0a097f57052488f0af0cd7166b3393
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
If one needed to listen to a signal just once, one had to
store the QMetaObject::Connection object returned by connect()
and use it to disconnect the slot after the first signal
activation.
This has led to a proliferation of using wrappers (and enough
TMP); they usually look like this:
1) create a shared_ptr<QMO::Connection>, allocating its payload;
2) create a lambda, capturing the shared_ptr by value;
3) in the lambda, disconnect the connection (through the shared_ptr),
and call the actual slot;
4) connect the signal to the lambda, storing the returned
QMO::Connection into the shared_ptr.
This is expensive, error prone for newcomers, and tricky to
support as a general facility inside one's projects.
We can do better, just support single shot connections right
in QObject.
[ChangeLog][QtCore][QObject] Added the Qt::SingleShotConnection
flag. When a connection is established with this flag set,
the slot is going to be activated at most once; when the signal
is emitted, the connection gets automatically broken by Qt.
Change-Id: I5f5feeae7f76c9c3d6323d841efba81c8f98ce7e
Fixes: QTBUG-44219
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
One could guess it by assuming that disconnecting for a destroyed
receiver and disconnect() with given receiver use the same
implementation, but without closely knowing the implementation a
reader of the documentation can't know for sure.
Also add a test to prove that what the new documentation says is
really true.
Also remove an unnecessary negation in the preceding sentence.
Change-Id: I9d24442bb1a4646b89f969bad1a4d0e1eafa7534
Reviewed-by: Mitch Curtis <mitch.curtis@qt.io>
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
This is in line with QMetaType and will be used to implement a mutable
QSequentialIterable. Later on, a QMetaAssociation will be added as
well, to implement a mutable QAssociativeIterable.
The code here represents the minimal set of functionality needed to have
a practical sequential container. The functionality is not completely
orthogonal. In particular, the index based operations could be
implemented in terms of iterator-based operations.
Task-number: QTBUG-81716
Change-Id: Ibd41eb7db248a774673c701549d9a03cbf2e48b6
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
When a signal/slot connection is activated, a lock on the receiver
object is taken (to be sure it doesn't get destroyed).
The path for blocking queued connections however took the lock on
the sender by accident, fix that.
Pick-to: 5.15 5.12
Change-Id: I8cd938a50eca2bf71e7bfb86768ee0c8431afdfa
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
We need to add these two classes at the same time, because
QAnyStringView makes all QUtf8StringView relational operators moot. We
might want to add some later, esp. for UTF-8/UTf-8 comparisons, to
avoid the pessimization that we can't early-out on size() mismatch in
QAnyStringView equality operators, but that's an optimization, not a
correctness issue, and can be fixed in a source-compatible way even
after Qt 6 is released.
To deal with the char8_t problem in C++20, make QUtf8StringView a
class template out of which two UTF-8 views can be instantiated: the
Qt 7 version, which depends on C++20 char8_t as value_type, and the Qt
6 version where value_type is a char. Use inline namespaces to map the
QUtf8StringView identifier to one or the other, depending on the C++
version used to compile the user code. The inline namespace names must
needs be a bit ugly, as their inline'ness depends on __cpp_char8_t. If
we simply used q_v1/q_v2 we'd be blocking these names for Qt inline
namespaces forever, because it's likely that inline'ness of other
users of inline namespaces in Qt depends on things other than
__cpp_char8_t. While inline'ness of namespaces is, theoretically
speaking, a compile-time-only property, at least Clang warns about
mixed use of inline on a given namespace, so we need to bite the
bullet here. This is also the reason for the QT_BEGIN_..._NAMESPACE
macros: GCC is ok with the first declaration making a namespace
inline, while Clang warns upon re-opening an inline namespace as a
non-inline one.
[ChangeLog][QtCore][QUtf8StringView] New class.
[ChangeLog][QtCore][QAnyStringView] New class.
Change-Id: Ia7179760fca0e0b67d52f5accb0a62e389b17913
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
High-DPI pixmaps are always enabled, and cannot be disabled.
Change-Id: I01a006b404e5431582b64ef812974c1c022b39ae
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Since 5.0: assignment/construction from QObject pointer
Since 5.14: data() to recover the packaged pointer
Change-Id: I5d6ab561ce39bc0d9d3e5035eb2ca38139cd76b6
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The Unicode table code can only be safely called on valid code-points.
So code that calls it must only pass it valid Unicode data. The string
iterator's Unchecked Unchecked methods only provide this guarantee
when the string being iterated is guaranteed to be valid UTF-16; while
client code should only use QString, QStringView and friends on valid
UTF-16 data, we have no way to be sure they have respected that.
So take the few extra cycles to actually check validity in the course
of iterating strings, when the resulting code-points are to be passed
to the Unicode table look-ups. Add tests that case mapping doesn't
access Unicode tables out of range (it'll trigger the new assertion).
Added some comments to qchar.h that helped me understand surrogates.
Change-Id: Iec2c3106bf1a875bdaa1d622f6cf94d7007e281e
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Port the high-dpi icon test to use the new pixmap()
API which takes the target devicePixelRatio as an argument.
This means we can run the fromTheme test at dpr = {1,2,3},
instead of at the current global devicePixelRatio only.
Task-number: QTBUG-85885
Change-Id: Iec7b21e04ed760e48964307d2048eaec1976ffe2
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Add Q_REQUIRED_RESULT to force callers to check the return; the
QTranslator object is unusable if load() fails.
Check the result in QTranslator's own test.
Task-number: QTBUG-85700
Change-Id: I07509c76470cc87626190670665cd3162bfb17e7
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
A large slice of it has been deprecated since 5.2.
Reflowed a doc paragraph pointed out, in the deprecation commit, as
having been left ragged by its edits.
Note: qSwap() is documented as \deprecated but not marked, where it's
defined, as deprecated.
Change-Id: Iaff10ac0c4c38e5b85f10eca4eedeab861f09959
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This attribute is now on by default.
Change-Id: I7c9d2e3445d204d3450758673048d514bc9c850c
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
The functionality should be available via QWidget::winId(),
and QWidget::createWindowContainer() + QWindow::fromWinId().
Any bugs in this area should be fixed by improving the general
wrapping APIs.
Fixes: QTBUG-83254
Change-Id: I86584a4a8138d17d65a50da39efd58039f10da91
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
Without an override for logicalDpi() the base class implementation
would use the geometry of the screen to figure out the DPI, and end
up with ~100, which combined with a 96DPI base logical DPI would
give a wrong scale factor.
Change-Id: I68aecce44d2ee672c7b707dfe5444af8f551e961
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
warning: all paths through this function will call itself [-Winfinite-recursion]
Change-Id: I70008710a7448ffd532bb6c3acb488fe439cb807
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
It was left out of the build in f08038fca7
but not removed. It doesn't make sense to keep it, as the classes it
tests have been removed.
Change-Id: Ic36a066eb092d8b9b40c645de649af9dcaab4ff2
Reviewed-by: Laszlo Agocs <laszlo.agocs@qt.io>
PCRE2 already uses size_t which we can now make full use of.
Change-Id: Icb5efd5c6ef27f2e31a9780bf62f5671ddc603cd
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
As there is now a chance that a QMetaMethod already contains the
metatypes for its arguments, we can just query it directly (and use the
fallback to name lookup logic that already exists there).
This also allows us to avoid creating a QList of names, and only
requires us to do a name lookup in case the connection actually fails.
Change-Id: Idda30bc4b538a94476ae6c533776c22340f0030d
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
While we initially wanted to just disable the functionality the way it
is currently built forces people to do a full build of Qt just to enable
it. Instead of doing this half-measure let's just remove the code
completely from QtBase and rather prepare QtNetwork to handle being a
plugin that can be compiled at any time.
Task-number: QTBUG-80340
Change-Id: I19155c8c167cf932088f01b2a9706d0e7ab792d1
Reviewed-by: Alex Blasche <alexander.blasche@qt.io>
I don't think we want to have implicit conversion
from a const char * or a QByteArray to a QUuid.
Change-Id: Idfe7450ce15b89e295aa7af7ccf1fc94f5acd4f9
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Restored previously deleted logic of setting GrowsBackwards flag for
prepend-like cases. This should be sufficient to fully enable prepend
optimization
Fixed QList::emplace to not use implementation detail logic. Updated
tests to cover changed behavior and its correctness
Task-number: QTBUG-84320
Change-Id: I4aadab0647fe436140b7bb5cf71309f6887e36ab
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
Introduced prepend optimization logic to QCommonArrayOps.
Trying to rely on original QList behavior
Task-number: QTBUG-84320
Change-Id: I46e6797b4edad804a3e3edb58307c9e96990fe01
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
Introduced allocation function in QArrayDataPointer with
interface similar to QArrayData::allocate that supports growing
strategies. This func is used instead of the original in cases
when prepend-aware storage is needed. Tried to follow Qt5 QList
policy in terms of space reservation
Updated QPodArrayOps::reallocate to be aware of growing
shenanigans. It doesn't look like a perfect solution but it is
rather close and similar to what Qt6 QList is doing when not
growing (e.g. reserve/squeeze)
Added initial QCommonArrayOps with helper function that tells
when reallocation is preferable over just using the insert-like
operation. This comes up later on when GrowsBackwards policy is
properly supported in operations
Essentially, 2/3 main data management blocks for prepend optimization
are introduced here. The last one being a generalized data move that
is done instead of reallocation when existing free space is not enough
Task-number: QTBUG-84320
Change-Id: I9a2bac62ad600613a6d7c5348325e0e54aadb73d
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
A comment is not good enough, Some of the enum members were even still
in use, or mentioned in documentation.
WA_ContentsPropagated, WA_WState_DND and WA_ForceAcceptDrops have been
deprecated since 4.5.1; and at least the last has been an \omitvalue
in the docs for even longer. (WA_ShowModal and WA_GroupLeader have
been similarly marked, but are in use, see QTBUG-85816.)
Push back to 5.15.1 in order to be able to remove these at Qt 6.
Pick-to: 5.15.1
Change-Id: I6ea3839767e5f5158b0fed508f65798470191908
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Add new QIcon::pixmap() overload:
QPixmap pixmap(const QSize &size, qreal devicePixelRatio, ...)
This function replaces the existing pixmap() function which
take a QWindow pointer, and should be more convenient in use.
Task-number: QTBUG-85885
Change-Id: Ie4ca96a266d9278864678dc61bdfc2836cabdb93
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
viewOptions returned a QStyleOptionViewItem object. Such a method
can never support newer versions of the option structure.
Most styleable QWidget classes provide a virtual method
initStyleOption that initializes the option object passed in as a
pointer, e.g QFrame, QAbstractSpinBox, or QComboBox.
Follow that API convention, but name it initViewItemOption, as the
QStyleOptionViewItem struct contains information about the item as
well as the widget itelf.
This is a source incompatible change that will go unnoticed unless
existing subclasses mark their overrides as 'override', or call
the removed QAbstractItemView::viewOption virtual function.
[ChangeLog][QtWidgets][QAbstractItemView] The virtual viewOptions
method that previously returned a QStyleOptionViewItem object has
been renamed to initViewItemOption, and initializes a
QStyleOptionViewItem object that's passed in through a pointer.
Change-Id: Ie058702aed42d77274fa3c4abb43ba302e57e348
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
