The plan for QObject::connect() (perfect) forwarders, such as
QWidget::addAction(), was to just use a variant of the Detection Idiom
to see whether QObject::connect() with the arguments as given would
compile and SFINAE out the forwarder otherwise.
It turns out that the "functor" overload of QObject::connect(), in
particular, is severly underconstrained and accepts e.g. QKeySequence
as a function object, only erroring out via a static_assert() in the
body of the function, and thus at instantiation time and not, as
needed, at overload resolution time.
At the same time, we don't really want QObject::connect() to SFINAE
out on argument mismatches between signal and slot, because the
resulting error messages would be ... unkind to users of the API. We
would like to keep the static_assert()s for easier error reporting.
Reconciling these two contradicting requirements has so far eluded
this author, so for now, to unblock progress, we explicitly black-
and, in one case, white-list possible arguments. Because QKeySequence,
in particular, is implicitly constructible from int(!), and therefore
any enum type(!), incl. Qt::ConnectionType, we need to do way too much
coding in the addAction() constraints. Hopefully, we'll be able to fix
the issue at the root cause, in QObject, before Qt 6.3 is out, but
until then, this is an ok-ish stop-gap measure.
Add thorough overload set checks (positive ones only, for now) to
tst_qwidget and tst_qmenu.
Change-Id: Ia05233df818bc82ecc924fc44c1b349af41cbbf1
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
All other equality comparisons except floating-point are value
comparisons, not interpretation comparisons.
[ChangeLog][QColor] Equality on HSL colors is now raw value
based like other color-formats, instead of being partially
interpretation based.
Pick-to: 6.2
Change-Id: I391464da172af6a2d69178b94203d36f7ce2c07d
Reviewed-by: Eirik Aavitsland <eirik.aavitsland@qt.io>
Disabled (moved into the legacy provider) DES-CBC results in a crash,
when setting key length.
Pick-to: 6.2 6.1 5.15
Change-Id: Ie0b49424f11d8042ebecebfd3b6346263f730551
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
All the getters of QFutureWatcher are consistent with the getters of the
corresponding QFuture, except for the isFinished() method, which returns
'true' only after the finished() signal is delivered. This behavior
might be unintuitive for the users. In particular, isFinished() returns
'false', even if it's called immediately after waitForFinished().
[ChangeLog][QtCore][QFutureWatcher][Important Behavior Changes] The
QFutureWatcher::isFinished() method now indicates if the related
QFuture is finished, instead of indicating if the finished() signal was
delivered. This makes it consistent with the future that is being
watched.
Fixes: QTBUG-91048
Pick-to: 6.1 6.2
Change-Id: I6ae9b882b23e06198a82c95b026491bd480b3bf0
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
The fontFamilies and fontStyleName have an asymmetric getter and setter.
The setter takes QStringList, but the getter returns QVariant (?), for
no reason whatsoever (the underlying QFont APIs take QStringList,
and the docs say that they should return QStringList).
It's not entirely obvious that we can fix this without introducing
badly-named getters (getX) or somesuch, so for the time being, mark
this to be fixed in Qt 7.
Change-Id: I49706af00906a2c5d950ada4be3cbc76dbee432c
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: David Faure <david.faure@kdab.com>
One was long redundant. The other two, relatively recently added,
should be redundant thanks to the former enum member MSECS_PER_DAY
being changed to a constexpr qint64 by
commit 4059af81d3
Change-Id: I601ed8dae905fed6399ce29c25545ec651e9838a
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Eirik Aavitsland <eirik.aavitsland@qt.io>
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
QAbstractButton::setChecked is not virtual, so QToolButton cannot
override to synchronize the default action's checked state. This resulted
in button and default action not being in sync when the checked state
of the button was changed programmatically, while changing the checked
state on the action kept the button in sync.
Connect to the button's own toggled signal instead to keep the state of
the default action in sync. Make it a unique connection to allow multiple
calls to setDefaultAction, which are used by QToolButton to keep the
button updated if properties of the default action change.
Add a test that confirms that button and action are synchronized both
ways, and that we only get single signal emissions when changing either
programmatically.
Fixes: QTBUG-95255
Pick-to: 6.2 6.1
Change-Id: I0e027faf1da763ef1878e46e85bfa70073c8bf82
Reviewed-by: Mitch Curtis <mitch.curtis@qt.io>
CMake 3.21.0 introduced a regression where per-config sources can't be
specified for object libs when building Qt with Ninja Multi-Config
and cross-config mode on.
We hit that in our Windows resource compiler handling.
To work around the issue, pass the config-specific sources directly to
the target rather than via an object library, when using CMake 3.20 or
later. The original issue of not being able to do that has been fixed
in 3.20.
Amends ba6175eb73
Amends a1ccedeb44
Amends 657fa0462d
Pick-to: 6.2
Fixes: QTBUG-95229
Change-Id: Idea6d5bcc54b3124c66c45538c2e06e92f288f5f
Reviewed-by: Cristian Adam <cristian.adam@qt.io>
No Qt API throws exceptions, ever. Kill the throwing paths, leaving the
normal paths in which a permission request is denied.
Pick-to: 6.2
Change-Id: I6bebaf155c5232444bf5014e25e9c469d5dccfa6
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
The correct variable to check is WIN32 rather than WINDOWS.
This affects which plugins get automatically linked in a static
Qt build.
Amends a3b58a7844
Amends d385158d52
Pick-to: 6.1 6.2
Fixes: QTBUG-95283
Change-Id: Idf78c78a1029f4d13fb460c07bef3d2669e55b09
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Explicitly specify that calling this method for an empty set or with
an invalid iterator results in undefined behavior.
On a debug build an assert is triggered in such case, but on a release
build it will access the incorect index of an array.
Task-number: QTBUG-91736
Pick-to: 6.2 6.1
Change-Id: Ibc3e91512a0ad9d9779a41083fedb8a91780380b
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
While QFileSelector is documented to work on files, select accepts
arbitrary URLs. Moreover, the QML engine can end up intercepting
arbitrary (user provided) URLs, including to directories.
Prior to this change, passing "file:///" or "/" to the function would
break: We would temporarily get a "//" path, which is invalid, and thus
we would return an invalid in the end.
Prevent this by only appending a slash to the path when it doesn't have
one.
Fixes: QTBUG-85410
Pick-to: 6.2
Change-Id: I1d2807a9d225df611c3a5e871e3c1d90a6a25953
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Fixes static analyzer warning 12b19393e18b2394a398806f633c6eee, and
amends a1a6e3d21b.
In the process, replace the "int& *= double" with correct integer
arithmetic that'll produce the intended result without going via
double.
Done-with: Edward Welbourne <edward.welbourne@qt.io>
Pick-to: 6.2
Task-number: QTBUG-8096
Change-Id: Ib2aa8ae46a1bfd4d121e61cf99141c0311502215
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
QtConcurrent reduce functions were requiring to explicitly pass the
the result type when passing functors as reductor. This was because of
inability to deduce the result type from the functors. The result type
of the QtConcurrent reduce functions should match with the type of the
fist argument of the operator() when a functor is passed. Reused the
ArgResolver type trait (already used for QFuture and QtConcurrent::run)
to find out the result type in that case.
Task-number: QTBUG-88448
Change-Id: Ief0eeee197df8cb9c30f3403d71978f36e4fb0f2
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
layoutAboutToBeChanged must be called before
persistentIndexList as the user might create persistent indexes
as a response to the signal
Fixes: QTBUG-93466
Pick-to: 6.2 5.15
Change-Id: I73c24501f536ef9b6092c3374821497f0a8f0de4
Reviewed-by: David Faure <david.faure@kdab.com>
These signals allow monitoring where in the HTTP1/HTTP2
flow a request is currently in.
Fixes: QTBUG-71698
Fixes: QTBUG-18766
Change-Id: Icc2fe435afc9f680fa7a76c32731e25fcdfeb4b4
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
The Q_LOGGING_CATEGORY macro defines a function, and for each backend
the function would be defined (and defined differently).
Pick-to: 6.2
Change-Id: I6522b5e3baf731e5ead0e21da009c970d580fa02
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
The buffer might be large than needed and contain a lot of uninitialized
pixels. Scaling it down to the requested size is wrong, we need to scale
it by the correction factor.
Amends e56b8e1e59.
As a drive-by, make the correction factor helper function static.
Fixes: QTBUG-94733
Pick-to: 6.2 6.1 5.15
Change-Id: Ia5be2a77459321e30485d330b49cf57fdbb664d2
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
If the file already existed we simply removed the old one without
adjusting the size. So use the removeFile() function which takes care of
that.
Additionally, if the current size was non-null we previously increased
the size (presumably meant to be temporarily but wasn't) and called
expire() which would either:
1. not do anything and return currentCacheSize, if it was not greater
than the max size. This would mean that the size of the file would be
counted twice.
or,
2. discard currentCacheSize, measure the size of the items, and then
remove some items if the total size surpassed the max cache size
Neither of those branches need us to (temporarily) increase
currentCacheSize. It also doesn't attain the (presumed) goal of trying
to keep below the max cache size after having added the new item.
Fixes: QTBUG-95009
Pick-to: 6.2 6.1 5.15
Change-Id: I2b5b13ff473a7aa8169cf2aecfea783c97f2d09a
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Custom tab bar's close button looks outdated in Fusion style.
Common style's close button looks much better.
Task-number: QTBUG-94752
Change-Id: I9e9aba4a5cd1ee0d3b7fc230306c5e052b6ba3bd
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Reviewed-by: David Faure <david.faure@kdab.com>
Spotted during API review for 6.2
Pick-to: 6.2
Change-Id: Ib31020848ec6ffabc126c6984cc82547266d4a1d
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Move the static tabBarCloseButtonIcon helper into the standardIcon()
implementation, simplified by using QIcon::addFile and the
iconResourcePrefix helper.
Add corresponding support in standardPixmap().
Use standardIcon() in PE_IndicatorTabClose where tabBarCloseButtonIcon()
was previously used to get the icon, leave the caching in place.
Fixes: QTBUG-95264
Change-Id: I89c161734553db762caa29f8ab738e3c6e64ea0a
Reviewed-by: David Faure <david.faure@kdab.com>
The insert() overloads that took a const_iterator started by calling
std::distance(begin(), pos) - which has a cost linear in how far pos
is from begin() - in order to, after detach()ing, obtain an iterator
at the same offset from the new begin(), using std::next() - also
linear. This leads to quadratic behavior when large numbers of entries
are added with constEnd() as the hint, which happened to be tested by
tst_bench_qmap. That wasn't running, due to some assertion failures,
but once those were fixed the hinted tests timed out after five
minutes, where their unhinted peers completed comfortably within a
second.
Check whether detach() is even needed and bypass the std::distance() /
std::next() linear delay when it isn't. This brings the hinted tests
down to running faster than their unhinted equivalents.
Pick-to: 6.1 6.2
Task-number: QTBUG-91713
Change-Id: I6b705bf8fc34e67aed2ac4b3312a836e105ca2f2
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Declaration::lengthValue only supported 'px' sizes, but one can transform
any 'pt' value into 'px' by multiplying with 1.33.
Notes: this ignores display DPI, and instead follows the W3C definition
of 'pt' and 'px' as absolute lengths [1].
[1] https://www.w3.org/TR/css3-values/#absolute-lengths
1pt = 1/72th of 1 inch
1px = 1/96th of 1 inch
so the conversion is px = pt * (72/96).
Add unit test that verifies this using QPushButton's icon-sizes property,
also with changed font in preparation of adding support for 'em' and 'ex'
units in a follow up commit.
Task-number: QTBUG-8096
Pick-to: 6.2
Done-with: Cristian Maureira-Fredes <Cristian.Maureira-Fredes@qt.io>
Change-Id: I58782e7ad0e2ff9d89ed695f8a23b1e584cfed64
Reviewed-by: Andy Shaw <andy.shaw@qt.io>
The line edit receiving focus might require different input method
capabilities (e.g. be a password edit), so the input method needs to
be refreshed when focus is received. This implicitly happens on tab
focus when the text is selected or the cursor is moved within an
input mask, but neither of those might happen, and for click-focus
it never happened.
Fixes: QTBUG-86846
Pick-to: 5.15 6.1 6.2
Done-with: Volker Hilsheimer <volker.hilsheimer@qt.io>
Change-Id: I47c5275d3a9d4d190ebce64269a345431346c17c
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Qt has a hardcoded backlog value of 50,
this allows for applications to tune this
value. Modern kernels have the SYN cookie
feature that reduces pressure from an
flood attack, the backlog setting however
is then a queue for most likely real
completed (SYN/ACK) connections hence, it's
easy to get clients connections dropped
with this very small limit.
[ChangeLog][QtNetwork][QTcpServer] Added
QTcpServer::setListenBacklog() to be able
to have control over the listen backlog feature.
Change-Id: I1c78af6d99e012591e214b7e09fa85c485880d48
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Take two. f20edffcc8 only disabled the
backtrace preamble.
Pick-to: 6.2
Change-Id: I972d007af59d13ec6eb11da44fc2b5e25c36e46e
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
Remove read/write variants of the permission types to make the
API simpler and more versatile. If the user wishes to have more
control over the permission requests/checks, they can use more
platform-specific code.
Pick-to: 6.2
Task-number: QTBUG-94407
Change-Id: I2b72041aa3effaac7e7f7361237cf1146817b525
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
As Giuseppe pointed out in API change review, an opaque type should be
used.
Task-number: QTBUG-94407
Pick-to: 6.2
Change-Id: I862a6f52d284317e1243fd91f45bb0af130d154a
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
The qml files should now always be added via dedicated functions
provided by the qtdeclarative repo. Those functions take care of adding
the files to targets so that they will show up in IDE file lists, so we
should not add them to a separate target as well. This will also remove
one unnecessary build target, which will reduce visual noise in some
IDEs.
Pick-to: 6.2
Change-Id: I79bf1ea8fb8a2f15643e052d50ac582a91bdff1a
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
... instead of QSharedPointer.
QSharedPointer performs twice as many atomic operations per pointer
copy as std::shared_ptr, and this is private API, we're not bound by
BC constraints, so we can port to the more efficient version.
Change-Id: I2e2a02493565a7ca51c86ec0ed66b6ce7c763e41
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
The use of a freestanding function is not needed now that the name
doesn't alias the nativeInterface accessor function, and was just
adding complexity to the machinery.
People not familiar with the code will have an easier time following
the flow through the helper member function, and we no longer need
to declare our own export macros.
Pick-to: 6.2
Change-Id: I17530b7e89939cfc19ab8ffaa076b7129ae02dcf
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
In preparation of porting to shared_ptr.
Add a few strategic moves and remove a rather pointless comment
stating the obvious.
Change-Id: Ie1c0f3431af79bdb204e00d99323bf9f2d100d0d
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Projects sometimes set both CMAKE_CONFIGURATION_TYPES and
CMAKE_BUILD_TYPE regardless of what CMake generator they are using.
They shouldn't, but it is common enough that we should protect against
it. The GENERATOR_IS_MULTI_CONFIG global property is the only robust
way to know which type of generator we are using, so switch to that for
how we detect this.
Pick-to: 6.2
Change-Id: If5d6568da0d2adb3275bbaf9a7d3bd3e033dc636
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
Remove copying of json files in non-prefix builds.
Previously, this was done to mimic the directory layout of prefix
builds in a non-prefix build, but the extra complexity is not worth
it.
Keeping the files in the build directories of each repo in a
non-prefix build should be good enough.
As a result, we don't need to distinguish which file path should be
passed in INTERFACE_SOURCES. Now we have only 2 paths to consider, the
BUILD_INTERFACE one and the INSTALL_PREFIX one, rather than 3
different paths.
Move the code that handles installation closer together.
Rename the internal COPY_OVER_INSTALL option to
___QT_INTERNAL_NO_INSTALL.
We might want to expose it as a public option later, but it depends on
whether we will disable default installation for user projects.
Remove assignment of some properties that are not used anymore.
Amends c431e2d33c
Pick-to: 6.2
Task-number: QTBUG-94942
Change-Id: I3a26d1988987b2ce7c7d01d2311c13b3da3175a8
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
Change the copying of
${CMAKE_CURRENT_BINARY_DIR}/${target}_metatypes.json
to
${prefix}/lib/metatypes/${target}_metatypes.json
to happen as part of the command generating the file, rather
than copying the file as part of a separate custom target or
POST_BUILD event.
This ensures that the custom commands in qt6_qml_type_registration
that use those files as dependencies will cause them to be
generated before the qml type registration happens, thus eliminating
errors like
Error 5 while parsing
qtbase/lib/metatypes/qt6quick_metatypes.json: illegal value
Pick-to: 6.1 6.2
Fixes: QTBUG-94942
Change-Id: Idddd73786d1a622984965c60ac9b4c3bc2c13ab5
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
For SingleSelection, removing the selected item will select the nearest
item and, if autoScroll is enabled, ensures that the newly selected
item is visible in the viewport. This may result in scrolling.
For Multi- or ExtendedSelection, this should not happen, as having no
selection is perfectly fine in those modes.
However, QListView still tried to scroll to the current item in response
to the currentIndexChanged signal. Since the currentIndex is at this
point already hidden, the rectangle for it became invalid, and the
attempt to scroll resulted in a one-pixel up-movement of the viewport
(since the invalid rectangle has width == height == -1).
Fix this by not scrolling if the rect for the index is invalid. Note that
the index is still valid at this point, so we can't shortcut the call
stack earlier. Add test that exercises the different combinations of
ViewMode and SelectionMode, and demonstrates the one-pixel
movement without the fix.
Fixes: QTBUG-94788
Pick-to: 6.2 6.1 5.15
Change-Id: I1f36973eadb46e8c9b8b8068bc76ee09e9f490dd
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
As pointed out by Marc Mutz in another review, the Android branches of
its #if-ery amounted to a complicated no-op, so simplify the #if-ery,
add a TODO and then simplify the code thereby freed of the need to
accommodate the #if-ery.
In the process, initialize a set of filenames with the two filenames
that we read certificates from after looping over the set, which might
have left those files being read twice.
Change-Id: I2ee4ee3c3cf40226ee6a50afd6127fa4a71d2834
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
The switch condition expression has already dealt width all the
conditions, so a `default` label is not necessary (it introduces a
warning). Delete it.
Fixes: QTBUG-95223
Change-Id: Iac3d9bb5d0b6fd8372d4dc16d920ce0a7b0511db
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Replace a manually-managed pointer + size with QVarLengthArray,
allowing use of pre-allocated space.
Change-Id: Ife3abea0b3b0f2577bb76c348513d52d46f79641
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Make qtpaths also available under qtpaths6. In fef850c51a, qtpaths
got extended to replicate the 'qmake -query' behavior. This was meant for
IDE's like Qt Creator, so that they can use qtpaths instead of qmake to
identify Qt versions.
Anyhow, since qtpaths was also available in Qt 5, letting users just select
_any_ qtpaths executable won't work for this purpose. Adding qtpaths6 - like
qmake 6 - allows IDE's to query for a Qt 6 qtpaths.
Pick-to: 6.2
Task-number: QTCREATORBUG-25546
Change-Id: Ifc7e1a76b26cf706872db1bb1f9d3de5f006de24
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
Use unique_ptr to indicate ownership.
The QPainter dtor is now empty, = default it.
As a drive-by, fix a non-sensical if (p) p = nullptr;
Change-Id: I5c811eb7de6abf8a9f279cf3630d5f86db6b3864
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
We keep QVector2D storage Qt 6 BC (to avoid making QNativeGestureEvent
larger), but in Qt 7 we should return exactly the same value as given
(for what it's worth, in spite of this being overkill for panning a
reasonable distance).
Change-Id: Iecbd4c9b60ad9ae5e0466c7027b038ddb85b8c8b
Pick-to: 6.2
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
In QItemSelectionModel, items which are disabled or marked as not
selectable should not be considered as selected. But this was
not handled consistently.
The following methods considered only items which are enabled and
marked selectable: selectedIndexes(), rowIntersectsSelection(), and
columnIntersectsSelection(). The following methods considered only
items which are marked selectable, but did not check whether they
are enabled: selectedRows(), selectedColumns(), isRowSelected(),
isColumnSelected(), isSelected(). Finally there is hasSelection(),
which did not check for enabled nor for selectable.
This patch introduces consistent behavior. All methods check
both whether the items are enabled and whether they are selectable now.
[ChangeLog][QtCore][QItemSelectionModel][Important Behavior Changes]
All methods in QItemSelectionModel now consider only items which
are marked as enabled and selectable as part of the selection.
Fixes: QTBUG-93829
Pick-to: 6.2
Change-Id: I4725243ea6b0db4f289ce34ada22c7a9d3282713
Reviewed-by: David Faure <david.faure@kdab.com>
The members view, par, and itemFlags were initialized in each
of its nine constructors. This patch moves the initilaization to
the declaration to prevent this repetition.
Change-Id: I71d136e7615dca5d87e8a4dc161c877633017ad0
Reviewed-by: David Faure <david.faure@kdab.com>
The maximum size for a hash result is 64 atm. Even if, and esp when,
we'll get to 128 and 256 bytes in the future, there's no reason to use
dynamic memory, because the sizes will always be statically known.
So use, essentially, a std::array<char, 64> to hold the result
internally. Add a bit of convenience API on top to limit impact on the
rest of the code and add a few static_asserts that ensure this is large
enough. Then give users access to the internal buffer by adding
QByteArrayView resultView() const noexcept. The documentation snippet
is taken from QString::data(), suitably adjusted.
Use resultView() in a few places instead of result().
[ChangeLog][QtCore][QCryptographicHash] Changed to use a
statically-sized buffer internally. Added resultView() to access it.
Change-Id: I96c35e55acacbe94529446d720c18325273ffd2f
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
By switching out the static_assert for an enable_if we end up producing
a clearer error, at the call site:
/qt/qtbase/examples/gui/rasterwindow/main.cpp:69:9: error: no matching member
function for call to 'nativeInterface'
app.nativeInterface<QNativeInterface::QCocoaGLContext>();
~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/qt/qtbase/src/gui/kernel/qguiapplication.h:176:5: note:
candidate template ignored: requirement
'NativeInterface<QNativeInterface::QCocoaGLContext>::isCompatibleWith<QGuiApplication>'
was not satisfied [with NativeInterface = QNativeInterface::QCocoaGLContext, TypeInfo =
QNativeInterface::Private::NativeInterface<QNativeInterface::QCocoaGLContext>, BaseType =
QGuiApplication]
QT_DECLARE_NATIVE_INTERFACE_ACCESSOR(QGuiApplication)
^
By using SFINAE for the TypeInfo we can also ensure that it works for
types that are not native interfaces, such as if the user tries to
call nativeInterface<QString>().
Since we can no longer use decltype(*this) to resolve the base type
we need to change QT_DECLARE_NATIVE_INTERFACE_ACCESSOR to take the
type as an argument, as we do for other QT_DECLARE_FOO macros.
Pick-to: 6.2
Change-Id: Ie3f7e01ab7c3eb3dcc2ef730834f268bb9e81e0c
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
QMetaType now implicitly knows how to convert any QFuture<T> to
QFuture<void> without needing to manually register a converter
function.
QtWebChannel will make use of this to transparently support QFuture<T>
return types.
[ChangeLog][QtCore][QMetaType] QMetaType now supports converting any
QFuture<T> to QFuture<void>.
Task-number: QTBUG-92903
Change-Id: Ied7e71be37c346cc3d2c274ffb0d91a6821ab4d4
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
This replaces QLocale::system().collation(), which left client code
with no way to change this default. Since QLocale's default is the
system locale initially, the old behavior is retained until the first
call to QLocale;:setDefault(). Addition of a separate collation locale
in 5.14 changed from using the default locale to using the system's
collation locale; the present change restores the ability to override
this.
[ChangeLog][QtCore][QCollator] The default locale used by QCollator is
now the collation locale of the default QLocale. This restores the
ability (lost at 5.14) to control the locale used by
QString::localeAwareCompare(), while retaining the use of a collation
locale when the default is the system locale.
Task-number: QTBUG-95050
Pick-to: 6.2 6.1 5.15
Change-Id: I232772bb8f23fffd74573174de70ef2deae372a9
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
It now uses QtPrivate::condition_variable, it pulls in the correct
header for that, so #include <condition_variable> is no longer needed.
Separate opening braces of function bodies onto next line.
Change-Id: I08f721c4d52756932bb9409e34e51dcbb3eda104
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
In case of
QTranslator translator;
translator.load("somedir/file.qm");
and file.qm being a meta catalog file, the sub-catalogs in somedir
couldn't be located, unless "somedir" was set as second argument.
Pick-to: 5.15 6.1 6.2
Fixes: QTBUG-95013
Change-Id: I06103244ce2ff9800c2c64cb0c17f9bc7ef0e8de
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Pick-to: 6.2
Task-number: QTBUG-94407
Change-Id: Ie9c05dbe498cd372c015b5125e6cb8d59ca96b59
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
The ordering function used to sort the locale data generated for
QLocale attempted to sort the default territory for a given language
and script before other territories, but was too tangled for it to be
obvious this is what it was doing. The result turned out to be
non-transitive. Replace with code that implements the same preference
but only applies it where the result is compatible with transitivity.
This leads to a shuffling of the order of the Serbian-language
locales, which sorts the Cyrillic ones before the Latin ones. This is
consistent with my reading of the CLDR data, which fills in Cyrillic
and Serbia for Serbian; Serbian/Cyrillic/Serbia did previously sort
before all other Serbian variants.
Thanks to Ievgenii Meshcheriakov <ievgenii.meshcheriakov@qt.io> for
discovering the non-transitivity.
Pick-to: 6.2
Change-Id: I0ce9f78e620e714f980f32b85b7100ed0f92ad74
Reviewed-by: Ievgenii Meshcheriakov <ievgenii.meshcheriakov@qt.io>
Reviewed-by: Cristian Maureira-Fredes <cristian.maureira-fredes@qt.io>
... instead of QSharedPointer.
QSharedPointer performs twice as many atomic operations per pointer
copy as std::shared_ptr, and this is private API, we're not bound by
BC constraints, so we can port to the more efficient version.
Change-Id: I9572a8321aae381e5dbe4a51119f2c9494a8fbc7
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Both readChannelFinished() and disconnected() signals should be emitted
after closing the pipe. Otherwise, these signals do not correspond to
the state of the socket and may even be resent, if a slot connected to
one of these signals processes events.
[ChangeLog][QtNetwork][Important Behavior Changes] QLocalSocket on
Windows now emits both readChannelFinished() and disconnected() signals
after closing the pipe and emitting stateChanged(UnconnectedState),
which is consistent with the behavior on Unix.
Pick-to: 6.2
Change-Id: I1cc551b7897fdba3cec1fd6705f5396790818c7d
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
They will only expand what's currently in the model.
As a drive-by, remove superfluous : after the \warning macro.
Fixes: QTBUG-94981
Pick-to: 6.2 6.1 5.15
Change-Id: I85d38373fb7edf5a5407622b32870e7bcd5d9aeb
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
Use unique_ptr to indicate ownership, even though it's conditional (on
QPaintEngine::autoDestruct()). That just requires a custom deleter.
Change-Id: Icf8e356c333f9617b2e5172b14f13197e63c9502
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
Android 12 introduced a notification which is shown to the user each
time the app accesses the clipboard via getPrimaryClip.
Currently this notification is triggered, even if we just want to check,
if some clipboard data exists.
So lets not get the actual data and instead use getPrimaryClipDescription
to check for the existence of the correct mime type in the clipboard.
Pick-to: 6.2 6.1 5.15 5.12
Change-Id: I4800f5545ab46b7f6cade0ce9d78c04b50ae96cf
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
With a stylesheet that defines a rule for checked:selected menu items,
the rule was never rendered. Instead, the checked rule was used.
The 'act' variable was initialized based on State_Selected being set, but
that state flags was never set on the temporary QStyleOptionMenuItem
used to render the PE_IndicatorMenuCheckMark.
Note: QStyleSheetStyle falls back to QWindowsStyle for this draw path.
Task-number: QTBUG-73966
Pick-to: 6.2
Change-Id: I3044dbe8f60f1896f79af1afdb26094971704616
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
In QPanGesture this is called delta().
OTOH we have QWheelEvent::pixelDeltas().
Delta is a vector, and there's only one (with two components).
Native gestures hold incremental values: e.g. the pinch gesture event
provides an incremental amount of either zooming or rotation (so most
events have QNativeGestureEvent::value() very close to 0).
It's the same with the pan gesture's delta().
Add better docs for swipe and pan gestures.
Change-Id: Ia147c7c9a22e084c3700b1620dec46427d792bd1
Reviewed-by: Povilas Kanapickas <povilas@radix.lt>
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
It came up during 6.2 API review that we prefer all floating-point API
to be double-precision on 64-bit platforms, despite the awkwardness of
representing a displacement vector with something called a "point".
The docs for QPointF explicitly state "A QPointF object can also be used
as a vector: Addition and subtraction are defined..."
Amends 31f90e99b8
Change-Id: I01029661f2586640cbf846f49df164c176d17f7a
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
For a style sheet that defines a rule for the menu icon, but not for the
item itself, we ignored the icon rule.
Implement a separate rendering path for menu items with an icon.
That rule takes precedence over a rule for menu item check indicators,
as almost all styles reuse the icon for that.
Factor the icon positioning and rendering code out into a private member
function that we can reuse. Reduce amount of local variables to make
the list of arguments for that member reasonable, the bit checks are
cheap enough.
Fixes: QTBUG-73966
Pick-to: 6.2
Change-Id: I64b6f5181e35527d0a163d9633a7414b50319829
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
The base style should still draw the item and the selection highlighting.
Fixes: QTBUG-83619
Pick-to: 6.2 6.1
Change-Id: I1cd8c4fce48cb5543337415918fc99ffb5e19e00
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
When saving to a QIODevice, QImage and QImageWriter will automatically
deduct the file format from the filename if it determines that the
device is a QFile. That did not work for a QSaveFile device. Fix by
using the common ancestor, QFileDevice, in the implementation.
Fixes: QTBUG-89022
Pick-to: 6.2
Change-Id: Ie01d80df4f29ca0d4ff30bf7e1b77605293c070e
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
This prevents us from first reserve()ing Prealloc elements, and then
possibly reserve()ing a larger number, which leaves the first bucket
list's memory unused.
Consequently, deprecate reserve().
Change-Id: Ifc0a5a021097f4589557e7b5e45d9d0892797ade
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Was missed in the int -> qsizetype port for Qt 6.0.
Pick-to: 6.2 6.1
Change-Id: I1ae8190601f2e1a1bc02a736c12230a9c71acb18
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
...by providing std-compatible insert() functions via a local subclass
of QSet, reducing the #ifdef'ery somewhat.
Change-Id: Ib532a866b47b82e8e3b9f199e8d1e01a87ed016d
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
All callers use the shared-ptr version these days.
Change-Id: I77e9fc9ccb8a57bfebcad7883e9eaff3780748f0
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
They're now in QWidget itself. Remove them from the API, but not the
ABI.
The QToolBar case is straight-forward. QMenu is a bit more complicated:
Since QT_CONFIG(shortcut) builds changed the signature of an existing
function instead of adding/removing an overload, we have to deal with
two cases: In a QT_CONFIG(shortcut) build, these overloads that take a
trailing QKeySequence parameter have been deprecated and therefore
cannot be removed. In a !QT_CONFIG(shortcut) build, the same functions
are 1:1 copies of QWidget functions and can be removed (from the API).
[ChangeLog][QtWidgets][QMenu/QToolBar] The addAction() functions have
been moved down into QWidget.
Change-Id: I49997b3440c137a1d4e3858d1d27d34a191e1eed
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
Since any QWidget can have actions since at least Qt 5.0, it makes no
sense to keep the addActions(text) etc. convenience functions that
have traditionally existed in QMenu and QToolBar only in these two
classes, where, to add insult to injury, they were just copies of each
other, increasing library size for no real reason.
So, add them to QWidget, too. This will allow us to de-duplicate the
code between QMenu and QToolBar, too, which will be done in a
follow-up patch, subject to BC constraints.
[ChangeLog][QtWidgets][QWidget] Added the same addAction(text)
overloads that previously existed only on QMenu and QToolBar, with the
following two differences: First, the QKeySequence object, if any, is
now available on all overloads, not just the ones taking a slot, but
has changed position to allow, secondly, passing an optional
Qt::ConnectionType parameter which will be passed to the
QObject::connect() call. In addition, the function template overloads
are now properly constrained so that they exist if and only if the
corresponding QObject::connect() call does, too.
Change-Id: Ifc3c2789600cf603192de8224fecbf9c88d91970
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Use unique_ptr to indicate ownership. Use std::stack to enforce
stack-ness (QStack doesn't). Requires the addition of a clear()
function, for which we spin a SmallStack wrapper. It also helps to not
have to repeat the value_type...
Because we use unique_ptr, `state` and `states` are now exclusive
(before, `state` was always `states.back()`). Correspondingly, rename
`states` → `savedStates`.
As a drive-by, this fixes an off-by-one error in a qWarning() (checked
`> 1`, but reported one saved state too many).
Change-Id: I8faade59300401be802ddc52c64ce31b8f23dc52
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
QLineEdit was the only widget in Qt that flips the widget's layout
direction if a key press changes the document's layout direction.
This resulted in inconsistent behavior: setting text to RTL content
programmatically didn't set the widget's layout direction to RTL, but
the first key press (even if just a cursor key or a latin1 character)
did. Clearing the text with keyboard shortcuts changed the layout
direction, doing the same using the clear button or context menu actions
didn't.
In addition, the automatic changing overrode whatever the UI developer
set, or what Qt detected as the global layout direction based on the
active translation.
Remove the automatic flipping.
[ChangeLog][QtWidgets][QLineEdit] QLineEdit used to change the layout
direction on each key press, based on the text content. This feature
resulted in an inconsistent and erratic user experience, and has been
removed.
Fixes: QTBUG-95011
Pick-to: 6.2
Change-Id: Ib653350d35e25ce431cca1abe9b45bdf33d035fe
Reviewed-by: Fan RuiJie <fanruijie@uniontech.com>
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
The \sa was neglected
Amends d33040548f
Pick-to: 6.2
Change-Id: Ib298ac3390819f9ea4092a1f2bf470ea82c9e4fe
Reviewed-by: Andreas Buhr <andreas.buhr@qt.io>
Outside tests, all in-tree callers of QObject::findChildren() pass no
name to match, and in my experience that is also true for the vast
majority of out-of-tree users.
Avoid the temporary QString creation in the caller and the repeated
QString::isNull() checks in the implementation by overloading
findChildren() without a name argument and checking for name.isNull()
only once, forking off into separate helper functions.
Adjust in-tree callers that used an explicit `QString()` argument in
order to pass options, which goes to show that `name` should never
have been the first argument of findChilden() in the first place, even
though I appreciate the symmetry with findChild() (the use-cases of
which, however, are radically different).
Change a `findChildren().size() == 0` call found while scanning for
findChildren() calls to `!findChild()` as a drive-by.
Modernize loops in the various qt_qFindChild{,ren}_helper() overloads
to match how the new code looks.
[ChangeLog][QtCore][QObject] Added findChildren() overload taking no
name (thus optimizing this common case).
Change-Id: Ifc56e5438023d079b40c67f11ae274a3e128ad5e
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
We have space for so many elements, so reserve()ing anything less
makes no sense.
Pick-to: 6.2 6.1 5.15
Change-Id: I84d692b10a6a491c37661f84aa3fdd9af43d71e5
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Instead of just sizeof(T), we, of course, also need to take the
support structure into account, to wit: the bucket list and, in the
node, the next pointer and the stored hash value.
Pick-to: 6.2 6.1 5.15
Change-Id: I8227a95c49e316aacf3d4efd8f6170ea3bea1cf0
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
SSL_get_peer_certificate was deprecated in OpenSSL v3 and can be 'compiled-out'
using OPENSSL_API_COMPAT. Use SSL_get1_peer_certificate instead.
Pick-to: 6.2
Task-number: QTBUG-94596
Change-Id: Iedb2e06e673e981cab79d4bf0147ac6f5f90089a
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
QT6_ADD_RESOURCE_DECLARATIVE_EXTENSIONS is never defined now. All qml
files are expected to be added via dedicated functions provided by the
qtdeclarative repo rather than being given to qt_add_resources()
directly.
Pick-to: 6.2
Change-Id: I44feca84a2e3d4ef8ecc0140f88625ff192bfdfd
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
QSettings encodes QVariants as @Type(data) strings. If that data contains
a null-byte, we write the string as UTF-8 encoded CFData. When reading it
back we look for a @ prefix, and then pass it as UTF-8 through stringToVariant.
The problem arises then the user writes raw QByteArrays with a @ prefix.
We can detect this situation by checking the result of stringToVariant,
and if it's just a simple conversion of the string into a QVariant, we
know that stringToVariant hit its fallback path due to not finding any
embedded variants.
If that's the case, we return the raw bytes as a QByteArray.
Pick-to: 6.2 6.1 5.15 5.12
Change-Id: I4ac5c35d0a6890ebea983b9aca0a3a36b0143de2
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Cleanup: This patch does not change functionality.
In several places, sorting inside of the QTreeModel is disabled
and afterwards restored. There is a RAII class for this, which is
used in some places, but not in others. This patch introduces
more consistent usage of this RAII object.
Change-Id: I9802998ad31d0f9d4417824f72ff0abbb0c38a17
Reviewed-by: Ievgenii Meshcheriakov <ievgenii.meshcheriakov@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
... instead of raw pointers or QSharedPointer.
Raw pointers are, of course, a no-no in modern code. In particular,
when the result is then held in shared_ptr or QSharedPointer,
make_shared or QSharedPointer::create() should be used to reduce
number of memory allocations.
Since this is private API, we're free to use std::shared_ptr, which
does only half the atomic operations on copies, compared to
QSharedPointer, so is more efficient.
For either make_shared or QSharedPointer::create(), we need to work
around the private ctor, which we do by inheriting a member-function
local class from QColorTrcLut and make_shared'ing that. As a
member-function-local class, it has access to the otherwise private
parts of QColorTrcLut, including its default constructor. As a public
subclass, shared_ptr has no problem performing the derived-to-base
pointer adjustment in the return statement. This way, we can use
make_shared even though our target's class' ctor is private.
Change-Id: Icb11249b54cd5e544e692f6a0bf1f9dda1710454
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
Instead of special casing sRGB.
Pick-to: 6.2 6.1
Change-Id: I7ad20c4b9600210333d5e44c3a2d20ce7d39dd92
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
Use QByteArray::fromRawData() and QList::fromReadOnlyData() to return
the QList<QByteArray> without allocating memory.
Change-Id: Iefa3250ea2acc79df92400d2d225e450566477e2
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
...where applicable to use a known CMake-provided variable in favor of
our own cooked up one.
This affects four plugins that are not supposed to be built out of tree,
meaning QtBase_SOURCE_DIR is available (and identical to
QT_SOURCE_TREE).
Task-number: QTBUG-88090
Change-Id: I27f012e6c5fd99c3239a1dd65c0811c819dbcfe9
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
Use the BASE argument of qt_internal_add_resource to avoid the
specification of resource aliases.
Use QtBase_SOURCE_DIR in favor of QT_SOURCE_TREE to use a known
CMake-provided variable in favor of our own cooked up one.
Pick-to: 6.2
Task-number: QTBUG-88090
Change-Id: I0d8b30a1564495db475ae58596b52b373b2b173e
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
Originally, the template specialization QtPrivate::IsSequentialContainer
was done in Q_DECLARE_METATYPE_TEMPLATE_1ARG. This was conceptually
wrong and prevented re-using Q_DECLARE_METATYPE_TEMPLATE_1ARG.
This patch moves the specialization to
Q_DECLARE_SEQUENTIAL_CONTAINER_METATYPE, which makes it symmetrical to
Q_DECLARE_ASSOCIATIVE_CONTAINER_METATYPE and allows re-using
Q_DECLARE_METATYPE_TEMPLATE_1ARG with non-container templated types.
Change-Id: I3b952ed36e4c140bce3511aebc1e5599c4822e34
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
For generic interop with other parts of Qt, we need to be able to
extract the type-erased internal storage of QFuture<T>. In particular,
QtWebChannel needs this to transparently support QFuture<T> as a method
return type.
Task-number: QTBUG-92903
Change-Id: I763f054656b8810d58720262d364baf42c47eb37
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
SFINAE out the new-style connect() overloads for const char*
functions.
Before, we detected the problem in the body of connect(), by failing
the static_assert for signal/slot argument compatibility, but that's
too late. We don't want to select that overload at all, because we
want that decltype(QObject::connect(QAction*, &QAction::triggered,
QObject*, SLOT(something())) SFINAEs instead of returning
QMetaObject::Connect, because then all the classes that provide
convenience new-style connect syntax by wrapping the
QObject::connect() call can just use that decltype() to accept
anything that will succeed in the wrapped connect() call instead of
manually trying (and failing) to provide proper overloads themselves.
[ChangeLog][QtCore][Potentially Source-Incompatible Changes]
QObject::connect(sender, pointer-to-member, [receiver, ], SLOT(~~~))
is now SFINAEd out instead of triggering a static_assert(). That means
it is now possible to use Expression SFINAE
(decltype(QObject::connect(<sender>, <signal>,
std::declval<Args>()...) to properly wrap new-style QObject::connect()
calls with a single wrapper function, greatly simplifying the
implementation, incl. passing Qt::ConnectionType arguments. For more
information on how to use this in your own code, see the sources of
QWidget::addAction().
Change-Id: I4052a49eacd63a432a9648a667adc435744a633d
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
...lest some code tries to set it to no possible effect.
Change-Id: I7aed05200c1abeda1d2f9d88ceb99a5ce6132dbf
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
QTableView stored the current row/column selection start in an own
variable instead using currentSelectionStartIndex. This leads to an
inconsistent behavior when the selection is done with a click on the
header and then in a cell (and the other way round)
Fixes: QTBUG-92561
Change-Id: I4c8bda3a938de451b6eff2819141e86a6870fbef
Pick-to: 6.1 6.2
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Amends 556fbbea6e, which breaks
translucency of frameless tool windows. Possibly a bug in macOS, a
window being made miniaturizable should not have side effects on
background or translucency.
Pick-to: 6.2 6.1 5.15
Fixes: QTBUG-95042
Task-number: QTBUG-65637
Change-Id: I651196e3cf8a9c59b250e5f1b4064cc4ded02372
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Using the QTextFormat::FullWidthSelection property to select a line
would previously not take into account right-to-left text layouts.
With this patch, the whole line should now be drawn correctly for both
left-to-right, and right-to-left layouts.
Pick-to: 6.2 6.1 5.15
Fixes: QTBUG-91125
Change-Id: Ide7b340cd7b4060e7c00e55e0011a86ffdfb5eb4
Done-with: Volker Hilsheimer <volker.hilsheimer@qt.io>
Reviewed-by: Dmitry Shachnev <mitya57@gmail.com>
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
The workaround I had implemented based on alpha version of OpenSSL v3
and what developers (OpenSSL) said about DH back then is not going to work
in the end - they do remove all DH (struct and related functions) stuff
if you set a proper OPENSSL_API_COMPAT level. A proper re-write is required
to support OpenSSL v3, but then there is no reason to keep dead useless
code.
Pick-to: 6.2
Task-number: QTBUG-94596
Change-Id: Iae092dd08148521649a684879d30e190736e1abe
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
It was fixed quite some time ago for TLS counterpart (TLS_MAX_VERSION),
but somehow the similar fix was missing for DTLS. OpenSSL v3 deprecated
those constants and they can be compiled out by OPENSSL_API_COMPAT.
Pick-to: 6.2
Task-number: QTBUG-94596
Change-Id: Ia0246170a003db900c430b2fbfc177caf05a395a
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
None of the member functions use sharedFromThis().
Change-Id: I4d1fc0c3c18830df231833ce0ef631e1bb1fc27e
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
Cocoa renders the bevel of a push button always in the default height of
32 pixels. We handled the case that the button is larger than that by
switching to "squared" mode, but buttons that are smaller than that got
rendered incorrectly, with the bevel being clipped.
Fix this by always rendering buttons in "squared" mode when the height
of the widget is not the same as the standard. Adjust the default height
for "mini" push buttons based on visual testing. Make the equivalent
adjustment to the focus frame logic and general positioning so that
buttons and frames are centered compared to other controls.
As a drive-by, get rid of expensive QObject::inherits calls, and replace
with qobject_cast. The headers for all the relevant widget types are
already included.
Fixes: QTBUG-85972
Pick-to: 6.2
Change-Id: I4ad66fac2313fbee385c1eadcc243f69c9cace70
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
u8 literals changed type from const char[] to const char8_t[] in C++20
and Qt APIs that don't use QUtf8StringView aren't prepared for it. Use
a classical char literal for the time being.
Amends 857be50b2e.
Pick-to: 6.2
Change-Id: I7f194befc08ab50cdef321d8900ee3553599a19e
Reviewed-by: Andreas Buhr <andreas.buhr@qt.io>
If Qt was configured with -static-runtime which implies MultiThreaded
-MT flag, the plugin initializer object libraries were still compiled
with the default -MD flag.
When an application linked to Qt, that caused linking to fail with
mismatched symbol errors between the application symbols and the
plugin initializer object library symbols.
Make sure to set the MSVC_RUNTIME_LIBRARY property on both plugin
initializer and resource object libraries, depending on the value of
QT_FEATURE_static_runtime.
We did set the property for resources added by
qt_internal_add_resource, but not for the resource created by
the public qt6_add_resources counterpart.
Pick-to: 6.2
Fixes: QTBUG-95043
Change-Id: Ia543cd0241db94a12080be2655ad420fe9ad3f24
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
Amends c0b3c06a7d.
The QCompleter used by QFileDialog uses starts-with matching of
the path parts, for all parts of the path. It uses the EditRole of the
underlying model for the data to match.
The path parts used for a path on Windows are "C:", "Users", "etc...",
and then finally whatever input has been typed so far. If we try to
match that against a starts-with rule against "Local Disk (C:)", then it
will fail and the completer will never have anything to show.
So, for the EditRole, return just the volume drive letter as we did before
c0b3c06a7d, not the volume name. Neither
can be edited anyway. This happens to then match the native file dialog
behavior, which doesn't complete "L" to "Local Disk (C:)" if the Computer
contents (ie. volumes) are shown.
Augment the QCompleter test case to cover that scenario (it already has
everything set up for this particular combination, even though we are
actually testing QFileSystemModel returning the correct data).
Fixes: QTBUG-94799
Task-number: QTBUG-78043
Pick-to: 5.15 6.2 6.1
Change-Id: I14844d91601e9b16dc07cff2ca48713acb7cdd09
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
There are four messages that we, for some reason I don't understand,
have our own text for instead of using strerror(). But even though they
were marked for extraction, they weren't translated.
Fixes: QTBUG-95039
Pick-to: 6.2
Change-Id: I266f1bf9a4d84db39086fffd168f8c6dfe9c2cf4
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
The method was never tested, but it failed to compile after
QMultiHash was introduced as a separate class in 6.0.
This patch fixes it and adds some unit-tests to cover the case.
Task-number: QTBUG-91736
Pick-to: 6.2 6.1
Change-Id: I5dd989d4775efc6a9bb13c5ed1d892e499d95dc2
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
This script was written for Python 2, which is deprecated now.
Change-Id: I534c93922776254a4c822ac60c6fd77ca7cb9760
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Reviewed-by: hjk <hjk@qt.io>
Let androiddeployqt write a dependency file so that
the build system knows when to re-run it.
Fixes: QTBUG-94567
Change-Id: I5985d707f257b22789013a74f0a6f7c4de6e5e88
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
The base implementation discards the data by reading into a dummy
buffer, which is slower than necessary.
Change-Id: Iabf0c4a25746af6cac5b61d7bda66d89501c808c
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Do not detach when find(key, value) is called on an empty QMultiHash.
As a drive-by: fix return value for QMultiHash::remove() in case of
empty QMultiHash.
Task-number: QTBUG-91736
Pick-to: 6.2 6.1
Change-Id: I1e32f359e7ee9ce8403dae79d02e0b88a20ec4a5
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Says GCC:
qdbusmarshaller.cpp: In member function ‘QDBusMarshaller* QDBusMarshaller::beginMap(QMetaType, QMetaType)’:
qdbusmarshaller.cpp:286:17: warning: ‘%s’ directive argument is null [-Wformat-overflow=]
286 | qWarning("QDBusMarshaller: type `%s' (%d) is not registered with D-BUS. "
| ^
Fix by manual checking. It's a False Positive, because
QString::asprintf() can handle nullptr strings just find, but let's
make GCC happy.
Since the code snippet appears multiple times, even though GCC only
warns about one of them, take this opportunity to factor the code out
into a cold helper function.
Change-Id: I1d642f2465f34b670b179646185cba05cb16e573
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Now that calling hashLengthInternal() is cheap, use it to factor
common code in switch statements. For sha3, that would have been
possible before, too. Reason for duplicating the case bodies is
unclear.
Change-Id: I281617546e0b3e701315eee2f10df8d26ada92ae
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
... instead of the "usual" rule of three: ctor, addData(), result().
Not only does it generate less code in the caller, it's now also
faster.
Change-Id: I67c7eeb01f527b90e80a08f60c1c7f2ec1e49dd4
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
char16_t, as returned from QStringView::utf16(), is guaranteed to be
unsigned, so there's no need to reinterpret_cast to ushort.
Change-Id: I7432dd26b4814c61fe70cc33ed307097ef46ce4e
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
We call evaluateRecursive_inline in setBinding, which in turns runs the
noSelfDependecies check. However, creating a binding resuting in a
binding loop must not crash, but instead result in the binding entering
an error state. To prevent a crash caused by the assert in debug builds
of Qt, we replace the assert with a warning for now.
A better approach in the future would be to ensure that we only run the
check in cases where we are sure that a self-dependency is really a
fatal error.
Pick-to: 6.2
Change-Id: I58158864ed81fa907132a4e7d6667c9b529e7e64
Reviewed-by: Maximilian Goldstein <max.goldstein@qt.io>
... so don't try to call it.
The call resolves to a fall-back implementation of toString() that
just returns nullptr, so we might as well pass nullptr directly, like
three lines above, and not confuse readers of the code.
Change-Id: I85d196aae7a0317050cfe16b206252df3fdffa2d
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
The pro2cmake.py conversion script faithfully reproduced the .pro files
for the plugins, which specified the libraries as public. But in CMake,
the implications of this are that public usage requirements should then
be propagated to consumers. We don't expect any consumers, since a
plugin is created as a MODULE library in CMake, so for Windows we don't
even have an import library to link with. The only exception to this is
for static builds where plugins are created as STATIC libraries
instead, but only in certain controlled situations do we then link to
plugins. Even then, usage requirements are not expected to propagate to
the consumers, so these relationships should always be specified as
private.
Pick-to: 6.2
Task-number: QTBUG-90819
Change-Id: Ibc7c2bcd3b6a9dc77df40c4c0c22ff254a80f33d
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
C++20 gave us constexpr std::swap, but it makes no sense for qSwap
not to be constexpr anyways.
[ChangeLog][QtCore][QtGlobal] qSwap is now constexpr.
Change-Id: I13f3cbf2870adf5770c62dc00e15004978fc85d9
Pick-to: 6.2
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
After dc794f7622, side widgets only got
space if they were not fading out, but the logic was not correctly
accounting for side widgets that never fade, such as buttons added via
QLineEdit::addAction.
Fix this to give visible widgets space, unless they are fading out. That
was the intent of the original change. Rename the variable to make its
purpose clearer, and reset it at the end of the fade-out animation.
Add a much-needed test that relies on private APIs to verify that the
effective margins are calculated correctly.
Fixes: QTBUG-94824
Pick-to: 6.2 6.1 5.15
Change-Id: If2ee6be52be9e4f9be1e91f72f27681ce27def6d
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Just create it on the stack, we know the lifetime.
Reduces memory allocations in static hash() from 2 to 1.
Change-Id: Ie0e22b023331da9a6f39c80b4cd1a5c016f63a87
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
These can be noexcept. Requires to add rvalue overloads for
QCborMap::fromJsonObject() and QCborArray::fromJsonArray(), too.
[ChangeLog][QtCore][QJsonValue] Added constructors taking
rvalue QJsonArray and rvalue QJsonObject.
[ChangeLog][QtCore][QCborMap] Added rvalue overload of
fromJsonObject().
[ChangeLog][QtCore][QCborArray] Added rvalue overload of
fromJsonArray().
Change-Id: I523c9a41166ea932fba4dd204072bd39d2192d2e
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Make sure the right variable is used
Pick-to: 6.2
Change-Id: Iec6d8392b320b704423ecbd4434270883fa9fd96
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Amends df31696321 to take care
of macOS 10.15 specifically, where the focus ring is a pixel
too high and leaves a gap that's visible on high-dpi displays.
On macOS 10.14 and before, and for editable comboboxes, the focus
frame is correct.
Fixes: QTBUG-94069
Pick-to: 6.2 6.1 5.15
Change-Id: I278aa30036808c06f8e71385d4797315e46d8f1d
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
...and use it internally.
Exported symbols are LD_PRELOADable, so the compiler might not
constant-fold calls to them. Besides, it's a requirement for a
follow-up change.
Change-Id: I437f46d7d42ed0a6921dee3027a471e2aa52baa1
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
It never changes, making it const might improve code-gen.
Change-Id: Ife8723e27ae9cf6cfcca48d58d46307003123354
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
The base implementation reads data using repeated calls to getChar(),
which is quite slow.
Change-Id: Ie46624df63791b2cdd3c8a28fe3327427d942505
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Incl. the static hash() function. Remove the QByteArray versions from
the API, but not the ABI.
Adapt some callers.
[ChangeLog][QtCore][QCryptographicHash] Replaced QByteArray with
QByteArrayView in addData() and static hash() functions.
Change-Id: Ia0e9bf726276305e05894d323d76a29e985f39eb
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
QIODevice::readLine() can also return partial lines, which was not
properly documented. Add an autotest for QLocalSocket to illustrate
and test this behavior.
Pick-to: 6.2
Change-Id: Ia2c1c438cc68d2672d34881e11fdf7837232f3b4
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Need to decrement 'remaining' (check), but also increment data (meep).
Testing is a bit complicated, as most algorithms are just too slow to
fit into the 5min QTestLib timeout. Picked the fast ones and Sha512
(which completes here in < 17s, with threads), at least.
Amends e12577b563.
[ChangeLog][QtCore][QCryptographicHash] Fixed a bug where presenting
more than 4GiB in a single addData() call would calculate the wrong
result().
Pick-to: 6.1 6.2
Change-Id: Ic72916ebc33ba087d58225af6d8240e46e41f434
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Marc Mutz
Allan Sandfeld Jensen
Timur Pocheptsov
Nico Vertriest
Sona Kurazyan
Giuseppe D'Angelo
Edward Welbourne
Volker Hilsheimer
Robert Löhning
Alexandru Croitor
Ivan Solovev
Fabian Kosmale
Luca Beldi
Markus Goetz
Mårten Nordheim
Nodir Temirkhodjaev
Tang Haixiang
Daniel Nicoletti
Tor Arne Vestbø
Assam Boudjelthia
Craig Scott
Øystein Heskestad
WenTao Song
Kai Köhne
Shawn Rutledge
Andreas Buhr
Arno Rehn
Alex Trotsenko
Mike Achtelik
Eirik Aavitsland
Joerg Bornemann
Christian Ehrlicher
Oliver Eftevaag
Thiago Macieira
Cristian Maureira-Fredes
Topi Reinio
Nicolas Fella
Friedemann Kleint
Liang Qi