qt6_enable_object_libraries_finalizer_mode is not needed anymore
now that static Qt builds require CMake 3.21 and thus CMake takes
care of placing object library object files at the beginning of
the link line.
Rename qt6_enable_import_plugins_finalizer_mode to a more generic
qt6_set_finalizer_mode that can enable or disable multiple
different modes.
For now the only available mode is "static_plugins" which handles
the behavior of the old function name.
The mode can be enabled by calling
qt6_set_finalizer_mode(${target} ENABLE MODES "static_plugins")
Note that the function is re-tagged as being in Technical Preview.
Ideally we would not need it at all. But we need to provide some
workaround for the limitations of linking Qt static plugins in CMake
on Linux-y platforms that use bfd ld or ld.gold.
The workaround does not work well with dependencies wrapped in
generator expressions which is why we can't confidently advertise it
as a proper solution.
Our hope is that this can be fixed in future upstream CMake versions
and the function can be removed.
See 6fcc272ac9 for details.
Adjust the tests due to the renamed and removed functions.
Amends 19e789bace
Amends cdbb390c4a
Amends a25027eecb
Amends 6fcc272ac9
Amends a3c430f390
Pick-to: 6.2
Fixes: QTBUG-95169
Task-number: QTBUG-95601
Task-number: QTBUG-95603
Change-Id: I51b85f776ec29fc04fed1a637eba7d1f60609e69
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
Building a static library Qt configuration will now require a
minimum CMake version of 3.21.
Using such a static library Qt configuration in user projects will now
require the same minimum CMake version of 3.21.
Qt builders and packagers can still opt out of the mentioned minimum
required version by configuring Qt with
QT_FORCE_MIN_CMAKE_VERSION_FOR_BUILDING_QT
and
QT_FORCE_MIN_CMAKE_VERSION_FOR_USING_QT.
Such a Qt configuration is /NOT SUPPORTED/.
Increase the upper range of CMake policies set to NEW to 3.21.
Amends 9aa0d99e66
Amends 6518bcc167
[ChangeLog][CMake] Building Qt as static libraries now requires
CMake version 3.21 or later. Building user projects with CMake using
that Qt installation also requires a CMake version of 3.21 or later.
Pick-to: 6.2
Fixes: QTBUG-95018
Change-Id: I717b6e1207df46627443b3e5ed430f3c964e2b48
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
When CMP00126 is set to NEW, set(CACHE) doesn't remove regular
variable bindings with the same name as the cache variable.
This introduced an issue in qt_find_package, which called
find_package(CONFIG QUIET) to try and find a package config file, but
did not clean up the non-cache _FOUND variable which is automatically
set to 0 by CMake if no Config file is found.
Make sure to unset both the regular and cache _FOUND variables if either
the package config file was not found, or if none of the provided
targets found even if the Config file was found.
Also move the _DIR cache variable cleaning into the same code block.
Amends 4c31ce68d5
Amends 34b1c1c23c
Fixes: QTBUG-95635
Pick-to: 6.2
Change-Id: I871987217526e0f0a20038a8da52625838e49488
Reviewed-by: Craig Scott <craig.scott@qt.io>
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
When qcompleter and qlineedit are used together, the currentcompletion()
of qcompleter is its first item by default. Therefore, when qlineedit
makes the initial value, then selects the text and enters, qcompleter
will modify the default first item to qlineedit text. The judgment that
completionprefix() of the completer is not empty is modified and added
here, because completionprefix() is always empty when there is no match.
Fixes: QTBUG-20894
Pick-to: 6.2 6.1 5.15
Change-Id: Id47f7f4da0a93d03a4e7b056ec731615b0803d13
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
Reviewed-by: Tianlu Shao <shaotianlu@uniontech.com>
ICCCM 4.1.4 says:
Clients that want to re-use a client window (e.g. by mapping it again) after
withdrawing it must wait for the withdrawal to be complete before proceeding.
The preferred method for doing this is for clients to wait for a windown manager
to update or remove the WM_STATE property.
This patch implements the required logic. Qt 4 had something similar. Without
this patch we are calling various setter functions on a native window while it
is in an undefined state.
Fixes: QTBUG-69515
Pick-to: 6.2 5.15
Done-with: Liang Qi <liang.qi@qt.io>
Change-Id: I377a66ad3d5e43f14465d0ea670b2f43f96ed7d3
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
For the first call of QXcbClipboard::clipboardReadProperty()
inside of clipboardReadIncrementalProperty() in getSelection(),
it will get a XCB_NONE reply before the contents arrived via
property change. Then we give a chance to read more.
Manually tested with following setups:
* examples/widgets/mainwindows/application with gvim(gtk3)
* examples/widgets/widgets/imageviewer with GIMP 2.10.18(based on
gtk2) and GIMP 2.99.6(based on gtk3 via flatpak)
Fixes: QTBUG-56595
Pick-to: 5.12 5.15 6.1 6.2
Done-With: JiDe Zhang <zhangjide@uniontech.com>
Change-Id: Ib45f08464d39ad79137b1da99808c89b7dca2d08
Reviewed-by: JiDe Zhang <zhangjide@uniontech.com>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Add Q_DISABLE_COPY_MOVE to QCalendarRegistry to silence the following
warning produced by QtStaticAnalysisBot:
class 'QCalendarRegistry' defines a non-default destructor but does
not define a copy constructor, a copy assignment operator, a move
constructor or a move assignment operator
QCalendarRegistry is a singleton so it does not need to be moved or
copied.
The warning was introduced by d0ae1ef33a.
Task-number: QTBUG-93004
Pick-to: 6.2
Change-Id: I5e018346415b9d0a1ebc3bbde2ab7c3ad5e6d9d0
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Add two tests for decoding IDNs when they encode values outside Unicode
(> 0x110000).
"xn--5p32g" decodes to "a" (all ASCII) before QTBUG-95577 got fixed.
"xn--400595c" decodes to the same value as "xn--097c" after the above
mentioned bug got fixed. This test is currently failing.
Task-number: QTBUG-95689
Pick-to: 6.2
Change-Id: Icab55c41e0233b34d57e38232fa90ac42f35a50a
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Iterate over Unicode codepoints instead of UTF-16 characters
when converting to/from Punycode as described in the specification.
Additionally reject strings with invalid surrogate pairs when
encoding to Punycode, reject strings with any encoded surrogates
when decoding.
Remove expected failure marking from the test for this issue
in tst_qurlinternal.
Fixes: QTBUG-95577
Pick-to: 6.2
Change-Id: I3dd68f95ada6d652e2fa5c0c3118dcfa0a5f4c4d
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Helped quite a lot with OpenSSL 3 not accepting some old algorithms.
Pick-to: 6.2 6.1 5.15
Task-number: QTBUG-95123
Change-Id: If4894fa86eba7b002465fa661d436ae6ea751989
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
CMake master / 3.22 introduced a behavior change in how conditions are
evaluated in the while command. It is now consistent with how if()
evaluates conditions.
This caused an issue in Qt code where a "(" opening parenthesis was
evaluated as part of the condition rather than as a string.
Fix this by wrapping the evaluation of the variable in quotes.
Pick-to: 6.1 6.2
Change-Id: I70c26dc91394f4a14f7a26419df264a069dc7dc5
Reviewed-by: Kai Koehne <kai.koehne@qt.io>
Reviewed-by: Craig Scott <craig.scott@qt.io>
Replace phrases like "future has been running", "parent" with more
precise descriptions.
Pick-to: 6.1 6.2
Fixes: QTBUG-95273
Change-Id: Ibd5a464007d41cc437da49ba250b9ea0a46078c6
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Although the function is called get{Dat,Tim}eFormat(), the content
it's filling the buffer with isn't a format, it's a formatted date or
time. So name the variable we transcribe the buffer into text instead
of misleadingly calling it format.
Change-Id: Ia3a51184963109a50dbc41e205bd8c83d83c04d7
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
QMetaType does. QVariant should do the same.
Pick-to: 6.2
Change-Id: I3419276b78b3b5ce8bd144dee92685195797d568
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Reviewed-by: Daniel Nicoletti <daniel.nicoletti@kdab.com>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Unlike an acceptable effect in QTableView + QTreeView
setAdjustPolicy(QAbstractScrollArea::AdjustToContents)
unfortunately didn't work for QListViews (and QListWidget).
This patch corrects QListViews AdjustToContents
behavior.
[ChangeLog][QtWidgets][QListView] A more correct implementation
of QListView::viewportSizeHint has been made. That
implies that setting the sizeAdjustPolicy to AdjustToContent
on QListView and QListWidget will now cause the view to
size after the contents and avoid scrollbars.
Pick-to: 6.2
Task-number: QTBUG-58749
Change-Id: I1675115f2348e2fcf0b2c39b451ef337e10eb872
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
The -coverage argument merely added compiler flags for the Qt build. It
was never properly ported to the CMake build, and it doesn't seem
feasible to have configure arguments for every possible compiler option.
The same can be achieved by passing the needed compiler option to CMake,
for example: CMAKE_CXX_FLAGS=-fsanitize-coverage=trace-pc-guard
Pick-to: 6.2
Fixes: QTBUG-86227
Change-Id: Ieef9acaedc0a839f9fb35b4403395eea28643864
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
Previous fix hit too widely so some valid horizontal and vertical
lines were affected; the root problem being that such lines have an
empty control point rect (width or height is 0). Fix by caculating in
the pen width.
Pick-to: 6.2 6.1 5.15
Change-Id: I7a436e873f6d485028f6759d0e2c6456f07eebdc
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
The {mouse,touch}DoubleXDistance variables were initialized to -1,
which forces them into the DATA segments.
But this value is completely unused, since these variables are also
set in initThemeHints(), called from QGuiApplicationPrivate::init(),
before their only user, QGuiApplicationPrivate::processMouseEvent(),
can possibly execute.
By changing the default value to 0, then, we allow the compiler to
place these variables into the BSS segment.
Change-Id: I86144559ee2410b844f70a4f2499c8d3140d14cf
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
This class is an algorithm, and shouldn't be copied or moved.
Change-Id: I6f517f53a242dbfd61975ef885c358eef139a272
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The +[NSObject new] method is a combination of alloc and init.
Fixes: QTBUG-95619
Pick-to: 6.2 6.1 5.15
Change-Id: I341f8a3958fb7a016cf4c346750ea6d46eeebe9f
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
The intention is to remove TYPE as a keyword completely before 6.2.0
release, but in case if that's not possible due to the large amount
of repositories and examples, just print a deprecation warning for
now and handle both TYPE and PLUGIN_TYPE.
Task-number: QTBUG-95170
Pick-to: 6.2
Change-Id: If0c18345483b9254b0fc21120229fcc2a2fbfbf5
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
We don't produce <failure> elements for them, so we shouldn't include
them in the total count of failures.
In the future we might produce <skipped> elements instead, but for now
remove the incorrect increment.
Pick-to: 6.2
Change-Id: Ia89a5cdaf79f6aa2f92be53180112c40ad6da2c6
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Add code to check if the calendar registry is destroyed to all
QCalendar methods that dereference QCalendarBackend pointer.
Pick-to: 6.2
Change-Id: I9b562355e2e0579396b52968f6065c6927cc9ca8
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
All calendar backend accounting was moved into QCalendarRegistry class
(renamed from Registry). Calendar backends are no longer registered
inside constructors, because in multithreaded environment
this may lead to incompletely initialized instances becoming visible
via QCalendar API in multithreaded environment. All system backends are
registered by QCalendarRegistry itself when they are needed. New method
QCalendarBackend::registerCustomBackend() is provided to register
any 3rd-party calendar backends.
Registration by names was also simplified. The list of names is now
passed to QCalendarBackend::registerCustomBackend(). The checks are
provided to ensure that all system backends have non-conflicting names.
Name conflicts for custom backends are resolved in favor of earlier
registered backends, as it is already the case in the existing code.
The documentation was updated to reflect that.
Method QCalendarBackend::names() was added to query the list of names
associated with a backend after it is registered.
Calendar backend deregistration was completely removed because it is
not possible to perform it safely without reference counting.
Fixes: QTBUG-93004
Pick-to: 6.2
Change-Id: I0ab1eccc02fdd1e1c66b5e5dd076c93de32d5a49
Reviewed-by: Paul Wicking <paul.wicking@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
LayoutDirectionAuto is the default in qguiapplication.cpp's
layout_direction. Because the numeric value isn't zero, it forces the
variable into the DATA segment instead of the more efficient BSS one.
At a less technical level, LayoutDirection{} should really produce the
default, Auto, instead of LeftToRight. APIs have been changed for
lesser violations of inclusivity :)
Change-Id: I06b42f0bc00ea539805280ef0aa9694e6d739f41
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Hopefully we won't need it but let's have it still at least for
consistency
Change-Id: I72289e65e5e5613174ad4d98cf8d614f9caae8e6
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Instead of appending to a QByteArray and then streaming that one, just
stream the components of a line directly. QTextStream's op<<(const
char*) is not subject to QT_NO_CAST_FROM_ASCII, etc., so can be used
unconditionally.
Change-Id: Idd97a75a1b5b939de7176d40880a2f328d01927d
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
The data itself is not indented, as consumers may read it as verbatim
data.
Pick-to: 6.2
Change-Id: Ia934616cea273feadc3a45d7c74726d4f804f0dc
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Use U+102F7 as an example. The current code is not able to handle conversion
of Unicode codepoints outside BMP correctly, so mark the test as expected
failure.
Task-number: QTBUG-95577
Pick-to: 6.2
Change-Id: Ifd50cf306bce4940f84cb412de148eac952e6c09
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
When Qt5Config.cmake was ported to Qt6, the assignment of
_Qt6_FIND_PARTS_QUIET was accidentally left out.
Make sure that when find_package(Qt6 QUIET COMPONENTS Foo) is
called, the underlying component find_package(Qt6Foo) call gets
a QUIET option as well.
This will silence short-form warnings like
-- Could NOT find Qt6Foo (missing: Qt6Foo_DIR)
It currently won't silence any warnings produced by the transitive
find_package calls in QtModuleDependencies.cmake. That will require a
separate change.
Remove usage of _Qt6_FIND_PARTS_REQUIRED which was always empty.
See details of QTBUG-95573 why it's currently not used.
Pick-to: 6.2
Task-number: QTBUG-95573
Change-Id: I3822c068eecbe84c306a1b04f4187278d35ec3d5
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
Provide better error messages when a Qt package / component is not
found.
Mention the location of the expected Config file on the file system,
whether it exists
Also mention the location of the Config file that was specified by the
user when configuring the project or which CMake computed and stored
in QtModule_DIR cache var.
Mention that a package is not found in case if the main target exposed
by that package is not found.
If the target is not found, mention that it might be due to
QT_NO_CREATE_TARGETS being set to TRUE (when it is set to true).
If it is set to true, the assumption is that the target must have been
defined by something else before the find_package call (either an
earlier find_package call without QT_NO_CREATE_TARGETS set to TRUE or
maybe it's the use case of Qt being built together with its examples
or it's a super build).
Unset _Qt_NOTFOUND_MESSAGE to ensure that the Qt6 not found error
message is not spilled into any subsequent find_package(Qt6) calls,
causing a cascade of unwarranted warnings / errors.
Make sure to unset it only if components were specified, so the
message is not shown or unset in any of the recursive
find_package(Qt6) calls which is a dependency for regular Qt module
packages.
This works fine, because find_package(Qt6) calls with components are
only done in project code and not done by the transitive dependency
code (which looks for Qt6Foo packages directly).
Remove some dead code.
Pick-to: 6.2
Task-number: QTBUG-95532
Change-Id: Ie93d18e25e33aa0fe9e9da9a60e3d3e4cd6adb58
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
The major use-case of the now private QX11Info from Qt X11 Extras was
getting hold of the Xlib display and XCB connection, for example in KDE:
https://lxr.kde.org/search?%21v=kf5-qt5&_filestring=&_string=QX11Info
A new native interface for QGuiApplication has now been added that
exposes these two properties, e.g.:
if (auto *x11App = app.nativeInterface<QX11Application>())
qDebug() << x11App->display() << x11App->connection();
To avoid type clashes one of the enum values of QXcbNativeInterface's
ResourceType had to be renamed.
Pick-to: 6.2
Task-number: QTBUG-93633
Change-Id: I2e366a2bb88bd3965ac6172ad000ae32209f43e7
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
We don't (yet) have a workable back-end for it.
Pick-to: 6.2 5.15
Task-number: QTBUG-95314
Change-Id: I8cf0fd11f315bc7d6cac351d485f4cbc676ed95d
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
Reviewed-by: Lorn Potter <lorn.potter@gmail.com>
The uint, particularly, was requiring some contortions to avoid
getting a "negative" value, where using qsizetype makes it harmless,
as the resulting loops are then no-ops. In the process, document the
slightly eccentric semantics of one of the retyped variables.
Change-Id: Idaad4cfde9ed9d24e1bcbf03c2bdb10b62e07916
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Two blocks of code branched on the same local variable and, in each
block, one branch acted on a C-locale text, the other a localized
version, but doing the same to each in the two branches. Branch to set
a variable to the selected text so that we only have to write the code
to do each branch's actions to that one text.
Change-Id: I8bbc1210f2c14b19f41a9974c7b6ec2ae612cc70
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
No user of the variable iterates over the container, they only
manipulate or inspect individual entries, so order doesn't matter. If
order doesn't matter, use the more efficient QHash over QMap.
Change-Id: Id28913a57bb14eb328e668624f236f76cad86def
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>