It's just a proxy-setter so we cannot really avoid the call
Pick-to: 6.3 6.2
Change-Id: I4230075a906c100ddb9722160859e7a2c8e26ebc
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
The code didn't make much sense:
- QHighDpi::fromNativeLocalPosition should not take global positions
- QPlatformWindow::mapFromGlobal() should not take local positions
Tested with a basic MFC window that has a QMainWindow as child
(parented with QWinWidget from QtWinMigrate). Menus would be off
when on a secondary screen with dpi factor > 1
Change-Id: Ib5cbf5d29ed97c786f7f7ba4e868b70cb5435239
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
the char array of argv passed to the app's main() function
need to be null terminated [*] and argv to be **char.
[*] http://eel.is/c++draft/basic.start.main#2
Fixes: QTBUG-101028
Pick-to: 6.2 6.3 5.15
Change-Id: I6e3a685f092a92a94f60d166357b75d2538af74c
Reviewed-by: Ville Voutilainen <ville.voutilainen@qt.io>
QDateTimeEdit's default constructor instantiates the start of
1752-09-14 as its default earliest time; however Friedeman has seen
this triggering an assertion. The QDTE tests should be picking that up
anyway, but let's overtly test it in QDate's startOfDay testing, too.
Change-Id: Ifae87f2695ac3a7993c173a7c21809c87d5daa71
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
A recent change introduced a cut-off in believing MS's claims that a
zone has always had a yearly cycle of DST, to stop believing it before
1900, as no actual zone adopted DST that early. However, this was only
applied to the data() method to get the data for a specified time;
failure to do the same for {next,previous}Transition lead to
inconsistent results and tripped up an assertion. The change to data()
also failed to consider the situation of a southern hemisphere zone,
for which the rule will claim the early part of the year was in DST;
and init() should also mark an initial DST-recurrence rule as starting
no earlier than 1900.
Add corresponding FIRST_DST_YEAR checks to {next,previous}Transition()
to be consistent with the one in data(), refine data()'s checking and
amend init().
This amends commit 9a83706046
Pick-to: 6.3
Fixes: QTBUG-100873
Change-Id: Ic9822b87caa01561bc0b7d0355963e4c80ef047c
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
The first transition in a year with DST is *to* DST precisely if the
year *doesn't* start in DST, not if it does.
Also fix the startsInDst() check it's using there, which neglected the
possibility that the year's DST transition might be faked.
This amends commit d98b43005e
Pick-to: 6.3
Change-Id: I17422fa9972f826b5bf7e325f5e579f41300fd97
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
In nextTransition() it wasn't being carried over from one rule to the
next, as intended. It needs to be defined at function level, set for
the first rule and updated for each new year tried. Also assert that
it does faithfully reflect the standard time in rules without
transitions.
In the process of fixing that, since it's computed from the
yearEndOffset() of a prior year, test whether that year is less than
the rule start rather than whether the present year is less than or
equal to it. As the prior year is immediately before the present year,
this makes no difference, but it clarifies the reasoning. Apply this
clarification also in data() and previousTransition() for consistency.
Pick-to: 6.3
Change-Id: I29c41e67784eaae13b83f6ae1ad16509e636c187
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
In QWinTimeZonePrivate::previousTransition(), an initial no-transition
first rule was interpreted as implying a transition to its standard
time at "the start of time". Do the same for times before the start of
the first rule even if it is a DST recurrence; and, in data(), treat
times before the first rule as being in its standard time.
In the process, restructure data() to do that early return first,
instead of in an else clause, so as to dedent the other branch of its
code and make clearer how it fits into its outer loop.
Pick-to: 6.3
Change-Id: I21482d904c33542bf04f6510b974c01817d7aa5f
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
A transition outside the range of qint64 would be mapped to
invalidMSecs(), the same as the fake-detection sets a fake transition
to. This would have lead to a year at the boundary of qint64's range
being mistaken for a fake-DST year. So replace the fakesDst() method,
that compared transition times to invalidMSecs(), with an actual
boolean member that gets set when detecting fake DST, so that the
detection is correctly done.
Pick-to: 6.3
Change-Id: Iadc80973bc033915733c4a4f4ccfdd3863025fb4
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The end-of-year checks previously presumed they would accompany a zero
daylight-savings change there; however, this misunderstands what a
fake transition is doing. A fake at the end of the year means that the
transition during the year is out of the time the fake claims it goes
into, so the offset nominally in effect after the fake should match
the offset the year started in, the same as for a year-start fake.
However, implementing that exposed the fact that Bangladesh's brief
flirtation with DST in 2009 ended at the end of the year, and MS's
data has no other instance (aside from entirely bogusly claiming that
Bhutan did the same in 2009) of an end-of-year transition, in
particular no fake transitions. So the end-of-year case for faking
does not arise and I've eliminated it from the code here. This
greatly simplifies the code to detect fake DST rules.
At the same time, in the ruleToData() wrapper, use the transition
implied by isDst only if it isn't the fake one; the other is the only
real transition in the year anyway, however described. Also, extract
the detection of start-of-year transition times to a function and make
it more exact.
Pick-to: 6.3
Change-Id: Ide160d28e7cca37c224f40164ebe7f9a17f028e7
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The minMSecs() itself is one more than the type's min(), which is used
as invalidMSecs(). As (at least) the Windows back-end uses minMSecs()
as the time of a start-of-time transition (that we'd like to find as
our current transition), use one more than it as lower-bound for where
to search from for a previous transition, so that we do find that
first ever transition.
Pick-to: 6.3
Change-Id: Iae861e740e02bd38ffb2af77aff625d3b48182d2
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
The methods give them more appropriate names. This revealed one place
where the min() that's actually invalidMSecs() was being used for a
time that should have been in the supported range, so amend that to
use minMSecs(). Replaced a use of invalidMSecs() + 1 with minMSecs(),
to which it is equal, as that was the meaning it was used for.
At the same time, make those methods constexpr (because they are) and
[[nodiscard]], since their values should be used, while dropping their
fatuous inline (the bodies are inline in the declarations).
Pick-to: 6.3
Change-Id: Idcd51c55850573372b44e6fcf08d5d2665b8a60e
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The test crashed for some reason, so other testcases are also not
executed. Skip the test for now to enable more tests in the CI
Task-number: QTBUG-89402
Pick-to: 6.3 6.2
Change-Id: I5ad38645d1b8f86c64da7208c0ae4f66d126c7d9
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
Most of the blacklisted tests were already fixed earlier.
The tst_QPlainTextEdit::adjustScrollbars() test needed a small fix
to show the window non-fullscreen, so that the scrollbar could appear.
Task-number: QTBUG-87423
Task-number: QTBUG-89402
Pick-to: 6.3
Change-Id: I849f411a5798053742323fc4db3fe30f2b690a8b
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
This reverts commit 293b7fab55.
Reason for revert: Since Qt 6.2 MinGW 8.1 is not anymore supported
Pick-to: 6.2 6.3
Change-Id: I9b2e5e08bee663d5289a69218493911544671e79
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Remove the unholy bool out parameter and make QtPrivate::to{Double,
Float,{Signed,Unsigned}Integer}() return a struct instead.
The struct contains what we'll most likely need for a full
QParsedNumber in the future: the value, an error code (always zero
atm), and a pointer to the first character that wasn't parsed (always
nullptr atm), so we don't need to change the ABI when QParsedNumber
eventually lands.
As an immediate positive contribution, even without the backend ported
away from bool out parameters, the functions can now be marked as PURE
and, in case of the FP versions, also noexcept (the int versions have
a narrow contract d/t the base argument, which, unlike the return
value, can be fixed later, by overloading).
Pick-to: 6.3
Change-Id: I67945af80a9b53d6f170502a6df3384895e82d3e
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This overload set may come in handy to cushion some of the concerns
regarding replacing QString functions with QAnyStringView ones.
Overloading with a Q_WEAK_OVERLOAD QAnyStringView function requires
users to jump through hoops in order to avoid the QString overload,
but with the Q_WEAK_OVERLOAD roles reversed, the QAnyStringView
overload becomes the preferred version, relegating the QString
overload to a fall-back to facilitate sharing where it makes sense
(e.g. for QObject::setObjectName()).
Pick-to: 6.3
Change-Id: Ic65ead505beee627976a306e2d430e800540a600
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
Rename the resize() function of its internal buffer into the possibly
better resizeForOverwrite(). The point is that unlike an ordinary resize
the data in the buffer is NOT initialized, and therefore must be
overwritten by the caller, "or else".
Change-Id: I7d82821e92f59f7eeab18c51d98003826ffe164b
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Reviewed-by: Øystein Heskestad <oystein.heskestad@qt.io>
Use two template parameters for the haystack and needle types, to test
all possible combinations of all argument types.
Note that the tests for QByteArray::count() are removed: it doesn't
make sense to have them in tst_qstringapisymmetry, and we already have
the symmetry tests for QByteArray in tst_qbytearrayapisymmetry.
Change-Id: I33901fd135eb7433f0d45300a7248aef4d40324a
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Don't use the haystack as needle when testing count() for QLatin1String.
This wasn't caught earlier, since QLatin1String has no count() yet, and
the codepath was never tested.
Pick-to: 6.3 6.2
Change-Id: I2764070894ddce047eceaea52456e5a521252dab
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
We have a few lambdas now. However, GCC probably inlines everything
anyway in release mode, so this is likely moot.
Change-Id: Ic15405335d804bdea761fffd16d50573621446ab
Reviewed-by: Kai Koehne <kai.koehne@qt.io>
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
They usually come in groups, so this avoids extra work.
Change-Id: Ic15405335d804bdea761fffd16d4fb5c41e122f0
Reviewed-by: Kai Koehne <kai.koehne@qt.io>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Also removed the Qt::CaseInsensitive searching, which was slow.
Change-Id: Ic15405335d804bdea761fffd16d4f863847c360b
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Explain what that number 8 is by way of a comment and a constant.
Pick-to: 6.2 6.3
Change-Id: Ic15405335d804bdea761fffd16d4f7567089c575
Reviewed-by: hjk <hjk@qt.io>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
We had proper CMake tests for all we needed, so we can drop the old
detection we were forced to use way back when. Said test (CMake's own
FindBacktrace.cmake) even found QNX's offering, which we reported as
"yes" in our configure output, but never actually used because it was a
different API. Now we'll report "no" for QNX.
Pick-to: 6.3
Change-Id: Ic15405335d804bdea761fffd16d4f73cfe4cf38a
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Kai Koehne <kai.koehne@qt.io>
Reviewed-by: Alexey Edelev <alexey.edelev@qt.io>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
File and function names are UTF-8 on Unix.
Pick-to: 6.2 6.3
Change-Id: Ic15405335d804bdea761fffd16d4fb71ec2b0d71
Reviewed-by: Kai Koehne <kai.koehne@qt.io>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
The test is no longer failing on Android in dev and 6.3, but still
failing in 6.2.
I think it's fixed by 63a35898f4
which is integrated into dev and 6.3
Task-number: QTBUG-87396
Pick-to: 6.3
Change-Id: I82e0aac1547f8e43353f0948cd3f91b4b8f9720e
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
On Android orientation changes are reported from the
DisplayListener.onDisplayChanged() method, while the screen size
changes are reported from QtLayout.onSizeChanged().
In practice these callbacks come in random order, so rotating the
screen multiple times might result in inconsistent order of signals
and events.
This patch makes sure that size change events always happen before
orientation changes. This is done by caching the new orientation
values and reporting them only when needed. At this point we also
need to use QMetaObject::invokeMethod() for orientation change,
like it is done for geometry change. Otherwise the orientation
update can still be processed earlier than the geometry change.
Also note that at some point we might get an orientation change
without a size change (for example Qt::LandscapeOrientation ->
Qt::InvertedLandscapeOrientation). That is the reason for
isSimilarRotation() helper function.
As a drive-by: ignore size changes with inconsistent values when window
size is reported to have old orientation, while the screen has already
been rotated. In such cases a new size change will be triggered shortly
with normal value.
Task-number: QTBUG-94459
Pick-to: 6.3 6.2 5.15
Change-Id: I5c98e526d0370d380344b2297169d5e0c0ee8ea7
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
Screen has 3 parameters:
- physical size
- screen size (or screen geometry)
- available geometry
Before this patch, they were reported in 3 different methods. Screen
size update and available geometry update both were generating the
"screen geometry changed" event, that updated the user-visible
parameters. As a result, at some point we could end up in an
intermediate state, when screen size was already updated, but
available geometry was not yet updated.
This was specially visible while changing the screen orientation:
at some point screen geometry could be reported for the new
orientation, while the available geometry - still for old orientation.
This patch introduces a new method to set all geometry-related
parameters together and generate only one "screen geometry changed"
event. This allows to maintain consistency between 'geometry' and
'availableGeometry' properties of QScreen.
Task-number: QTBUG-94459
Pick-to: 6.3 6.2 5.15
Change-Id: I844f6d0db87df8d5e6e9bcce5d27126384a5a0de
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
This is useful for Qt for Android builds, where we want to build Qt on
a single host, but make the Qt installation usable on any host
(Windows, Linux, macOS).
There are only two flavors of shell scripts, unix ones that use sh and
Windows batch files.
If Qt is configured with
-DQT_GENERATE_WRAPPER_SCRIPTS_FOR_ALL_HOSTS=ON
then we generate both of them regardless of the current host platform.
Note that the target_qt.conf file still needs to be patched to specify
a correct HostSpec value so that qmake operates correctly.
Other target_qt.conf values might also need path adjustments depending
on use case (like HostPrefix and HostData).
Pick-to: 6.2 6.3
Task-number: QTBUG-101357
Change-Id: Ic86caaa8b318467528cc82dc7fbfecde998cdb71
Reviewed-by: Jörg Bornemann <joerg.bornemann@qt.io>
The operators are declared in the Qt::Literals::StringLiterals
namespace, to avoid collisions in the global namespace.
[ChangeLog][QtCore][QLatin1String] Added literal operator""_L1 that
converts string literals and chars to QLatin1String and QLatin1Char.
Fixes: QTBUG-98434
Change-Id: Ia945a6acf4b8d4fbbb5f803264e4d79d7b17a8da
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
We can use the new org.freedesktop.appearance color-scheme setting in
the XDG Desktop Portal Settings to implement this in a DE-independent
way.
In the future we can also use the related "SettingChanged" signal in the
Desktop portal to support changing the theme during runtime.
Useful links:
Spec: d7a304a006/data/org.freedesktop.impl.portal.Settings.xml (L33-L45)
Blog post: https://blogs.gnome.org/alexm/2021/10/04/dark-style-preference/
Task-number: QTBUG-93955
Pick-to: 6.2 6.3
Change-Id: Ia5df22fb215ee94d68d2787da97a1c9f8f2c6b0c
Reviewed-by: Shawn Rutledge <shawn.rutledge@qt.io>
The implementation for extracting a QMetaEnum from an enumeration
type was duplicated from QMetaType::fromEnum, just use that one.
Also add an explicit cast for the valueToKeys call.
Change-Id: I51b1a214dbab2101ae704954727acf0219eb1604
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
QT_TYPESAFE_FLAGS allows explicit casts only to QFlags<T>::Int, which
is either int or unsigned int. The cast to the resp. other type fails.
To fix, first convert to QFlags<T>::Int with toInt(), and only then
cast to int or unsigned int.
Fixes: QTBUG-101399
Pick-to: 6.3
Change-Id: Ie74d53adc601cdf19708265b040092780676058f
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Lowering the architecture all the way to the i386 baseline without SSE2
was causing warnings about disabling SSE FP math. We don't need to go
that low for 32-bit.
Task-number: QTBUG-101198
Change-Id: Ibf4acec0f166495998f7fffd16d682d0f5beec11
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
That option makes printf() and some other functions become always_inline
functions with extent checking. Unfortunately, that also means GCC fails
to inline them when we're not compiling for the same target as
X86_BASELINE.
Fixes: QTBUG-101198
Change-Id: Ibf4acec0f166495998f7fffd16d682c7a7be94d8
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
We set the default verbosity of configure output to NOTICE in
regular Qt builds.
To preserve the STATUS verbosity in CI runs, we now pass it
explicitly when configuring all repos / tests.
Pick-to: 6.2 6.3
Change-Id: Ib63739c05855cfd5951d38ce3f8a7fc8d1218cd3
Reviewed-by: Alexey Edelev <alexey.edelev@qt.io>
Only show the more verbose configure output when configuring
with -developer-build (which matches --log-level=STATUS)
Otherwise in a non-developer build, restrict the output
to NOTICE+ message (includes WARNINGs and ERRORs).
Developers can still pass a custom log level when configuring.
For example -DCMAKE_MESSAGE_LOG_LEVEL=STATUS or
--log-level=STATUS.
The former method will be cached, while the latter is only applied
to the current configure invocation.
Also show the build instructions hint message only when configuring
for the first time.
[ChangeLog][CMake][configure] The configure output verbosity of
non developer-builds of Qt is now reduced by default. Pass
"-- --log-level=STATUS" to configure to make it verbose again.
Pick-to: 6.2 6.3
Change-Id: I7583a9c92142e0b1d7c5411b06403f40d8ebaf20
Reviewed-by: Alexey Edelev <alexey.edelev@qt.io>
Our associative containers' iterator's value_type isn't a destructurable
type (yielding key/value). This means that something like
for (auto [k, v] : map)
doesn't even compile -- one can only "directly" iterate on the
values. For quite some time we've had QKeyValueIterator to allow
key/value iteration, but then one had to resort to a "traditional" for
loop:
for (auto i = map.keyValueBegin(), e = keyValueEnd(); i!=e; ++i)
This can be easily packaged in an adaptor class, which is what this
commmit does, thereby offering a C++17-compatible way to obtain
key/value iteration over associative containers.
Something possibly peculiar is the fact that the range so obtained is
a range of pairs of references -- not a range of references to pairs.
But that's easily explained by the fact that we have no pairs to build
references to; hence,
for (auto &[k, v] : map.asKeyValueRange())
doesn't compile (lvalue reference doesn't bind to prvalue pair).
Instead, both of these compile:
for (auto [k, v] : map.asKeyValueRange())
for (auto &&[k, v] : map.asKeyValueRange())
and in *both* cases one gets references to the keys/values in the map.
If the map is non-const, the reference to the value is mutable.
Last but not least, implement pinning for rvalue containers.
[ChangeLog][QtCore][QMap] Added asKeyValueRange().
[ChangeLog][QtCore][QMultiMap] Added asKeyValueRange().
[ChangeLog][QtCore][QHash] Added asKeyValueRange().
[ChangeLog][QtCore][QMultiHash] Added asKeyValueRange().
Task-number: QTBUG-4615
Change-Id: Ic8506bff38b2f753494b21ab76f52e05c06ffc8b
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Simplify a conditional to save some repetition, purge some spurious
parentheses.
Change-Id: I727d22ce81733e765c4ee951475ccdb599063399
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Apparently the class docs didn't get an update when systemTimeZone()
was added (at 5.5). Also add QCalendar to the \sa.
Prompted by Jaishree pointing out a typo ...
Pick-to: 6.3
Change-Id: If6d38040fff4badc3c0bb765889c1289c560c2b0
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
[ChangeLog][Network][QNetworkCookieJar] It is possible to use
system's copy of publicsuffix database when it is available.
This behavior is enabled by default on Linux and can be
controlled using new command line switches -system-publicsuffix,
-qt-publicsuffix, -no-publicsuffix, and -publicsuffix=all.
Fixes: QTBUG-95889
Change-Id: I911e1a13c1422cdc35851953309fff064e7c5f26
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
