The VxWorks implementation of std::variant is broken.
std::visit throws std::bad_variant_access when called on a variant with
duplicated types.
Skip the tests that are affected by it.
Pick-to: 6.7
Task-number: QTBUG-115777
Change-Id: I45227cc543ef7db2217b1d53313c2e2b140988d6
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
(cherry picked from commit 0e9c0c0847bd09c8a2b5166a2bafc103db0d3abc)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
A locale's UI language includes the script, e.g. the QLocale::uiLanguage
list for the Australian locale en_AU is {"en_Latn_AU", "en_AU", "en"}.
The old code iterated over each language in the outer loop, and for each
language tried several times, removing a segment from the back with each
try in the inner loop. The de-facto search order was then
en_Latn_AU
en_latn_au
en_Latn
en_latn
en
en_AU
en_au
en
Usually, translation files are provided for country_Territory, i.e.
en_AU. But if an en file (for a generic English translation) was also
present, then en_AU was never tried.
Fix this by breaking the logic into two loops: first, create a list of
candidates by removing segments from each UI Language. Then sort that
list of candidates so that entries with more segments come first. The
search order is now:
en_Latn_AU
en_latn_au
en_Latn
en_latn
en_AU
en_au
en
This way, en_AU gets loaded correctly, before en is tried.
Adjust the test, which was essentially duplicating the logic from
QTranslator to generate files. This only tested that two identical loops
result in the same thing. Instead of using the system locale, make the
test data-driven, and explicitly try en_US and en_AU, with candidate
files generated based on a hardcoded list for each data row.
Pick-to: 6.7 6.5
Fixes: QTBUG-124898
Change-Id: I6bdcff289d2843e61c9053c116e955b79e09e95a
Reviewed-by: Mate Barany <mate.barany@qt.io>
(cherry picked from commit 9a11273b745a30cebb5cd648c89eb224e9704492)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
The test failure seems to be related to code optimizations and not
to the c++ standard that is used. With a recent version of MSVC
(>=19.40) the test fails on both x64 as well as ARM64 if the build
configuration does code optimizations (aka we have a release
build).
Pick-to: 6.7
Fixes: QTBUG-126349
Change-Id: I01781ea5b20f80a6cbd7ccb284d30b5d078ad958
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
(cherry picked from commit d8cb2044e3e86633c732f1f1b28926062b08b76c)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
When the Java array that we hold a reference to is changed on the Java
side, then our reference continues to be valid, and references a
different object than the new array in Java.
Change-Id: If8743cc1b2fabb254b9c0009f41377455719ec3e
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
(cherry picked from commit 37bb00d8507a4def05b5d85a6ef4ea23cca70ab6)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
This avoids that we or users have to declare e.g. String or Uri in
several places in Qt. This also prevents problems where multiple
declarations (possibly from different headers) cause build errors.
As a drive-by, remove some unnecessary type declarations (e.g.
UriType, which had the same class string as Uri).
To ease the submodule update process, define a preprocessor symbol
that submodules can use to conditionally declare the type locally.
Once the dependency update is through, the symbol can be removed
and submodules can use the declaration from qjnitypes.h.
Change-Id: I7d96edf644a54246302b5c5cb478e66fa615e73e
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
(cherry picked from commit 457a1c973d68e705f9cf72ac72b19fc26cdb2917)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
Make the macro variadic. For N > 2 arguments, interpret the first
N-2 arguments as nested namespaces that are declared as inline
namespaces. We can then construct the signature string from those
namespaces, so
Q_DECLARE_JNI_CLASS(String, "java/lang/String")
is equivalent to
Q_DECLARE_JNI_CLASS(java, lang, String)
except that with the second overload, the C++ type will be in the
QtJniTypes::java::lang namespace. Since both the java and the lang
namespaces are made inline, QtJniTypes::String still works.
This helps with avoiding conflicts, as there might be Java classes
in different packages, but with the same name.
Change-Id: I0355239e7fc3c91cb26e041f19c4e0853d3678ac
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
(cherry picked from commit 2b1312bc954b72f59d0c1d7b1192e2f029016012)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
This fell out of my tracking of a failure of the new QPointer
conversion feature, and it would be a waste to throw it away unused.
Pick-to: 6.7 6.5 6.2 5.15
Change-Id: Ie255b4a6432b4763071e5712d92e9ccd57927052
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
(cherry picked from commit 0e40a3f6645a684458cbb09bef1353df2eec5e0b)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
Dogfooding the new QtTest 6.8 feature, and greatly simplifiying the
source code.
Change-Id: I0731d7291048e6bbed19666d1b6672e5cccaf27a
Reviewed-by: Ahmad Samir <a.samirh78@gmail.com>
(cherry picked from commit 157fcdf5afbe4e6b421713947ccdb4ee35116ee4)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
Dogfooding the new QtTest 6.8 feature.
This not only helps with the sourceModelBinding() test with its manual
currentTestFailed() calls, but also the old verifySubSetOf() helper
function that's there from Qt 4 times.
Use Ye Olde Scope Guard Trick™ to replace the qDebug() messages
recording where failures occurred.
Change-Id: I4d1460bde5315dfcc3b261b6e09c7293bbd9ff89
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
(cherry picked from commit 32fc8b936d1d036ba5b70ec1cac7681e1c9359c8)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
QFileSelector tests fail, because VxWorks is not properly categorized.
Even though it is not a unix-like system, it does provide a
compatibility layer that enables some unix utilities.
It has also been treated as unix on the Qt 5 customer branch.
Solve the issue by assigning VxWorks to the unix category in
QFileSelector and its tests
Task-number: QTBUG-115777
Pick-to: 6.7
Change-Id: Icab80764b66b121995f51ddf149de55dc8c9eb55
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
(cherry picked from commit da5de2926ac9573b13d0d743e43e685cbdb2abaa)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
This is expected behavior for default-constructed containers.
Also add the missing isEmpty() member function to QJniArrayBase.
Change-Id: I23111f6906ef5476567272cb23746fec962afa35
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
(cherry picked from commit e5516191b25d70ee9d6e87246d932fa559ef0b59)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
The pattern "*debug*" also matches libinvalidplugin_debug.dylib
on macOS. As the CMake target for the plugins include the 'plugin'
part, let's look for that as well when finding matching plugins.
Change-Id: Icf3177d42c104829890a8c404a6b9782bf190bb4
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
(cherry picked from commit 087ace615a6399f6bf7a020bce07365689bd4784)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
We can't do anything about a failed write at these code locations. In
particular, we can't QVERIFY() their success, so suppress the warning
with the idiomatic cast to void.
Amends 13a1995e9d,
bd32c7d705 and
c5221f6be0.
Pick-to: 6.7
Change-Id: Ia7ef57f01eeb8608079a12724c0b4f03a59081c9
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
(cherry picked from commit 97f31a68795e1d492f98a9942cce5228cbe64cf8)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
GCC 14 doesn't understand that neither q_check_ptr() nor
qCompare<int*> dereference the pointer.
Since the missing initialization isn't part of the test (except,
maybe, to provoke this warning if the implementation indeed changed to
dereference the pointer), fix by initializing the int, like we do for
the string a few lines below.
Amends the start of the public history.
Pick-to: 6.7 6.5 6.2 5.15
Change-Id: I6033e026f041b03821b776daa8e65586b8054136
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
(cherry picked from commit 8e914e07fbac8b40e51d0e6da764fa13a6af8181)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
tst_QCoreApplication test fails due to leaking object. Investigation
about the reason of this issues was started, but no conclusion was made
and reason for object leak is still unknown. To prevent this test from
crashing exclude this check for VxWorks as well.
Pick-to: 6.7
Task-number: QTBUG-115777
Change-Id: I2c543a3b8320e2d96b2e7cd7e14c1980e500ce0d
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
(cherry picked from commit 92e420aefb373c2dc1dd39a05ff43037edb2a145)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
Dogfooding the new QtTest 6.8 feature.
Removes a lot of manual currentTestFailed() calls.
Change-Id: Id8bb0bad9b90b0a51315e9a9677d7009ebc090e4
Reviewed-by: Ahmad Samir <a.samirh78@gmail.com>
(cherry picked from commit 561a1cae9df85a7a03f73ce77948ef235fff7c5d)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
Commit 54c373faa4 updated the bitwise
operations to be more efficient, bypassing QByteArray and going straight
to QByteArrayData (a.k.a. QArrayDataPointer<char>). This meant we also
bypassed the initialization of the null terminator.
This wasn't caught in our unit testing and with some runtimes because
the memory we allocated happened to be zero or contain the information
we wanted. But with Visual Studio, the debug-mode runtime initializes
all newly allocated memory with pattern 0xcd, which showed up as a
problem.
[ChangeLog][QtCore][QBitArray] Fixed a regression introduced in 6.7.0
that could cause QBitArray to report wrong bit counts after a bitwise
operation.
Pick-to: 6.7
Fixes: QTBUG-126343
Change-Id: Icdc467f26dea4b05ad90fffd17d939c3b416adca
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
(cherry picked from commit 8ac57ff6bc778519bb1edc4036ce79ab8f688e27)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
This is far from complete, as per path coverage, but it should at
least have 100% statement coverage now.
Pick-to: 6.7
Change-Id: If2a9a3d61695cca253a4e12c33a505b5afe50aa4
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
(cherry picked from commit 171ff57be1b8fd1c1b33cffbffa389790f239b5c)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
It's harmless and enables the pointer-idiomatic assignment of nullptr
without having to resort to reset().
Found in API-Review.
Change-Id: Id3a02f55b3578a29943f729d8d734fc3b1d3af11
Reviewed-by: Rym Bouabid <rym.bouabid@qt.io>
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
(cherry picked from commit 62d3b3c680ed2c6ca417f30fb3ca57482339c2c1)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
The test checks if two paths test equal if they are pointing to the same
place, but one of them is not a canonical path.
Example of such paths: /tmp/.. and /
The test assumes that /usr is an existing directory on platforms other
than win, haiku and android.
The reason /usr is used instead of /tmp is that /tmp is a symlink on
macos and that interferes with the test (as described in 0b67ad9d).
However /usr does not exist on VxWorks (unless created by the user).
/tmp can (and is in our CI env) enabled by turning on the
INCLUDE_TMP_DIR component.
Solve the problem by using /tmp instead of /usr on vxworks.
Pick-to: 6.7
Task-number: QTBUG-115777
Change-Id: I98b0d46d1b879f404c0b5bd655031e7c3f19ba33
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
(cherry picked from commit 70e2fe2bba71ea0a4d1d1e533c1130c97f77df55)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
Strictly speaking, we did want to compare pointers, but char* are
special with QCOMPARE in that the pointed-to strings are compared, not
the pointers. So this wasn't doing what we wanted it to do.
We could static_cast to void* to do it, but subtracting one from the
other also gets us what we want, with simpler code. The drawback is that
if appendToBuffer() ever returns nullptr, we'll print a huge number as
the offset.
Found by Valgrind:
==3769588== Conditional jump or move depends on uninitialised value(s)
==3769588== at 0x483FEDC: strcmp (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==3769588== by 0x4C41EB6: qstrcmp(char const*, char const*) (in lib/libQt6Core.so.6.9.0)
==3769588== by 0x48975D8: QTest::compare_string_helper(char const*, char const*, char const*, char const*, char const*, int) (in lib/libQt6Test.so.6.9.0)
==3769588== by 0x12B12A: QTest::qCompare(char*, char*, char const*, char const*, char const*, int) (in tests/auto/corelib/text/qstringconverter/tst_qstringconverter)
==3769588== by 0x10EE94: tst_QStringConverter::invalidConverter() (in tests/auto/corelib/text/qstringconverter/tst_qstringconverter)
Amends c46ee7df57.
Pick-to: 6.7 6.5 6.2
Fixes: QTBUG-126107
Change-Id: I44265a5eb67e47a59fc8fffd17d64051657af529
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
(cherry picked from commit 9d5611dc97979dab1932b07f8cfab367c3872d24)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
The ctor is explicit so there should not be an assignment operator op=.
Changing the assignment to = Qt::totally_ordered_wrapper(ptr) is also
possible but if ptr is nullptr, I have to forward declare a pointer as
follow "T *n = nullptr" and then call Qt::totally_ordered_wrapper(n).
So I think adding reset(P) function is better.
Found in API Review.
Change-Id: I0acfcacc97a43f3cf8bfa65b2b16a65cae95b727
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
(cherry picked from commit dc49d84abc35dfbbc4028a5f86017786cff13ae7)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
Unlike the old Qt::compareThreeWay() overload for raw pointers, the new
overloads for the Qt::totally_ordered_wrapper didn't allow to compare
wrappers of different (but compatible) pointer types (base vs derived).
This patch fixes it.
Ideally the constraints on operators in Qt::totally_ordered_wrapper
should use std::common_type_t, but we hit a bug in VxWorks that
prevents us from using it, so simply demand that the types are
convertible pointers. For now that should be enough, considering
that Qt::totally_ordered_wrapper only expects pointers as wrapped
types.
Found in API Review.
Change-Id: I9f7184468bea3e1f2944ca5347f0b79eded2f4d3
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
(cherry picked from commit d73bdec05101b34745156c8cd4ea571f1223ad32)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
compareThreeWay() was supposed to be an operator<=>(), but for C++17.
The idea was that at some point when we unconditionally demand C++20,
people could just replace all the usages of compareThreeWay() with
operator<=>().
However, the Qt::compareThreeWay() overload for pointers is different
from the operator<=>() for pointers, because it is actually using
std::less{} or std::compare_three_way{} to do the comparison, thus
avoiding an UB. This is not bad as such, but can potentially lead to
UB when mass-replacing all compareThreeWay() calls with operator<=>().
To avoid this problem, deprecate the overload, and suggest to use
the Qt::totally_ordered_wrapper together with the respective overload
instead.
Found in API Review.
[ChangeLog][QtCore][QtCompare] Deprecate Qt::compareThreeWay()
overload for pointers.
Change-Id: I9c57871145dc3cb9656d6006db88b48a1553bef4
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
(cherry picked from commit b1ae4334ea11f6942c7ce37a80a3aa98a3447809)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
Provide the overloads as free functions in Qt namespace instead of
making them hidden friends of the Qt::totally_ordered_wrapper class,
because they should replace the to-be-deprecated overloads of
Qt::compareThreeWay() for raw pointers, so they should be
easily discoverable.
Also, we treat Qt::compareThreeWay() as a C++17 equivalent of
operator<=>(), so we need to have an overload for pointers (even
if it takes only the wrapped pointers).
Found in API Review.
[ChangeLog][QtCore][QtCompare] Added Qt::compareThreeWay() overloads
for Qt::totally_ordered_wrapper. These overloads do the comparison
using strict total ordering.
Change-Id: I2b5bc542c546330ca78c6284188c8167136a849e
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
(cherry picked from commit d6e1df3513c7d5b6e93d68583fc4c1febb7bcd85)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
Found more issues:
- quint8 and qint8 (signed and unsigned char) actually match an
integral overload (probably int/uint), and _not_ (char),
- char8_t (only in C++20 builds) and (at least on non-Windows) wchar_t
are incorrectly treated as numeric types,
- qfloat16 breaks if QFLOAT16_IS_NATIVE.
Created a few tasks to track these.
Task-number: QTBUG-126053
Task-number: QTBUG-126054
Task-number: QTBUG-126055
Pick-to: 6.7 6.5 6.2
Change-Id: Iaf64929e3fae257d5dd283f555bb71ea2908cc94
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
(cherry picked from commit 967736b96ba06bfdfdf553b2c3720d7adc4a921c)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
Makes sense to collect the stuff in a similarly-named header, and it's
not like any of the QTypeTraits stuff was needed for QTypeInfo.
Makes #include <qtypeinfo.h> _much_ lighter again, at the expense of
qminmax.h now getting <variant>, <tuple> and <optional> instead, but
qminmax.h is much easier to avoid in Qt headers (just use the std
versions) than qtypeinfo.h.
[ChangeLog][QtCore][Potentially Source-Incompatible Changes] The
qtypeinfo.h header no longer transitively includes <optional>, <tuple>
and <variant>.
Task-number: QTBUG-97601
Change-Id: Ied96113f38c1232fef3ec79847ee62f06c68f268
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
(cherry picked from commit b2eb422699118f4ae8370519b4c7bb3fe121beb2)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
Some compilers do not respect the
QT_WARNING_PUSH
QT_WARNING_DISABLE_DEPRECATED
...
QT_WARNING_POP
incantation, so rewrite the test to use `template <typename X>` as
an Attribute parameter of the macros.
This allows to get rid of annoying warnings when compiling
tst_qcomparehelpers.
Amends de16185068.
Change-Id: I0b3a0f4ad3189d39f5662ce128d5e5f427543393
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
(cherry picked from commit c6feed7e1d8b8ed991a81878c78bb361f038eaf6)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
The recent change to IANA ID lists extracted for Windows adds new IANA
IDs to various zones and shuffles the order in some cases, leading to
changes in offsets throughout the related tables (and two test-case
that lose an expected entry that was an alias for one of its
others). This is caused by CLDR doing a clean-up of its zone
information, deprecating some old names, combining some synonymous
zones and reversing some aliases to better match the IANA DB. As a
result, some prior expect-fail tests now pass.
The code that reads the tables is insensitive to order after the first
entry in each list; and the first entries only change where they were
CLDR's ID for a zone, replaced by the modern IANA ID, which is what we
should be using. Various aliases are omitted from lists, but can be
looked up via the alias table.
[ChangeLog][Third-Party Code] Updated timezone data from the Unicode
Common Locale Data Repository (CLDR) to v44.1 to match QLocale's data.
Change-Id: I8d89bef7370b93572a3dcf61f6713d4f21963e3d
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Linux kernel version 5.8 added support for the stx_mnt_id field in the
struct statx because "Systemd is hacking around to get it and it's
trivial to add to statx, so...". This allows us to much more neatly
match the lines in /proc/self/mountinfo.
The same kernel version added STATX_ATTR_MOUNT_ROOT so we can tell if a
given path is the mount point of a filesystem. We don't have a need for
that information for now.
We need to retain fallback code for two reasons: first, the user may be
running with an old Linux kernel, in which case we won't get the
STATX_MNT_ID bit set in stx_mask. Second, we may have failed to open()
the path in question, because the user may not have the necessary
permissions.
There's still a race condition because the mount IDs can be reused
immediately after something is unmounted. There's a 64-bit unique mount
ID (available since v6.8) but it's not reported in /proc/self/mountinfo,
so we couldn't us it right now. We can with 6.8's statmount().
Pick-to: 6.7
Task-number: QTBUG-125721
Change-Id: If3345151ddf84c43a4f1fffd17d3f7dbce4ff16b
Reviewed-by: Ahmad Samir <a.samirh78@gmail.com>
The test data is provided as float, so we must obey [conv.fpint]/1 when
converting from float to integer types. There are values in the rows
that are outside of the range of short and shorter integers, so just
check the range.
Fixes: QTBUG-125889
Pick-to: 6.7
Change-Id: If3345151ddf84c43a4f1fffd17d405ee0cd00d44
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
Provide the new comparesEqual() helper function as an implementation of
(in)equality operators.
Use QT_DECLARE_EQUALITY_OPERATORS_HELPER macro instead of
Q_DECLARE_EQUALITY_COMPARABLE for the comparison between QPointer<T>
and QPointer<X> to avoid creating the reversed version of the
operators in C++17 mode.
Use new \compares command in the documentation to describe the
comparison operators provided by QPointer.
Task-number: QTBUG-120306
Change-Id: Iff24d3ef5c790de7f06ab825f853e06186fa1471
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
This makes it possible for the stdCompatibility() test to cope with
the backend (based on the Windows Registry, which disagrees with MS's
ICU-based std::chrono::tzdb) successfully constructing a zone, by
finding a known alias, instead of failing because it doesn't know the
name. It may also be useful to client code when dealing with legacy
names.
Amended some existing tests to only expect what they now should and
added some tests specific to aliasMatches(). Also added some
explanative comments where it isn't needed.
Task-number: QTBUG-115158
Change-Id: I095bdbead78df339e29b29518d5010ef905fa8b2
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Follow-up to commit 69555b364d, using
QTimeZone::isTimeZoneIdAvailable() instead of trying to construct an
ad-hoc test for known unknonwn zone IDs.
Reopens: QTBUG-102187
Fixes: QTBUG-102187
Change-Id: I36ff4f1dfbc4035d73df5b8fd68376c1bc9641e1
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Also rename the subdirs in the hiddenDirs_hiddenFiles test dir,
normalDirectory/normalDirectory/ is a bit confusing.
Make the test more deterministic by comparing lists of files/dirs
instead of just counts.
Change-Id: I12fdb5428bbef8382d4ee591792d167abcd216cf
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This exercise found an inconsistency involving char16_t. Filed
QTBUG-125588 to keep track of this and decide how to deal with it.
Task-number: QTBUG-125588
Pick-to: 6.7 6.5
Change-Id: I259fecef34539e8841149570d7411d959b48d876
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
Extending the fromCharacter() tests to check for the construction to
be noexcept and implicit turned up a few bugs:
The constructor from char32_t was simply not marked as noexcept, even
though all its operations were. So just mark it noexcept.
The constructor from QChar, otoh, was never called, (certainly not for
rvalues), because QString happens to be constructible from QChar (and
QLatin1Char and QChar::SpecialCharactor), so due to perfect
forwarding, the if_convertible_to<QString> ctor won, allocating a
QString.
This is a regression of Qt 6.5 compared to Qt 6.4.
To fix this, exclude arguments that convert to QChar from matching the
if_convertible_to<QString/QByteArray> ctors, taking care to not match
those arguments that are already compatible_char<>s.
This, in turn, creates a problem for implicit QASV construction from
QLatin1Char and QChar::SpecialCharacter, because now that we've
excluded them from the if_convertible_to<QString> ctor, calling the
existing QChar or QString non-template constructors for these types
would require two user-defined conversions (from said type to
QChar/QString and from QChar/QString to QAnyStringView). That works
for explicit construction, but we need implicit convertability.
So bring out the big guns once more and add a perfectly-forwarding
ctor for anything convertible to QChar. QChar itself is actually
already handled by compatible_char<>, so the old QChar non-template
ctor can go. The extra copy of the QChar argument will be optimized
away by the compiler.
[ChangeLog][QtCore][QAnyStringView] Fixed a regression where
constructing a QAnyStringView from a char-like type (QChar,
QLatin1Char, ...) would first construct a QString and only then
convert that to QAnyStringView.
Amends 812a0d3125.
This change is forward and backward BC and SC, but not
behavior-compatible: certain operations detectably change
noexcept'ness, and some arguments may now cause the resulting
QAnyStringView to have a different encoding (though I really tried
hard to avoid that).
Since it's a regression, I proposed to pick this to the affected
branches, 6.7 and 6.5 (6.6 is already closed at this point).
Pick-to: 6.7 6.5
Fixes: QTBUG-125735
Change-Id: I86f37df5d80bee36db27e529c017cb73995a6831
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Łukasz Matysiak
Volker Hilsheimer
Oliver Wolff
Marc Mutz
Tor Arne Vestbø
Michał Łoś
Thiago Macieira
Liang Qi
Rym Bouabid
Ivan Solovev
Jari Helaakoski
Edward Welbourne
Ahmad Samir