../../src/corelib/io/qurlidna.cpp: In function ‘QString qt_ACE_do(QStringView, AceOperation, AceLeadingDot)’:
../../src/corelib/io/qurlidna.cpp:2543:23: error: ‘int __builtin_memcmp_eq(const void*, const void*, unsigned int)’
reading 8 bytes from a region of size 2 [-Werror=stringop-overflow=]
if (memcmp(result.constData() + prevLen, acePrefixUtf16, sizeof acePrefixUtf16) == 0)
~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1plus: all warnings being treated as errors
In function ‘bool operator==(const QByteArray&, const QByteArray&)’,
inlined from ‘virtual void (* QLinuxFbIntegration::platformFunction(const QByteArray&) const)()’
at ../../src/plugins/platforms/linuxfb/qlinuxfbintegration.cpp:185:18:
include/QtCore/../../../../src/corelib/text/qbytearray.h:571:45: error:
‘int __builtin_memcmp_eq(const void*, const void*, unsigned int)’ reading 17 bytes from
a region of size 1 [-Werror=stringop-overflow=]
The warnings/errors are bogus. Fix them by using QStringView::sliced() and de-inlining the
comparison operator for QByteArray.
Change-Id: I24956fe74a7989e75cd03d717570b8fca493ab23
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
After API discussions, agreement was that from(n) is a bad name
for the method. Let's go with sliced(n) instead.
Change-Id: I0338cc150148a5008c3ee72bd8fda96fb93e9c35
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This keeps the API symmetric with what we have in our string
classes.
Change-Id: I94c5b39b718ca2472f9ca645e7a42e4314636f67
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The change creates a slight source incompatibility. The main
things to take care of are
* code using printf statements on list.size(). Using qsizetype in
printf statements will always require a cast to work on both 32
and 64 bit.
* A few places where overloads now get ambiguous. One example is
QRandomGenerator::bounded() that has overloads for int, uint and
double, but not int64.
* Streaming list.size() to a QDataStream will change the format
depending on the architecture.
[ChangeLog][QtCore][QList] QList now uses qsizetype to index into
elements.
Change-Id: Iaff562a4d072b97f458417b670f95971bd47cbc6
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Also adjust qCalculateBlockSize() to be able to handle large
allocations.
QVector::length() is currently still limited to 2G items, that will
get changed in a later commit.
Change-Id: I3a92fbfd7f281d30844c5fafa3b9a474bc347c19
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This avoids ambiguities in our API when someone e.g. writes
vector.insert(0, ...).
It requires a slight workaround in qlalr, where std::search()
for libc++ doesn't like that our difference_type is qsizetype.
Change-Id: I40aa1040781ffbdd12d04410078207969b3bde53
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This is a next step towards making QList, QString
and QByteArray able to deal with large sizes.
Change-Id: Icad49b33f503401ac4912678b2f88584c6f91a63
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Remove the last places where those got used and avoid
allocations when we resize to 0.
Change-Id: Ib553f4e7ce7cc24c31da15a55a86d18bdf1cc5c3
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Those can simply be handled as compile time constant strings
pointing to the empty (Q)Char.
Change-Id: I1f6f6ab923a30c68a720003ca68c34c572aa29da
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
As a side effect, data() can now return a nullptr. This
has the potential to cause crashes in existig code. To work
around this, return an empty string from QString::data()
and QByteArray::data() for now.
For Qt 6 (and once all our internal issues are fixed), data()
will by default return a nullptr for a null QString, but we'll
offer a #define to enable backwards compatible behavior.
Change-Id: I4f66d97ff1dce3eb99a239f1eab9106fa9b1741a
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
In case of forward iterators, call std::distance just once and not
twice. In case of non-forward iterators, don't call
reserveIfForwardIterator -- as the name says, it doesn't make sense
on non-forward iterators.
Change-Id: I7e6a603205286c05f7bc7c47fd1f1e0d92705b20
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
The hand-written special member functions did exactly what the
compiler generated ones would do anyhow.
Change-Id: I66439178460d30957135aac44680dd3109ada62a
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
The iterators are quite heavy objects (>100bytes), don't pass them
by value.
Change-Id: I4c9d1f64d14419a35bd067884d7e8bca2589f9b9
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Properly use the new QStringConverter API and not an internal
qFromUtfEncoded method that was buggy after the changes.
Take the oppportunity to clean up and remove qFromUtfEncoded, as
QClipboard was its only user.
Fixes: QTBUG-85417
Change-Id: I8540d12056bf3f448c1f628ce0bd0ad462a6447d
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
Amends dcdfb6908d which failed to take the
workaround in destroy_current_thread_data into account.
Since pthread_getspecific was completely replaced with the thread_local variable
currentThreadData, the workaround has no effect anymore. Therefore we need to
replace it with a workaround that makes sure currentThreadData is set inside of
the destructor function.
This prevents a leak, where QThreadPrivate::finish() tries to access the
thread data, but since it already is null, recreates it without ever deleting it.
Pick-to: 5.15
Change-Id: I3811d262a411a6bde9d6eb90f8d17e0bbc5de657
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This change only happens to files touched
by the commit to add missing ; to Q_UNUSED.
Task-number: QTBUG-82978
Change-Id: I10e6993a2bb3952cf9a262708b8573550e0dbe63
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
Also add a ; where it is missing.
Task-number: QTBUG-82978
Change-Id: Ic5d2a07363c25ab641d234baca89bc62238458cb
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
There's a bunch of similar and overlapping logic in QCocoaKeyMapper
already. Moving it to the same place allows us to easier find ways
to reduce the overlap.
None of the exported functions were used outside of the plugin.
Change-Id: I6953690cdfda5ee8265b33ccbf919184c3a1700f
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
It looked a lot like it needed an update to its Unicode data (in
tables and functions) but Thiago tells me this would be misguided,
although we do need an upgrade to IDNA 2008, at some point.
So document why this doesn't get updated along with UCD.
Task-number: QTBUG-85371
Task-number: QTBUG-85323
Change-Id: I764667db9c24bf05371e8a3c2601ccbf48f99711
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
No client of QDateTimeParser actually uses it unless datestring was
enabled, nor is it any use without datestring. Various methods
conditioned on datestring are broken unless datetimeparser is enabled.
We can't condition public API on datetimeparser, as it's a private
feature, but client code can condition use of it on the private
feature. All string-to-date/time conversions that use a string format
(this includes all locale-specific formats) depend on feature
datetimeparser.
Change #if-ery (or add it) in all client (including test) code to test
the right feature.
Tidied up some code in the process. Killed some already-redundant
textdate #if-ery. Renamed a test whose name claimed it involved
locale, which it doesn't, in the course of #if-ing it.
This simplifies the condition for feature datetimeedit (which overtly
depended on textdate, redundantly since it depends on datestring which
depends on textdate; its dependence on datetimeparser now makes its
dependency on datestring also redundant).
It also removes the need for assorted datestring checks in
QDateTimeParser itself.
Change-Id: I5dfe3a977042134b2cfb16cbcc795070634e7adf
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
And only implement it for QPodArrayOps, as that's the only case where
we should be using it.
Change-Id: If48f3e4b142c322d3451309d6d1cf68aee569ea2
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The change introduced crashes in some tests that only surfaced
in certain CMake configurations.
This reverts commit 76c3eee402.
Pick-to: 5.15
Task-number: QTBUG-85357
Change-Id: Ief93aa41e2d487d73b879133e7df0fd5ce0451bd
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
The API is available by including qopenglcontext.h as usual,
but scoped in the QPlatformInterface namespace. The namespace
exposes platform specific type-safe interfaces that provide:
a) Factory functions for adopting native contexts, e.g.
QCocoaGLContext::fromNative(nsContext, shareContext);
b) Access to underlying native handles, e.g.
openGLContext->platformInterface<QCocoaGLContext>->nativeContext()
c) Platform specific functionality, e.g.
static QWGLContext::openGLModuleHandle()
openGLContext->platformInterface<QEGLContext>->doSomething();
The platform interfaces live close to the classes they extend,
removing the need for complex indirection and plumbing, and
avoids kitchen-sink modules and APIs such as the extras modules,
QPlatformFunctions, or QPlatformNativeInterface.
In the case of QOpenGLContext these platform APIs are backed
by the platform plugin, so dynamic_cast is used to ensure the
platform plugin supports the requested interface, but this is
and implementation detail. The interface APIs are agnostic
to where the implementation lives, while still being available
to the user as part of the APIs they extend/augment.
The documentation will be restored when the dust settles.
Task-number: QTBUG-80233
Change-Id: Iac612403383991c4b24064332542a6e4bcbb3293
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
QPluginLoader does not call unload() on the QLibraryPrivate without an
explicit call to QPluginLoader::unload(). QFactoryLoader should behave
the same. We want to avoid unload() if possible as we generally don't
unload plugins and the resulting behavior varies between platforms. In
particular, macOS does make the address space of libraries only the
plugin links to inaccessible on close() even if RTLD_NODELETE is given.
For code that actually wants to unload(), an explicit function to do so
could be added. As QFactoryLoader is private API, there is no need to do
that until such a case is found.
Change-Id: I4e57259a9dcb4ceb60dfbfeda55abc0b995f436a
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
The qmake we're building is always built for the target platform.
Therefore, QT_CONFIGURE_CROSSBUILD can always be set to a falsy value.
Change-Id: I0f03c4ce0c75d3b4e97be5141adf742276131dcb
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
idx has already been tested for being >=0, so it's pointless
retesting it.
Change-Id: I2f5d7e1b7a70097de2601c1ed83752f6aa707cd9
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The insert() overloads have generally a very wide contract. The very
next line accepts negative positions, so remove the related assert.
Change-Id: I89b67615c59287825942047a28572bf896cf30e3
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
... and add include for it.
Amends ffb73175e6
Change-Id: I709a5aed13f6f62017b9e4116a03a4dfaae4bb13
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
This will never work, not unless libc implements it
themselves, since the child process is not allowed to return
from the function that does the vfork(), as subsequent use
of the stack would trash the frozen parent's return address,
and in our case that's syscall(). Instead, we may add a
vforkfd() function that takes a callback function that will
be called in that context, like the glibc clone(3) wrapper
does.
Pick-to: 5.15
Change-Id: I1dba29bc0f454df09ca1fffd161800b453c00593
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
It's supposed to return the same as toLocalFile(), for local files,
which means passing QUrl::FullyDecoded just like QUrl::toLocalFile()
does.
But a few code paths were testing component formatting options without masking
other FormattingOptions like RemovePassword, so this had to be fixed.
Fixes: QTBUG-84594
Change-Id: I82f15148b6d93516200f9ad6258d474e7f10924a
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Omitting state machine and docs for now.
Task-number: QTBUG-84469
Change-Id: Ibfa5e7035515773461f6cdbff35299315ef65737
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
Lars Knoll
Giuseppe D'Angelo
Jarek Kobus
Lars Schmertmann
Mike Achtelik
Paul Wicking
Thiago Macieira
Tor Arne Vestbø
Edward Welbourne
Alexandru Croitor
Ulf Hermann
Kai Koehne
Joerg Bornemann
Samuli Piippo
Nico Vertriest
David Faure
Albert Astals Cid