If one destroys a running QThread, so far the behavior has been to crash
(à la std::thread) -- assuming the thread hasn't already signalled that
it has finished. This behavior is hostile to solutions such as using
QThread::create(), which always require a wait() before destroying the
thread object.
We can use the opportunity to change the behavior without breaking any
valid code. Instead of crashing, inside QThread's destructor we can ask
the new thread to quit, and then join it (à la std::jthread). This
simplifies the implementation of long-living runnables and the code that
manages them.
Deploying this solution for the whole QThread class may not be entirely
painless. While no correct code would work differently with the proposed
changes, incorrect code that deletes a running thread would no longer
crash "loudly" -- instead, it might deadlock "quietly", have memory
corruptions, etc.
Hence I'm limiting this approach to only the threads created by
QThread::create(), at least for the time being. This also side-steps
perhaps the biggest problem of generalizing the approach, which is that
placing such interrupt+join logic into~QThread's destructor would cause
it to be run _after_ a QThread subclass' own destructor has run,
destroying the subclass' data members too early. This might create
an antipattern if one chooses to subclass QThread. With create(), a
subclass in question exists, and it indeed has NSDMs, but it's entirely
under our control (in fact, I'm placing the logic just in its dtor).
[ChangeLog][QtCore][QThread] Destroying a QThread object created by
QThread::create() while the thread that it manages is still running will
now automatically ask that thread to quit, and will wait until the
thread has finished. Before, this resulted in a program crash. See the
documentation of QThread::~QThread() for more details.
Change-Id: Ib268b13da422e277ee3ed6f6c7b2ecc8cea5750c
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
The PRINT_2ARD_TEMPLATE macro expansion alone is responsible for about
50% of the compile time and RAM requirements of tst_qmetatype.cpp. By
factoring it into its own TU, we reduce the maximum memory load on my
machine from 4.0GiB to 2.5GiB, provided we don't parallelize the
build, then we take 0.5GiB more.
This is a quick-fix for the QNX build problems currently plaguing the
CI. Going forward, we should probably have a better solution, whatever
that may be.
Task-number: QTQAINFRA-4669
Change-Id: I2732b4c25b369b15cce1c7afe222d041ecb6795a
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
it avoids global conan configuration in conan.conf
or via environment variables
Pick-to: 6.2
Change-Id: Ia54cbfe95e0864e7107ea3bb53ff9ba4f1df7b60
Reviewed-by: Iikka Eklund <iikka.eklund@qt.io>
This reverts the QtWidgets changes made in commits
22634e0079 and
5b09346cf4 while keeping the auto tests
introduced in the former commit.
Both commits introduced rendering errors when moving widgets out of
or into areas in which they are obscured. Before we apply any further
optimizations to this code we need thorough auto test coverage.
Task-number: QTBUG-98151
Task-number: QTBUG-26269
Pick-to: 6.2
Change-Id: I9cb82b73776daed59ea0e9f51ff7ddef1c7265b6
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
remove(i) is the same as remove(i, 1), and the extra 'n' argument is
of trivial type, so it's ok to default it instead of overloading.
Change-Id: Id926cd63fde518e002684a41e055edc1004247a4
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
... so it can be used by multiple .cpp files.
Task-number: QTQAINFRA-4669
Change-Id: I7212b9b08cd3bfa44ee741ee4789d1d0024e4708
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Previously, only the printf-style overloads were marked as such.
Saves ~7KB in QtCore text size on Linux AMD64 GCC 11 C++ 20 builds.
Change-Id: I4ed480e48060b87968f3d15bb5a84bdbcf70a647
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Currently the compositor draws the Qt logo by default;
check if the window has an icon and draw that instead
if set.
Fixes: QTBUG-86052
Pick-to: 6.2
Change-Id: Ia4f9c03562c15993c86cb8717f27e47ab669353d
Reviewed-by: Lorn Potter <lorn.potter@gmail.com>
Factor the assertions relating to sizes and indexes into an inline
function, verify(). This hopefully reduces the string data generated
for assertions to the bare minimum.
Change-Id: Iad5c2e587ad189e7e1a62b9ca61ada4282c44f51
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
When deleting the last item in a chain, without it being the last item
in the chain, then we re-use the iterator which was passed in as an
argument. This is wrong if we detached earlier in the function, and
means we return an iterator to the previously shared data.
Pick-to: 6.2
Change-Id: I7da6309e23a32073da59e7da0cbfd1d16734f1ca
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
Last pieces of code guarded by these defined were cleaned up back in 2016.
Task-number: QTBUG-98650
Change-Id: I52d971c2b476b7d8d362be7807b71b71e657c70b
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Open the file only if matching on content is needed.
Use QFileInfo::filePath() instead of QFileInfo::absoluteFilePath() in
QMimeDatabase::mimeTypeForFile(). filePath() does much less work, and so
is faster. Thiago Macieira helpfully explained in a review comment why
the absolute path is not useful for correctness here: "Nothing needs
absolute paths within the same application that would resolve the
relative path to absolute. You only need an absolute path if you're
communicating with another application that may be in a different
directory."
QMimeDatabase::mimeTypeForFile() checks fileInfo.isDir(), so the
fileName.endsWith(QLatin1Char('/')) check in
QMimeDatabasePrivate::mimeTypeForFileNameAndData() was redundant when
called from this function. The other two callers of that function now
check this condition before opening IO devices. This improves
performance of the two QMimeDatabase::mimeTypeForFileNameAndData()
overloads in the corner case.
Refactor and optimize QMimeDatabasePrivate::findByFileName() and its
usages. Previously each caller constructed a QFileInfo object and passed
QFileInfo::fileName() into this function. Now the callers simply pass an
absolute or relative path to a file into this function, which then uses
QFileSystemEntry::fileName() to exclude the path. Constructing QFileInfo
is relatively expensive, so this change slightly improves performance.
Optimize QMimeDatabasePrivate::loadProviders() by calling static
QFileInfo::exists() instead of constructing a QFileInfo object and
calling the non-static QFileInfo::exists() overload. Note that the
QFileInfo object was always created, even if QFileInfo::exists() under
an `if` and an `#if` was never called.
The following table contains the average results of the added benchmark
tst_QMimeDatabase::benchMimeTypeForFile() on my GNU/Linux system before
and at this commit. The numbers denote milliseconds per iteration.
data row tag before at
MatchDefault:
archive 0.029 0.016
OpenDocument Text 0.029 0.015
existent archive with extension 0.039 0.025
existent C with extension 0.033 0.020
existent text file with extension 0.033 0.020
existent C w/o extension 0.076 0.074
existent patch w/o extension 0.11 0.105
existent archive w/o extension 0.069 0.066
MatchExtension:
archive 0.012 0.0115
OpenDocument Text 0.0115 0.011
existent archive with extension 0.017 0.016
existent C with extension 0.011 0.011
existent text file with extension 0.011 0.011
existent C w/o extension 0.016 0.0155
existent patch w/o extension 0.013 0.012
existent archive w/o extension 0.013 0.012
MatchContent:
archive 0.019 0.012
OpenDocument Text 0.019 0.012
existent archive with extension 0.053 0.051
existent C with extension 0.056 0.0545
existent text file with extension 0.058 0.056
existent C w/o extension 0.0605 0.059
existent patch w/o extension 0.10 0.099
existent archive w/o extension 0.057 0.054
Change-Id: Idb541656e073a2c4822ace3f4da412f29f2351f8
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: David Faure <david.faure@kdab.com>
Amend 97cfd49401, which resulted in
incorrectly laid out tables in Text elements.
Reported by the baseline test for the scenegraph. Not reproducible using
a QtWidget text widget, so no unit-test added here.
Pick-to: 6.2 5.15
Task-number: QTBUG-86671
Task-number: QTBUG-97463
Change-Id: I607b1fea3fb5923f81f05a3646bafebec3dcf3b5
Reviewed-by: Eirik Aavitsland <eirik.aavitsland@qt.io>
This reverts commit 064c3d35e6.
Reason for revert: Causes crashes on querying. Better to endure the non-NULL representations of null QDateTime values.
Change-Id: I33dd3b95ab35d8e3accb864aec13d21764b1270d
Reviewed-by: Andy Shaw <andy.shaw@qt.io>
After the windows font engine was no longer marking everything as
scalable we started limiting the font size of requests to the maximum
of Courier when it was requested. This was a regression from 5.8 and not
in agreement with our documentation.
The problem is that we would only make the switch from Courier to
Courier New after having already gone through the foundry-lookup and
found a closest-available font size for Courier.
With this sanitization step in the backend we can make these changes
early enough that we haven't yet adjusted e.g. the font size.
Pick-to: 6.2 5.15
Fixes: QTBUG-58995
Change-Id: I319e93e6b78c7c3c5539964ac5ab4e05f8902ab6
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@qt.io>
- GHS compiler is not fully compliant with iec559. Therefore we need
to update is_iec559 checking for GHS case.
Pick-to: 6.2
Change-Id: Ia094509d26bf5f0109f2937547a056267019cffb
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Janne Koskinen <janne.p.koskinen@qt.io>
When running a test using -callgrind, we recurse into a child process,
run under valgrind, to which we pass -callgrindchild; and we only want
the output from the child process, since the parent won't actually be
running any tests. We also won't be using the global data table for
the test in the parent process. So bypass the set-up and tear-down of
both logging and the global data table in the parent process.
Prior to commit 3ee6d8d336, these parts
of the set-up and tear-down were skipped in the callgrind parent
process, but that refactoring split qExec() up into qInit(), qRun()
and qCleanup() to enable QtQuick to recombine them to implement an
equivalent of qExec(), calling qRun() once for each built-in style. It
needs these pieces of set-up to happen in qInit(), and of tear-down in
qCleanup(), to avoid repeating them for each style. Leave a comment in
qExec() that might help future readers to understand why it's done the
way it is.
Change-Id: Ieaca9a125c713b8fcf8dec8f9be0c024a798d504
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Its twoHundredMillionInstructions() test has #if-ery to limit it to
gcc and x86; however, it was testing only __i386 for the x86 part,
where gcc defines __x86_64 instead on modern 64-bit systems. In the
process, invert the condition and the branches it controls - positive
tests are easier to comprehend.
Change-Id: I8e906c606c48aa5034e02e3ed5d042fbb1f2ecbc
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Development web server for web applications. Supports
http and https. Sets COOP and COEP headers.
The web server script supports certificate generation using
mkcert (github.com/FiloSottile/mkcert). Briefly, mkcert
supports generating server certificates for the current
local ip address(es), using a certificate authority
which can be installed on devices where the app should
run.
The COOP and COEP headers are required to enable the
SharedArrayBuffer API, which is required for Emscripten’s
pthreads implementation.
The server serves the current directory on localhost
by default. Use the “-a” argument to bind to an additional
address, or "--all” to bind to all IPv4 addresses found on
local network interfaces. Change the directory by passing
it as a positional argument.
Task-number: QTBUG-79087
Change-Id: Id0cba649e42af53ed8106e336a77e78398bcf901
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: David Skoland <david.skoland@qt.io>
We can expect that users use Qt 6 to try out Qt 6 examples.
Pick-to: 6.2
Change-Id: Ifa7bd9471a712fa1d63107b7ff2d392b7aefdcfe
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
Currently the compositor draws the Qt logo by default;
check if the window has an icon and draw that instead
if set.
Fixes: QTBUG-86052
Pick-to: 6.2
Change-Id: Ia4f9c03562c15993c86cb8717f27e47ab669353d
Reviewed-by: Lorn Potter <lorn.potter@gmail.com>
It's a precondition and we might just as well check it, given that
people have actually got the order of the arguments wrong (cf.
QTBUG-69330).
Unfortunately, Q_ASSERT is defined below qBound in qglobal.h, so I had
to reshuffle some code around.
Change-Id: I82e52bcb863ff8c96594817e0cd5d7d2abe2e57e
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
A default constructed QPageLayout has an invalid QPageSize. When
asking for its size, an invalid QSize is returned (-1, -1). This
size is then used to calculate the maximum margins the page layout
can have, resulting in negative margins.
I'm not sure if that makes sense for QPageLayout (in principle, a
negative margin is acceptable, meaning to "grow" the size), but then
this margin is passed as an upper bound to a series of qBound calls in
QPageLayoutPrivate::clampMargins.
To fix this, change the calculations of the maximum margins by using
a lower bound of 0.
Change-Id: I429104acfb2296d9eb1ee54c113e9d7e22d9b6ab
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
If a QListView's model is reset to an empty one, its columnCount()
below the root is going to be 0. Therefore, the code was doing a
qBound(0, d->column, -1)
which is meaningless (high < low). Instead, do the two logical
operations explicitly: first do an upper bound on d->column
(using qMin) and then lower bound the result by 0 (using qMax).
The code worked by chance, because 0 was eventually the correct
number to use as a bound for d->column.
Change-Id: Ic32077cdab01eaa715137c05ed1f9d66c8eb2f67
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
These are left-overs from the initial qmake2cmake conversion.
Pick-to: 6.2
Change-Id: Ie15c9ff022ea4566d10c1ba74599de9af83d29a7
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
[ChangeLog][QtCore][QStringBuilder] Added support for QByteArrayView.
Change-Id: If2c23549d533dd31c320f3ee455fcd01ea5b460a
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
[ChangeLog][QtNetwork][QNetworkRequest] Added
QNetworkRequest::Http2CleartextAllowedAttribute which controls whether
HTTP/2 cleartext (h2c) is allowed or not. The default is false. This
replaces the QT_NETWORK_H2C_ALLOWED environment variable.
Task-number: QTBUG-98642
Change-Id: I43ae1cc671788f6d2559cd316f6667b412c8e75e
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Visual Studio 2019 introduced the dubbed FH4 feature
which can make C++ exception handling smaller on x64.
According to the article [1], it's enabled by default
for UWP applications, and Microsoft also use it in
their own widely-known commercial products such as
Office to reduce the binary size.
So make use of this feature for Qt when possible, to
get smaller binary.
As a drive-by, add "/EHs-c-" explicitly to the flags
when we want to disable C++ exception handling.
[1] Official article that introduces dubbed FH4:
https://devblogs.microsoft.com/cppblog/making-cpp-exception-handling-smaller-x64/
Change-Id: I2e3330de477f78372cf7903d0ef7a732b09552a9
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: André de la Rocha <andre.rocha@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Oliver Wolff <oliver.wolff@qt.io>
The new overload allows creation of files with non-default permissions.
This is useful when files need to be created with more restrictive
permissions than the default ones, and removes the time window when
such files are available with less restrictive permissions.
[ChangeLog][QtCore][QFile] Added QDir::open() overload that
accepts permissions argument.
Fixes: QTBUG-79750
Change-Id: Iddfced3c324e03f2c53f421c9b31c76dee82df58
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
The new argument allows atomic creation of files with non-default
permissions.
Task-number: QTBUG-79750
Change-Id: I4c49455b41f924ba87148302c8d0f77f5de0832b
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
After 556511f9f3, which moved the resolve
mask storage into the palette's d-pointer, modifying the resolve mask
requires a detach. As of now, we only detached when setting a different
brush, but always modified the resolve mask, which broke palettes that
shared the d-pointer (likely the global default palette).
However, detaching has negative side effects when styles set brushes on
temporary palette objects and then use that palette object's cache key
to build a cache of pixmaps. As each drawing would detach the palette
(even if the palette doesn't change, which is likely), the cache key
changes with each detach, and the cache would quickly increase in size.
This was addressed in changes d7bcdc3a44
and 1e75dcf251.
We can either detach and find other ways to address the issues from
QTBUG-65475, or we can not change the resolve mask when the brush doesn't
change and completely ignore the call.
Since QFont ignores the setting of any attribute to a value that is
identical to the current value, and since it's possible to force that
the resolve-bit is set by calling setBrush twice with different brushes,
ignoring the call seems like the better solution.
[ChangeLog][QtGui][QPalette] Setting a brush on a palette that is
identical to the current brush no longer sets the resolve mask bit for
that particular role, so items using the palette will continue to
inherit changes from parent items.
Fixes: QTBUG-98762
Task-number: QTBUG-65475
Pick-to: 6.2
Change-Id: Ife0f934b6a066858408ef75b7bb7ab61193ceb47
Reviewed-by: Simon Hausmann <hausmann@gmail.com>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Allows changing the font used to display a font family
[ChangeLog][QtWidgets][QFontComboBox] Added the setDisplayFont() function, in order to be able to control the font used to render the font name and sample text (when previewing the fonts).
Change-Id: I94bfef43142c5346237e3069449bd19dbacb7420
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@qt.io>
Allows setting the sample text to be used
[ChangeLog][QtWidgets][QFontComboBox] Added the setSampleText() function, in order to be able to control the sample text displayed by the combobox (when previewing the fonts).
Change-Id: Iedcd3bb72bc6aec9a178f14f7fbbe324ba0223ef
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@qt.io>
Apparently this code is needed on other platforms now, not only macOS.
Pick-to: 6.2
Fixes: QTBUG-98504
Change-Id: Ie0a7e38609e8fc8c11915784dd3652a3517bb639
Reviewed-by: Paul Olav Tvete <paul.tvete@qt.io>
Reviewed-by: Jan Arve Sæther <jan-arve.saether@qt.io>
Only use shellQuote() if the path is actually passed to a native API.
For QFile API, use the unquoted path.
Fixes: QTBUG-97649
Pick-to: 5.15 6.2
Change-Id: I9d8131819010bbd2faa8a81eef367245d90a767f
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
addApplicationFont and addApplicationFontFromData basically do the same
thing (with one reading the data from a file and another reading
it directly from QByteArray)
The documentation is not quite identical, so fix that by using \include
magic of qdoc
As a drive-by, remove outdated \note within addApplicationFont() description
Fixes: QTBUG-98752
Pick-to: 5.15 6.2
Change-Id: I29738e7411d241c26a4b2e1dd91d0af867fcc111
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@qt.io>
Says GCC -std=c++20:
src/gui/util/qlayoutpolicy_p.h:127:30: warning: bitwise operation between different enumeration types ‘QLayoutPolicy::Policy’ and ‘QLayoutPolicy::PolicyFlag’ is deprecated [-Wdeprecated-enum-enum-conversion]
127 | if (verticalPolicy() & ExpandFlag)
| ~~~~~~~~~~~~~~~~~^~~~~~~~~~~~
Fix by making Policy a QFlags<PolicyFlag>, which we can do because the
class is private. In QSizePolicy, this would break BC.
Since a QFlags cannot be opened for constants like an enum, we need to
define the ex-Policy-enumerators as static inline constexpr objects
instead.
Pick-to: 6.2
Change-Id: I29bc938f86508deed3f99ad9d1c1892547206c05
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Less noisy compiler error messages.
[ChangeLog][Potentially Source-Incompatible Changes][QSizePolicy]
qHash() is now a hidden friend and can only be called by unqualified
(qHash(sp)), not by qualified lookup (as in, say, ::qHash(sp) or
QT_PREPEND_NAMESPACE(qHash)(sp)).
Task-number: QTBUG-98863
Change-Id: I37d26c661c2d8bb74350eb9378bd19fa426678d2
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
This can be useful for some cases when the Android APIs have calls
that expects a FileDescriptor instead of a Uri or an int file
descriptor (like a case in Qt Multimedia with MediaRecorder).
Pick-to: 6.2
Task-number: QTBUG-96081
Task-number: QTBUG-96957
Change-Id: I0ab8d37a43b7cb94f6ebb5d48014e5a7903aadc7
Reviewed-by: Ville Voutilainen <ville.voutilainen@qt.io>
Reviewed-by: Rami Potinkara <rami.potinkara@qt.io>
The method getUriWithValidPermission() now returns null only if the
Uri couldn't be parsed, otherwise, returns the parsed Uri and fail later
when used if it doesn't have permissions for example.
Also, Don't use use hardcoded strings for method names, and print the
exception message instead of the whole stack.
Pick-to: 6.2
Task-number: QTBUG-96081
Task-number: QTBUG-96957
Change-Id: If7ba88265c9683e6660cb483cfee6c22016f2976
Reviewed-by: Ville Voutilainen <ville.voutilainen@qt.io>
Reviewed-by: Rami Potinkara <rami.potinkara@qt.io>
QApplication hides the fact that the reason is never set by several
QPA plugins, but Quick items don't receive the correct reason on
Windows, Android, the offscreen plugin, and other platforms.
Add relevant scenario to the QFocusEvent test case, and fix the
plugins to always set the focus reason when handling window activation
changes. Exclude the minimal plugin from the test, it seems largely
unmaintained anyway.
Task-number: QTBUG-75862
Change-Id: I5404a225b387fc9a3851b6968d0777c687127ed1
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Giuseppe D'Angelo
Marc Mutz
Eric Lemanissier
Friedemann Kleint
Volker Hilsheimer
Morten Johan Sørvig
Mårten Nordheim
Ivan Tkachenko
Igor Kushnir
Ulf Hermann
Edward Welbourne
Tatiana Borisova
Gatis Paeglis
Kai Köhne
Dmitry Shachnev
Shawn Rutledge
Tor Arne Vestbø
Yuhang Zhao
Ievgenii Meshcheriakov
Albert Astals Cid
Andrei Golubev
Assam Boudjelthia
ChunLin Wang