When Daylight Time transtion goes from Standard Time to Daylight Time
there is a "missing" hour, i.e. at 2am CET the clock goes forward to
3am. Currently QDateTime ignores this gap and considers the time to be
valid and able to be manipulated. This change respects the transition,
so any time set in the missing hour is considered invalid, and any date
maths returns valid results.
The validity in the current time zone needs to be checked every time
isValid() is called in case the system time zone has changed since the
last time it was checked. This is done by calling mktime to check the
returned result matches the expected result. This could be very
inefficient, but the returned offset value is cached each time so
mktime is not required to be called again within each method call,
effectively meaning mktime is called the same number of times by
each method. Note that this means any new methods added must be
careful to ensure either isValid() or refreshLocalTime() is called
first by any method needing to use the UTC value.
[ChangeLog][QtCore][QDateTime] The Standard Time to Daylight Time
transition for Qt::LocalTime is now handled correctly. Any date set
in the "missing" hour is now considered invalid. All date math results
that fall into the missing hour will be automatically adjusted to a
valid time in the following hour.
Change-Id: Ia652c8511b45df15f4917acf12403ec01a7f08e7
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Change from storing the date and time as QDate and QTime to a serialised
msecs format. This format is a direct translation of the QDate and
QTime values, it is not the actual msecs since the Unix epoch. This
msecs format ensures we are always able to recreate the original QDate
and QTime values, but should still simplify the code and improve
performance.
Because we no longer store the explicit date and time we need to store
their isNull()/isValid() status separately.
The changes in storage results in the same memory footprint as before.
Note that this change does not optimize the code nor set out to fix the
known bugs, it only seeks to maintain the current behavior, although
some bugs are fixed implicitly. More bug fixes and optimizations will
follow.
[ChangeLog][Important Behavior Changes] The supported date range in
QDateTime has been reduced to about +/- 292 million years, the range
supported by the number of msecs since the Unix epoch of 1 Jan 1970
as stored in a qint64, and as able to be used in the
setMSecsSinceEpoch() and toMSecsSinceEpoch() methods.
Change-Id: I98804d8781909555d3313a3a7080eb8e70cb46ad
Reviewed-by: Sérgio Martins <sergio.martins@kdab.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
On linux using QFileSystemWatcher with inotify backend when a watched
file is moved and added again to the watched files its path is not
replaced with the new one. This behavior prevents the emission of
the fileChanged signal with the wrong file path.
Task-number: QTBUG-33211
Change-Id: Ib45d8efdf5afbf8b8f6b4b26e43f3d6ee740aca6
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Similar to commit 8b062418 (MetaType: Fix operator{+,-}(int)
with the type-erased const_iterators., 2013-09-11), explicitly
create a copy of the iterator and intialise it.
Change-Id: I8b9edef40ca00c826f72768cba4a0992e55371f8
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@digia.com>
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Up to now, it was only possible to connect to functors in a direct
way, without being capable of using Qt::ConnectionType. This patch
allows for specifying a receiver for Functors and function
pointers, hence making it possible to specify effectively the
connection type.
To do this properly, it was needed to add an enum in FunctionPointer
representing whether the considered function is a member function
or not, to reduce ambiguity upon overloaded calls.
Moreover, now senders are checked for the existence of a slot obj
as well. This way, should the context be freed, the slot obj and
the functor contained in it will be freed as well.
On a side note, connecting to a static slot
(like QCoreApplication::quit) specifying the receiver object is
now compiling.
Change-Id: I46474099413b1dc6ca4db9934191d469baeef070
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
For easy cross-platform and cross-device UIs, automatic asset swapping
based on filename is being developed. This API provides the logic for
the swapping, so that applications can use it themselves with the same
logic as any automatic swapping done in application templates.
Selector set is initially minimal, aiming for just platform selection
and enabling a common selection mechanism for Qt platforms to use.
Change-Id: I219517d740fa7385e923a9e09cb7e241378f857a
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The responsibility of sendWindowSystemEvents() is to process events from
the window system. Historially that logic was part of the QPA/QWS event
dispatcher, which naturally also sent posted events. Through refactoring,
the code at some point ended up in in the QWindowSystemInterface class,
still with the posting of events in place.
This resulted in QPA event dispatchers adopting a pattern of just calling
sendWindowSystemEvents(), as that would cover both posted and window system
events. Other event dispatchers would call sendWindowSystemEvents(), and
then use a base-class implementation from QtCore for processing events,
resulting in two calls to QCoreApplication::sendPostedEvents() per
iteration of processEvents(). This breaks the contract that processEvents
will only process posted events that has been queued up until then.
We fix this entanglement by removing the sendPostedEvents() call from
QWindowSystemInterface::sendWindowSystemEvents() and move it to the
respective event dispatchers. For some EDs it means an explicit call
to sendPostedEvents, while others were already doing sendPostedEvents
though a separate source (GLib), or using a base-class (UNIX/BB), and
did not need an extra call.
We still keep the ordering of the original sendWindowSystemEvents()
function of first sending posted events, and then processing any
window system events.
Task-number: QTBUG-33485
Change-Id: I8b069e76cea1f37875e72a034c11d09bf3fe166a
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
The old code smply copied 100 shorts from the pointer passed into the
ushortarray ctor, regardless of the actual bounds of the original array.
Fix by making the ctor take the array by deference, deducing the size
as a template parameter, and only copying that much.
Fixes asan trace:
==18660==ERROR: AddressSanitizer: memcpy-param-overlap: memory ranges [0x7fff3c56de00,0x7fff3c56dec8) and [0x7fff3c56dd60, 0x7fff3c56de28) overlap
#0 0x457161 in memcpy asan_interceptors.cc:330
#1 0x4c40fe in ushortarray::ushortarray(unsigned short*) qtbase/tests/auto/corelib/io/qurlinternal/tst_qurlinternal.cpp:62
#2 0x4b0437 in ushortarray::ushortarray(unsigned short*) qtbase/tests/auto/corelib/io/qurlinternal/tst_qurlinternal.cpp:63
#3 0x47b643 in tst_QUrlInternal::idna_testsuite_data() qtbase/tests/auto/corelib/io/qurlinternal/tst_qurlinternal.cpp:119
...
Change-Id: Ie497bc8d337bc680a562482ca71ace535797ffb3
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
'val' was set to
QList<...>() << ... << ...
further down.
Fixes a Clang warning.
Change-Id: I5fe80d87dbe2c1d50652dfd7b6c5f4a9198cd467
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Q_STATIC_ASSERT gives better error with C++11 enabled.
Aslo the qt_check_for_QOBJECT_macro had warning on some compiler since
it used null reference
Change-Id: Ic6115da800064b00c50a5762f0b79f5f656bf750
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
A number of QDateTime functions directly use or modify the data stored
in the private, but future changes to store msecs and status make this
maintenance more complicated. Where possible simplify this code to use
the standard msecs functions, standard constructors, or public api
instead. This greatly simplifies the functions and the following msecs
storage code changes.
This is an intermim step towards storing the time in msecs. Some
functions will be slower as a result of this change, optimization
will take place after all the msecs changes are completed.
Note this also removes a test that used valid QDates outside the range
of msecs, this change in behavior will be documented in the final
mscs change.
Change-Id: I6ef710f24babc7024091010064082e9be0b5bbfe
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Test the Daylight Time transitions. QDateTime does not correctly deal
with many of these scenarios so those tests are marked as QEXPECTFAIL.
These bugs will be progressively addressed in coming commits.
Change-Id: I01eba9d6143a792f081542cb198e221efcf28e98
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
The "base64url" encoding is defined in RFC 4648, which is the newest
version of Base64. There are also a few situations where the ending
'=' is not desired.
Change-Id: I9bb9fa55817b57d125189e4e795d6fde97caea6d
Reviewed-by: Richard J. Moore <rich@kde.org>
Add new public api to get and set the number of msecs since the start
of the day. Modify QDateTime to use the new msecs api.
[ChangeLog][QtCore][QTime] Added new methods fromMSecsSinceStartOfDay()
to create a new QTime from an msecs value, and msecsSinceStartOfDay()
to return the QTime as the number of msecs since the start of the day.
Change-Id: I285b725b883f1f5524fda87ca81bd64ed99fe6f4
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
This avoids dynamic construction of the private class. According to
the benchmark we go from 4,550 to 3,900 instruction reads per iteration.
(without change 32629676 the baseline is 5,600)
Change-Id: I5df925e30dbd49bdde87173e481820574ce5abe1
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Make sure we don't modify the lhs. Instead copy it and advance the
copy.
Change-Id: I3440e8e175bfc299f8f83b816faca54fa3f79e43
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Any known smart pointer in a QVariant can be handled in this way. The
metatype system can be informed of new smart pointer types using an
existing macro which is now documented.
This is very similar to the existing infrastructure
for containers.
Change-Id: Iac4f9fabbc5a0626c04e1185d51d720b8b54603d
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
New functions left, right, mid were missing in the api.
Change-Id: I3590a84431555d009d5012b204c111385bdceed3
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
which would interpret 'path' as a hostname.
The check is in the public setPath so that the internal one can still
support parsing URLs such as ftp://ftp.example.com//path.
[ChangeLog][Important Behavior Changes][QUrl and QUrlQuery]QUrl now
normalizes the path given in setPath, removing ./ and ../ and duplicate
slashes.
Change-Id: I05ccd8a1d813de45e460384239c059418a8e6a08
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
-DKEY="VALUE" was correctly turned into -DKEY=VALUE,
but -DKEY=\"VALUE\" was turned into -DKEY=\VALUE"
due to the erroneous check ('quote' is still 0 when
encountering the first '\' character).
This fixes QCoreApplication::arguments() as used by
moc.exe after porting to QCommandLineParser.
Further investigation shows that double-quotes and single-quotes
are handled differently. The tests now ensure that this parser
respects what Windows sends in argv, and in particular that
QTBUG-15379 doesn't regress, as well as fixing QTBUG-30628.
Task-number: QTBUG-15379, QTBUG-30628
Change-Id: I95062c9a6022632b321b2f6fae3089f07be7b5c6
Reviewed-by: Richard J. Moore <rich@kde.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Replace that need with a macro so that registration of the
container helper conversions is done at the time of registration
of the container (usually when it is put into a QVariant).
Change-Id: I823fb3fdbce306ebc9f146675ac43724cec678d5
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
To be consistent with signals which are public since Qt5.
Change-Id: I633077e37d0851b118c22da0681e8b8b1892ddbb
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
QMap explicitly sorts its entries by key value. For an ordered container
it's (often?) useful to access the first or last entry, for instance to
quickly compute the next key of the mapping. The first entry is easily
accessible by the STL begin() method, but for accessing the last entry
pretty ugly iterator arithmetics must be applied: *(end() - 1). With
their first() and last() accessors the container classes QList and
QVector provide a much nicer method of accessing extrema, so for
consistency this syntactical sugar also should be applied to QMap.
Change-Id: Ibd544acbad8c3ac16f12a1e74362207ea1694375
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Thorbjørn Lund Martsum <tmartsum@gmail.com>
Change the debug format from Qt::TextDate to a more detailed ISO style
format including better time spec output.
[ChangeLog][QtCore][QDateTime] The debug datastream is now an ISO-like
format instead of Qt::TextDate
Change-Id: Iddbb8199c3bfbf7bca845482617e7a85da43259d
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Before this patch,
qRegisterMetaType<QSharedPointer<double> >("QSharedPointer<double>")
without a metatype declaration fails to compile, whereas it works
with Qt 5.1 (ie, before commit e9a69c3ba9)
Change-Id: I9408f711c9df810ff29b879b7696dab81c1160f1
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Clean-up the implementation of toString() and fromString() methods in
QDate, QTime and QDateTime code to be more consistent in ISODate and
TextDate behavior, especially when handling TimeSpec.
Reformat some code so all methods are consistent in appearance and
function to make maintenance easier.
This changes some corner-case behavior in TextDate and ISODate, but
this either fixes bugs or makes the behavior match the documentation.
Change-Id: I457aa1d7cd4f448cd9f8a2e80ec635f3cb98e58c
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
Switch the implementation of toString() methods in QDate, QTime and
QDateTime to use the QLocale formatter, and remove the now redundant
QDateTime formatter.
Change-Id: Ie4f17c8a6e31acde3ce066f19835bb2b83351ce8
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
Modify the QLocale date formatter to be consistent with the QDateTime
date formatter and able to replace the QDateTime formatter in a
subsequent change.
Fix the treatment of negative years.
The internal QLocale::timeZone() has been replaced by the
QDateTime::timeZoneAbbreviation() to ensure the correct tz for the
date/time is used rather than always the current system default.
Change-Id: I2ef26700856e2e69b979069226aa504ecbb50071
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
this enables forwarding standard input from the parent process.
Change-Id: I7ee72b9842acc96320d4da693b95dd15d9a7b4d4
Reviewed-by: Simon Hausmann <simon.hausmann@digia.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Oliver Wolff <oliver.wolff@digia.com>
The destructor of the ConverterFunctor calls unregisterConverter. If
the instance is static (as it is in qmetatype.h), then this method
can be called after the QGlobalStatic has already been destroyed.
Change-Id: I33b70734cbe09dfa888e887280c349e43ad1b855
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
forwardedChannelsOutput() actually tests forwarding. also having the
older test only eats CPU cycles.
rename the new test to the old name for clarity.
Change-Id: I16a326c8692bb594d7df7897bc53d31b23c66d90
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
close stdin and let the process finish right away instead of first
waiting for a reaction (which is not supposed to come) and then
finishing.
Change-Id: Ifcf200eead5ed95217843e105f9d2dbb5398d646
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
§24.4.1 [iterator.traits] of N3337 makes it *mandatory* to define the
various types
std::iterator_traits<Iterator>::{value,pointer,...}
for any given iterator.
§24.4.1.2 makes it enough to define them within our iterator class.
Change-Id: I1305c8392d224889b09395e30ecc77b194020a15
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This reverts commit 5b9006bbdb.
Also revert "Doc: Enable documentation for QScopedPointer's rvalue ref functions"
This reverts commit 5f8416ec65.
Adding a move contructor to QScopedPointer makes no sense, because moving
means 'escaping the scope', which breaks the fundamental point of
QScopedPointer.
Change-Id: I4ac1b108bf199af6e436fa1629aa2d3b93c27724
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
To ease interruption of long running tasks, a new method
QThread::setInterruptionRequested() can be called.
The task can check QThread::isInterruptionRequested()
and act upon it by stopping itself.
These methods are designed to replace the use of a global variable
and other hacky ways to stop a task running in another thread.
Change-Id: I17622dd60d2262078210e7e4294ad6c53a6dc179
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The WaitForMoreEvents flag only guarantees that we will process _some_
events -- either if they are in the event queue already, or by sleeping
and then waking up to process an event. This event might be a system
event, not the timer firing, so a single call to processEvents() is
not enough to guarantee that the timer has fired. Instead we do a
Q_COMPARE with a timeout, where we continiously process events until
we see that the timer fired.
Change-Id: I5dc04377f04190f3505be22e877af73d11b7547d
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Add a new method to return the time zone abbreviation for the current
time spec. For LocalTime this is the abbreviation returned by mktime.
This new method will later be used in changes to the date formatter
and QTimeZone.
Note this change does not implement WinCE support.
[ChangeLog][QtCore][QDateTime] Add method timeZoneAbbreviation() to
return effective time zone abbreviation.
Change-Id: I265a5e96c72eb7236974f80f053f1fb341e3c816
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
A handful of tests lack QT_NO_PROCESS guards, making them non-compilable
on Qt builds with no QProcess support. This commit does not change
QProcess-specific tests, which should be left out of the build using the
.pro file mechanism.
Change-Id: Iac8562428abc1f59ccbb23bf5c3a919785e41f12
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
QTEST_NOOP_MAIN was removed in Qt 5, but was still in the source of the
QProcessEnvironment auto test.
Change-Id: I0048c330c9d4e99c46192cc05ffd3b77d404fb6d
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Jason McDonald <macadder1@gmail.com>
Since Qt5, the QMetaObject do not contains the string name of the
builtin types, but only the QMetaType id. QMetaMethod::typeName
convert back from the id to the string. But if the type is aliased,
the string of the main type is returned.
This was the case for example for qint64 which is transformed to
"qlonglong".
This causes a regression in QMetaType::invoke when trying to invoke a
method which return an aliased type, since the string comparison would
fail.
Fix the problem by also comparing the metatype id.
Changelog: QMetaMethod::invoke: Fix return of aliased meta type
Task-number: QTBUG-33222
Change-Id: Iec7b99dcbf7b23eb818de74f413e4451ce510ac4
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@digia.com>
(dacf9961da)
In order to see 2 active threads, we must check activeThreadCount() before
the runnable finishes.
Change-Id: I1a48b41e0c1fd81a65d915b9bd1e741ff267ed2b
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
This extends the QLinkedList auto test. This is copy paste
with small edits from the QList auto test.
Change-Id: I5bce7ee304998d1c7258f11b2f31a7fb87706936
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Add convenience methods for fromMSecsSinceEpoch() and
fromTime_t() to enable direct creation of other time specs
than LocalTime without the overhead of unncessary conversions.
For example instead of:
QDateTime dt = fromMSecsSinceEpoch(12345).toUtc();
the following saves two conversions:
QDateTime dt = fromMSecsSinceEpoch(12345, Qt:UTC);
This will improve the performance of the new QTimeZone class.
[ChangeLog][QtCore][QDateTime] Added convenience methods for
fromMSecsSinceEpoch() and fromTime_t() to take time spec to be used in
returned datetime.
Change-Id: I133635bfe3d35ee496a287257e13b2d600225a38
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
The Qt::OffsetFromUtc TimeSpec was made public in Qt 4.4 but the access
methods were never made public in the apidox effectively meaning the
feature was never used. The implementation was also incomplete and
inconsistent.
This change cleans up the implementation and exports new public API for
using the TimeSpec using new method names consistent with the new
QTimeZone support.
This change increases the QDataStream Version number for Qt 5.2 to 15.
The behavior of one constructor has changed slightly to be consistent
with the rest of the feature, but this behavior was never documented.
[ChangeLog][QtCore][QDateTime] Fully implement support for Qt::TimeSpec
of Qt::OffsetFromUTC, added new methods for offsetFromUTC(),
toTimeSpec(), and toOffsetFromUTC().
Task-number: QTBUG-26161
Task-number: QTBUG-29666
Change-Id: If3cc7fc9778ca2b831644408ae749448d5975a3a
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
It's a nice feature to have.
MSVC also complains about using doubles to create enum values, so
the ugly workaround is:
enumValue = MyEnum(qRound(json["myEnumValue"].toDouble()));
[ChangeLog][QtCore][QJsonValue]Added QJsonValue::toInt().
Change-Id: I1a200b912abf66b2e96390b1980caff26cfa2685
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
The QCommandLineParser class provides a means for handling the command line options.
QCoreApplication provides the command-line arguments as a simple list of strings.
QCommandLineParser provides the ability to define a set of options, parse the
command-line arguments, and store which options have actually been used, as
well as option values.
Done-with: Laszlo Papp <lpapp@kde.org>
Change-Id: Ic7bebc10b3f8d8dd06ad0f4bb897c51d566e3b7c
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Before a call to erase on a shared instance would imply that the
item was inserted into the shared data (i.e all instances)
Change-Id: I655ccf04b1ad9bf82e6bfade58929538fa7df000
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This patch ensures correct detach when insert with an iterator is
called on a shared instance (i.e same behavior as QVector)
Change-Id: Id660eacd3cc7b633456dfa989997bbad747e1df2
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Before a call to erase on a shared instance would imply that the
item was removed from the shared data (i.e all instances)
This patch improves the behavior to detach and erase the item
specified by the iterator (i.e same behavior as QVector)
Change-Id: Ib3cfb5363c86b400886c80b75b0c20ca854ce801
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Before calls to erase on a shared instance would in release mode
imply that items were removed from the shared data (i.e all instances).
In debug mode it would assert.
This patch improves the behavior to detach and erase items
specified by the iterator(s) (i.e same behavior as QVector)
Change-Id: I89b69446cb1ffd43a98402b7ab1ec9a59bceb8e6
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Before a call to erase on a shared instance would imply that the
item was removed from the shared data (i.e all instances)
This patch improves the behavior to detach and erase the item
specified by the iterator (i.e same behavior as QVector)
Change-Id: Ia44db84fc1388d92308bf0d2b32539ac4d53850b
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Before a call to erase on a shared instance would imply that the
item was removed from the shared data (i.e all instances)
This patch improves the behavior to detach and erase the item
specified by the iterator (i.e same behavior as QVector)
Since QSet uses QHash it improves QSet the same way.
Change-Id: I850b1efcf7bdfc85ceddb23128b048af95f75063
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QTBUG-21051 has a testcase where activeThreadCount() could actually
end up at -1 (converted to an autotest in this commit).
The reason was: start() calls tryStart() which returns false due to
too many active threads (reserveThread() causes this), so it calls
enqueueTask() - which actually wakes up the waiting thread, but
it didn't decrement the number of waiting threads.
Note that tryStart() is "if I can grab a waiting thread, enqueue task and wake it"
while start(), in case tryStart() fails, wants to "enqueue, and then if I can grab
a waiting thread, wake it". This is why enqueue shouldn't wake; waking must happen
only if we can grab a thread (d->waitingThreads > 0).
Task-number: QTBUG-21051
Change-Id: I3d98337103031c9bdf0bf365295f245be0c66aa7
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QUrl("http%3A%2F%2Fexample.com") has only a path of
"http%3A%2F%2Fexample.com". In Qt 5.0 and 5.1, the %3A would get decoded
to ':', which in turn makes the URL invalid (colon before first slash).
Found via discussion on the interest mailing list.
Change-Id: I7f4f242b330df280e635eb97cce123e742aa1b10
Reviewed-by: David Faure <david.faure@kdab.com>
This fixes the wrong value for path() and fileName() when a
path or file name actually contains a '%'.
userInfo() and authority() are not individual getters, they combine
two or more fields, so full decoding isn't possible (e.g. username
containing a ':').
[ChangeLog][Important Behavior Changes][QUrl and QUrlQuery]QUrl now
defaults to decoded mode in the getters and setters for userName,
password, host, topLevelDomain, path and fileName. This means a '%'
in one of those fields is now returned (or set) as '%' rather than "%25".
In the unlikely case where the former behavior was expected, pass PrettyDecoded
to the getter and TolerantMode to the setter.
Change-Id: Iaeecbde9c269882e79f08b29ff8c661157c41743
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The unit tests were caching the original TZ value to restore later
after testing with different TZ values. The problem is reading TZ will
return a null value if no override TZ value is set, and if you then set
the TZ to null the system assumes UTC and not the system time zone.
Instead we need to unset TZ if it was null to start with.
Change-Id: Ib0625b1712e565f9fdfa99e2ffe1e5d74f059354
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QThreadPool::clear() method removes all queued QRunnable.
When a large number of long-running tasks are queud in a
QThreadPool its destruction, which calls waitForDone(), can
be quite long.
QThreadPool:clear() removes (and deletes when appropriate)
all QRunnable that have yet to be started from the queue
enabling a faster interruption.
Change-Id: Ie5d6028ad3cfe7e439d1db068c8d0936ff818db9
Reviewed-by: David Faure <david.faure@kdab.com>
Invoking waitForStarted() on a QProcess before or after an unsuccessful
call to start() (e.g., with an empty command), would execute FD_SET with
an invalid file descriptor and cause the process to abort.
The bug can be reliably reproduced on OSX.
Task-number: QTBUG-32958
Change-Id: Id25b7781168489281645e21571361ca1a71d43e3
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Follows a similar include-pattern as the qguieventdispatcher test.
Change-Id: Ie8669a5bc155abd6687e81526f2b95d0d19b009e
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
A QCoreApplication may run a different event dispatcher, such as the
QEventDispatcherBlackberry, and QGuiApplications will have the GUI
dispatcher provided by the QPA plugin, such as QCocoaEventDispatcher.
Neither support X11ExcludeTimers.
Change-Id: Id5ea1c7dd74a127e13fa4d2eaa9a1bd2715a9dbb
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
This is a flaky test on Windows 8 64-bit, so marking it as
XFAIL if it is expected to fail.
Task-number: QTBUG-30943
Change-Id: Idd276f80b54fcd5cf295a7e1adebcf0020eaa8ca
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
QStandardPaths::enableTestMode has a verb in the imperative ("enable")
as the core word in the name. That indicates an action. The function
should not have had a parameter.
Instead, add a Qt-style setXXXEnabled function.
[ChangeLog][QtCore][QStandardPaths] QStandardPaths::enableTestMode is
deprecated and is replaced by QStandardPaths::setTestModeEnabled.
Change-Id: Ib26ad72d7c635890d2cb22ae9d44cbda08a6f17c
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: David Faure (KDE) <faure@kde.org>
Mark some tests as expected failures on OS X 10.8
- tst_QSettings::ctor(native)
- tst_QSettings::ctor(ini)
- tst_QSettings::ctor(custom1)
- tst_QSettings::ctor(custom2)
- tst_QSettings::rainersSyncBugOnMac(native)
Task-number: QTBUG-32655
Change-Id: I54928d991a8ccf300b40747feaa6fda9d124781b
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
As Qt still thinks that all UNIX filesystems are case sensitive, which
is not the case for eg Mac, where they might be both, as well as for
mounts of other filesystems.
Change-Id: I07b8550685bfa17ac407c20ac991dc54df040942
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
This prevents deadlocks in case the destructor re-enters.
(Example: a functor containing a QSharedPointer of a QObject)
This also fixes a leaked slot object in disconnectHelper.
Change-Id: Ia939790e3b54e64067b99540974306b4808a77f2
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Change-Id: Icd6a09402c3cf14286f4ba1f8f4c99ac483ec1a3
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
Added the compiler options, we need the info of sdk.
Task-number: QTBUG-32715
Change-Id: I70612f36a16e0ab5025194a10ce399822e159c7c
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
By not assuming that the file system is case insensitive. OSX supports
both.
Change-Id: I11a4ac4cdff97b97b183dd319757a42ae14bb52d
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Make qabstractproxymodel and qidentityproxymodel build and run
even if -no-widgets is used since they don't depend on Qt Widgets.
Change-Id: I48bc2f6a78812b1bf0083f76c6a4e106f4e38650
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
For some time, we've assumed that the URL specification had a mistake in
that it didn't allow the "#" character to appear decoded in the
fragment. We've gotten away with it so far.
However, turns out that the CoreFoundation NSURL class doesn't like it.
So we have to be stricter.
[ChangeLog][Important Behavior Changes][QUrl and QUrlQuery] QUrl no
longer decodes %23 found in the fragment to "#" in the output of
toString(QUrl::FullyEncoded) or toEncoded()
Task-number: QTBUG-31945
Change-Id: If5e0fb37bae84710986c9ca89bd69ec98437cd63
Reviewed-by: David Faure (KDE) <faure@kde.org>
It's a good practice to always replace bad UTF-8 sequences with the
replacement character. It could be considered a security issue too.
Change-Id: I9e7d72e4c4102cdb8334449b5e7f882228a9048f
Reviewed-by: David Faure (KDE) <faure@kde.org>
Those sections contain more than one components of a URL, separated by
delimiters. For that reason, QUrl::FullyDecoded and QUrl::DecodedMode do
not make sense, since they would cause the returned value to be
ambiguous and/or fail to parse again.
In fact, there was a comment in the test saying "look how it becomes
ambiguous".
Those modes are already forbidden in the setters and getters of the full
URL (setUrl(), url(), toString() and toEncoded()).
[ChangeLog][Important Behavior Changes][QUrl and QUrlQuery] QUrl no
longer supports QUrl::FullyDecoded mode in authority() and userInfo(),
nor QUrl::DecodedMode in setAuthority() and setUserInfo().
Change-Id: I538f7981a9f5a09f07d3879d31ccf6f0c8bfd940
Reviewed-by: David Faure (KDE) <faure@kde.org>
The longer explanation can be found in the comment in qurl.cpp. The
short version is as follows:
Up to now, we considered that every character could be replaced with
its percent-encoding equivalent and vice-versa, so long as the parsing
of the URL did not change. For example, x:/path+path and
x:/path%2Bpath were the same. However, to do this and yet be compliant
with most URL uses in the real world, we had to add exceptions:
- "/" and "%2F" were not the same in the path, despite the delimiter
being behind (rationale was the complex definition of path)
- "+" and "%2B" were not the same in the query, so we ended up not
transforming any sub-delim in the query at all
Now, we change our understanding based on the following line from
RFC 3986 section 2.2:
URIs that differ in the replacement of a reserved character with
its corresponding percent-encoded octet are not equivalent.
From now on, QUrl will not replace any sub-delim or gen-delim
("reserved character"), except where such a character could not exist
in the first place. This simplifies the code and removes all
exceptions.
As a side-effect, this has also changed the behaviour of the "{" and
"}" characters, which we previously allowed to remain decoded.
[ChangeLog][Important Behavior Changes][QUrl and QUrlQuery] QUrl no
longer considers all delimiter characters equivalent to their
percent-encoded forms. Now, both classes always keep all delimiters
exactly as they were in the original URL text.
[ChangeLog][Important Behavior Changes][QUrl and QUrlQuery] QUrl no
longer decodes %7B and %7D to "{" and "}" in the output of toString()
Task-number: QTBUG-31660
Change-Id: Iba0b5b31b269635ac2d0adb2bb0dfb74c139e08c
Reviewed-by: David Faure (KDE) <faure@kde.org>
7721c3d27c fixed the case where two similar definitions are in the same
directory. This commit fixes the case where two similar definitions are
in different directories, both in the search path (GenericDataLocation).
If the file extension gives us the same mimetype twice, there's no conflict,
i.e. no reason to fallback to determination from contents.
Change-Id: I72c56004b6d5e88964159e53ec160ce8b06c2264
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
On some platforms (like BlackBerry) qDebug doesn't write to stderr,
so we directly write to stderr with fprintf.
Change-Id: Ib86211c98cf4da1fa2dbea4600a78e2013dc1a5a
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Halley's method to get a better approximation is omitted, if it
would include a devision by zero (INFINITY/NaN is worse).
Change-Id: Ida09326e2b5892d7cb21bcb956631c289e5b56ba
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
It makes sense for a QScopedPointer to be movable, for instance
for allowing instances to be returned from a function.
Ownwership of the managed pointer is still tied to one (and one only)
QScopedPointer instance.
Moreover, a move assignment operator makes sense as well, as it
implementing the equivalent of
this->reset(other.take());
only when other is a rvalue and not a lvalue (so either it's a temporary
or it's getting explicitly moved in with std::move).
This makes QScopedPointer API's a bit closer to std::unique_ptr's one.
Task-number: QTBUG-29754
Change-Id: If1ac0c688327a67af4ad5b7ad45b439b022ed1c6
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QDateTime can now be converted to strings compliant to RFC 2822.
Additionally, it supports RFC 850 and RFC 1036 during parsing.
By having them all together, all type of dates found in exchanged
messages on the internet (including USENET) get supported.
Change-Id: I771066c23f409d20b31b7d802f37852ea68ca2a0
Reviewed-by: David Faure <david.faure@kdab.com>
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
Codecs registered by creating new QTextCodec instances should be listed
there.
Task-number: QTBUG-32500
Change-Id: I56c00e0d6bbfef55a6cbd571bcf9aa2cf333ef3a
Reviewed-by: David Faure <david.faure@kdab.com>
Currently tst_QEasingCurve::setCustomType() is always failing on CI
when ran on 32 bit 11.10 Ubuntu.
Task-number: QTBUG-32432
Change-Id: Iaf346c14985f14716692fe996714b7040fb70930
Reviewed-by: Sergio Ahumada <sergio.ahumada@digia.com>
We don't know what it might be used for. The RFC for URI says it's an
HEXDIG, and since we uppercase all other HEXDIGs already (in
percent-encodings...).
Change-Id: I56d0a81315576dd98eaa2657c0307d79332543a5
Reviewed-by: David Faure (KDE) <faure@kde.org>
So far, this function hasn't been used for input coming in from the
user, so it wasn't necessary. But we may want to do it, or we may
already be doing it accidentally somewhere that isn't triggering the
failed assertions during unit testing.
So let's be on the safe side and allow it. And test it too.
Change-Id: Ib63addd8da468ad6908278d07a4829f1bdc26a07
Reviewed-by: David Faure (KDE) <faure@kde.org>
We have no idea what it might contain, but test it anyway to make sure
it works. Turns out there were a few bugs the unit tests have now
caught.
Change-Id: I0a6c868365feec31c2360b3c341c8ca6944f4352
Reviewed-by: David Faure (KDE) <faure@kde.org>
Registered names and IP addresses can only contain unreserved
characters (letters, digits, dots, hyphens, underscores) and the
colon, which is a gen-delim. For registered names and IPv4 addresses,
we can simply use the default config -- if anything that remains
percent-encoded, it means it's not a valid hostname anyway.
For IPv6, we just need to decode the colon.
Change-Id: If8083d47f6e5375f760e7a6c59631c89e4da8378
Reviewed-by: David Faure (KDE) <faure@kde.org>
It's not necessary to check at every point if we know the minimum file
size: it must contain at least the header, one segment (__TEXT) and one
section (qtmetadata). Most files have more than one segment and more
than one loader command, so this check does not mean we can eliminate
the checks further down.
Also be more resilient against corruptions in the header data: check not
only the additions, but the values themselves. For example, an offset +
size addition could be smaller than the file size when the addition
overflows in 32-bit. Another thing is that the cmdsize fields could be
corrupt too.
Change-Id: I7968a769c1cbe9150270c91823cafc4f8f833876
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
We already had an ELF decoder, which helped us greatly to find the
metadata and that catches most Unix systems (Solaris, QNX, HP-UXi, and
all of the free Unixes). On other Unix systems, aside from Mac OS X,
we simply scanned the entire file for the signature. On Windows, even
without a COFF-PE decoder, we use a LoadLibrary trick to load the
plugin without loading the dependent libraries. In most cases, that
works.
Unfortunately, on Mac OS X we didn't have a decoder and nor could we
do the file scan: because Mac OS X binaries could be fat binaries, we
wouldn't know which architecture's signature we had found.
No more. This adds a full Mach-O decoder to QtCore. It is also capable
of finding the boundaries of the architecture's binary, but that
functionality is disabled since all Qt 5 plugins have plugin metadata
sections.
Change-Id: I2d5c04c5ecf024864b8a43f31ab6b7e6c5eae9ce
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The rule for a new override is that it must still work if the old
implementation is called. The catch is that any class that derives from
QProcess and isn't recompiled will still have QIODevice::open in its
virtual table. That is equivalent to overriding open() and calling
QIODevice::open() (like the tests).
In Qt 5.0, QProcess::start() called QIODevice::open directly, not the
virtual open(), so there's no expectation that a user-overridden open()
be called. With that in mind, simply fix QProcess::start to not call the
virtual open at all.
Similarly with QLocalSocket, the calls to open were always non-virtual.
Task-number: QTBUG-32284
Change-Id: I88925f0ba08bc23c849658b54582744997e69a4c
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
This is a bit like QDir::cleanPath(), but for URL paths.
The code is shared with QDir::cleanPath(), by extracting the common parts
it into a helper, qt_normalizePathSegments().
Change-Id: I7133c5e4aa2bf17fba98af13eb5371afba64197a
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The deadlock is caused because the QEvent is destroyed while holding the
event list mutex. And the QEvent may have a custom destructor that will
re-enter the event handlng code.
The QScopedPointer that should destroy the event must be created after
the MutexUnlocker.
Regression introduced by commit f9035587b9
Task-number: QTBUG-31606
Change-Id: I6b2cbc2656eacdec61b641886953f00bf5b3ff36
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Makes for more informative debug output when the tests fail.
Change-Id: Ib07dd79452a56413c711394dd72aa37dbb4a70d7
Reviewed-by: Simon Hausmann <simon.hausmann@digia.com>
This allows to find the parent directory url using
url.adjusted(QUrl::RemoveFilename).
Change-Id: I1ca433ac67e4f93080de54a9b7ab2e538509ed04
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
qt_ACE_do(".co.uk") was returning an empty string because of the
leading dot. Allow leading dots from topLevelDomain, but not from
other calls.
Change-Id: I757d9960708e205d30554cd2bbcf618c8624792b
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Calling waitForNotified on an uninitialized notifier will print a
warning and return false. The autotest has been adjusted.
Change-Id: I85e18d6d0a8a5462e1a5d451613add941d89b5fb
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Need to store 17 decimal digits for binary64, IEEE 754 double formats.
Autotest is included. Test cases from TC39 test suite for ECMAScript.
Task-number: QTBUG-31926
Change-Id: I546398f21ea7ff5e40e89fc9de8703f628f55df9
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@digia.com>
Do not lower case file names to generate hash keys since
QString::toLower() converts some characters with context
which the Windows file system will not.
Task-number: QTBUG-31341
Change-Id: I285bfedef3c1ca9d59083229e61974dd378c72ae
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@digia.com>
Instead of calling different wait functions on different platforms,
we use QTRY_COMPARE to check the process state.
Change-Id: I6489cabce9e63f9c8b1036f3cccbf35b52df72e7
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Andreas Holzammer <andreas.holzammer@kdab.com>
This reverts commit e3fa266623b08e837cb4ccc7fe59da243d03dd27
That commit applied a change at the wrong place in the code.
Change-Id: I21e3045a3af14ad2f90c5fe338815c35a2d27ae6
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: David Faure (KDE) <faure@kde.org>
Leading and double dots are bad, but trailing dots are fine. The ASCII
part of a hostname is supposed to be LDH (letters, digits, hyphen) only,
but we accept '_' (underscore) as an exception too.
Change-Id: I79957ddec4da78a0e2357fe50c8687db03e1c99e
Reviewed-by: David Faure (KDE) <faure@kde.org>
qt_ACE_do(".co.uk") was returning an empty string because of the
leading dot. This has always caused issues in KDE code too, where ACE
normalization needs the dot removed, and re-added afterwards.
Change-Id: Id9fcea0333cf55c14d755a86d4bf33a50f194429
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
The issue is already fixed in 5.0 but let's be nice and ensure the issue
won't be reintroduced later.
Task-number: QTBUG-30931
Change-Id: Ia6944acaf6e7217f8d0f1fa75d0e9977db11d892
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
QDateTime.toString() is writing out milliseconds since change
15da0a5af2. Unfortunately this breaks
QDateTime::fromString() with Qt::TextDate which can't handle the new
format.
Fix by making QDateTime::fromString split up seconds and milliseconds
on a period, if any. Now
QDateTime dt = ...;
assert(QDateTime::fromString(dt.toString(), Qt::TextDate) == dt)
works again.
Change-Id: Ibfe9032e357ceaf894e33f3e33affe94f56dbf5c
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
During a call to QDir::mkpath(), the same path could be created
by another process, in which case the OS mkdir will fail with EEXIST.
But the docs for mkpath() state that it's not an error if it
already exists, whereas for mkdir() it is an error. So
QFileSystemEngine::createDirectory should accept the EEXIST error
silently if it occurs while creating the sequence of parent directories
and the final leaf directory, but should fail if EEXIST happens when
it was called from QDir::mkdir(), which is when the createParents
parameter is false. We assume the operating system mkdir() and
CreateDirectory() are atomic, so there should be no race condition
in QDir::mkdir(). It's not necessary for mkpath() to call stat()
at each level, only to check whether an existing entry is a directory
or a file. Also added to the autotest to verify that if the
path is an existing file, creating a dir with the same name will
fail in either mkdir or mkpath.
Task-number: QTBUG-30046
Change-Id: I926352f10654fdf3b322c8685bb85ad8b8844874
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
There is a mismatch how QML and C++ converts QJsonValue. This patch
unifies conversions by adding QJsonValue support in QVariant::convert().
Change-Id: I8a1db3d77c517945ef48064b4b66ba03aa4f2fd0
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
BlackBerry OS does not allow for explicitly controlling the locale through the
"LANG" environment variable. Locale is controlled by the underlying PPS
Service instead.
Change-Id: I22154e39f81a9467ad7fdb90a042396390398b1b
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
Reviewed-by: Kevin Krammer <kevin.krammer@kdab.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Mehdi Fekari <mfekari@blackberry.com>
In Qt 5, we declared that throwing from event handlers is undefined
behavior. So stop testing this.
We will try our best to capture and pass along std::bad_alloc, but even
that might not work, depending on compiler settings. In particular,
after the upgrade to MinGW/GCC 4.8 with DW2, this test stopped working.
Task-number: QTBUG-31615
Change-Id: Ibf5fb2ce0c48b983549096bf7aac434b6ed3ac2e
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
qt_nameprep is tested by tst_qurlinternal. We just need to be sure that
QUrl handles them correctly.
Change-Id: Ic563004870d2cf2fa7a31ce49fff7280d5ffb5f3
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
