While working with HTTP/2, we are not re-sending failed requests.
In case we receive a GOAWAY frame, we properly handle it by
processing some active streams if possible, and aborting streams
that will not proceed further with ContentResendError. But it's
possible that some server failed to send us GOAWAY (for example,
it died) or closed the connection not finishing the streams that
were still active and valid (ID <= value from GOAWAY frame).
Now that we will not re-connect, there is no reason to be quiet
about us not progressing - emit RemoteHostClosedError on any
remaining active stream/request we cannot process further.
Fixes: QTBUG-77852
Change-Id: I4cd68a1c8c103b1fbe36c20a1cc406ab2e20dd12
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
(cherry picked from commit 543769666f)
While working with HTTP/2, we are not re-sending failed requests.
In case we receive a GOAWAY frame, we properly handle it by
processing some active streams if possible, and aborting streams
that will not proceed further with ContentResendError. But it's
possible that some server failed to send us GOAWAY (for example,
it died) or closed the connection not finishing the streams that
were still active and valid (ID <= value from GOAWAY frame).
Now that we will not re-connect, there is no reason to be quiet
about us not progressing - emit RemoteHostClosedError on any
remaining active stream/request we cannot process further.
Fixes: QTBUG-77852
Change-Id: I4cd68a1c8c103b1fbe36c20a1cc406ab2e20dd12
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
I realized this is a potential scenario where we will have leftover
data, but it wasn't covered.
Change-Id: Ibaf1015bf2aee120e4a4d98888925b88ecb6ddfd
Reviewed-by: Jesus Fernandez <jsfdez@gmail.com>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
The minimum supported version is Windows 7. Remove
QOperatingSystemVersion::WindowsVista added by
b0cd007335 and replace with "true"
wherever it was used.
Change-Id: I08c0208467b655a921b6773f77d8bc099be69031
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Mark QCalendarBackend as internal since it is in a private header
and fix some issues in the QCalendar related classes.
src/corelib/time/qcalendar.cpp:201: (qdoc) warning: clang found diagnostics parsing \fn int QCalendarBackend::daysInMonth(int month, int year) const
error: incomplete type 'QCalendarBackend' named in nested name specifier
(repeats)
src/corelib/time/qdatetime.cpp:1426: (qdoc) warning: Unknown command '\override' (repeats)
src/corelib/time/qcalendar.cpp:642: (qdoc) warning: Undocumented enum item 'Last' in QCalendar::System
src/corelib/time/qcalendar.cpp:642: (qdoc) warning: Undocumented enum item 'User' in QCalendar::System
src/corelib/time/qcalendar.cpp:744: (qdoc) warning: Undocumented parameter 'year' in QCalendar::isLeapYear()
src/corelib/time/qcalendar.cpp:923: (qdoc) warning: Can't link to 'dateTimeString()' (repeats)
src/corelib/time/qcalendar.cpp:893: (qdoc) warning: No such parameter 'year' in QCalendar::partsFromDate()
src/corelib/time/qcalendar.cpp:893: (qdoc) warning: No such parameter 'month' in QCalendar::partsFromDate()
src/corelib/time/qcalendar.cpp:893: (qdoc) warning: No such parameter 'day' in QCalendar::partsFromDate()
src/corelib/time/qdatetime.cpp:1425: (qdoc) warning: Undocumented parameter 'nmonths' in QDate::addMonths()
src/corelib/time/qdatetime.cpp:1467: (qdoc) warning: Undocumented parameter 'nyears' in QDate::addYears()
src/corelib/statemachine/qstatemachine.cpp:2522: (qdoc) warning: Undocumented enum item 'StateMachineChildModeSetToParallelError' in QStateMachine::Error
src/corelib/kernel/qtimer.cpp:602: (qdoc) warning: Undocumented parameter 'connectionType' in QTimer::callOnTimeout()
src/corelib/time/qcalendar.cpp:159: (qdoc) warning: Undocumented parameter 'name' in QCalendarBackend::QCalendarBackend()
src/corelib/time/qcalendar.cpp:159: (qdoc) warning: Undocumented parameter 'id' in QCalendarBackend::QCalendarBackend()
src/corelib/time/qcalendar.cpp:529: (qdoc) warning: Can't link to 'registerCalendar()'
src/corelib/time/qcalendar.cpp:529: (qdoc) warning: Can't link to 'fromName()'
src/corelib/time/qcalendar.cpp:178: (qdoc) warning: Can't link to 'QCalendar::fromEnum()'
src/corelib/time/qcalendar.cpp:405: (qdoc) warning: Undocumented parameter 'jd' in QCalendarBackend::dayOfWeek()
src/corelib/time/qcalendar.cpp:405: (qdoc) warning: Can't link to 'weekDayName()'
src/corelib/time/qcalendar.cpp:405: (qdoc) warning: Can't link to 'standaloneWeekDayName()'
src/corelib/time/qcalendar.cpp:268: (qdoc) warning: Can't link to 'daysInMonth()'
src/corelib/time/qcalendar.cpp:268: (qdoc) warning: Can't link to 'isLeapYear()'
src/corelib/time/qcalendar.cpp:295: (qdoc) warning: Can't link to 'daysInMonth()'
src/corelib/time/qcalendar.cpp:313: (qdoc) warning: Can't link to 'isLuniSolar()'
src/corelib/time/qcalendar.cpp:313: (qdoc) warning: Can't link to 'isSolar()'
src/corelib/time/qcalendar.cpp:313: (qdoc) warning: Can't link to 'isLunar()'
src/corelib/time/qcalendar.cpp:340: (qdoc) warning: Can't link to 'daysInMonth()'
src/corelib/time/qcalendar.cpp:357: (qdoc) warning: Can't link to 'daysInMonth()'
src/corelib/time/qcalendar.cpp:544: (qdoc) warning: Can't link to 'fromName()'
Change-Id: Ia2fabefb917f8e4cfa361044d9b754717276f4aa
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Paul Wicking <paul.wicking@qt.io>
The _q_error slot has a special case for RemoteHostClosedError,
where the current channel's state is 'idle' and no request/reply
is in progress. The comment states that:
"Not actually an error, it is normal for Keep-Alive connections
to close after some time if no request is sent on them. No need
to error the other replies below. Just bail out here. The _q_disconnected
will handle the possibly pipelined replies."
_q_disconnected, indeed, takes care about pipelined replies ... calling
'ensureConnected' even if we have 0 replies in pipeline, which makes
zero sense to me and results in QNAM endlessly trying to re-connect
to the server.
Fixes: QTBUG-77852
Change-Id: I6dcb43b36a6d432bc940246a08f65e1ee903fd24
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
From the API review.
Amends cd816d4b6a.
Change-Id: I3d3e2ef331501fa498545c5eec0e321544165b0d
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
When the end range of a byte range in a HTTP request is skipped the download
manager adds 0 it its place when resuming that download.
When there is no end range given the value is skipped.
Task-number: QTBUG-77867
Change-Id: I52358c94cf56c88217fcc91abb102ed393ac7242
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
* initialize the structure used as argument to ioctl SIOCGIFMEDIA
The ioctl in ifType(...) returned -1 and set errno to EFAULT, thus
the code to determine the interface type was not executed.
Task-number: QTBUG-76408
Change-Id: Ib69c2c0cd4dda3604d568adf9746e0c9a280e9a0
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Karsten Heimrich <karsten.heimrich@qt.io>
Reviewed-by: Christian Ehrlicher <ch.ehrlicher@gmx.de>
Through some misconceptions about ComPtr I ended up thinking that it did
not call AddRef internally on creation. But it does. This lead to always
having > 1 ref-counter.
Also stop the monitor on destruction if it hasn't already been stopped.
As was already done for QNetworkStatusMonitorPrivate.
Change-Id: Ic72a2f5cb3325f86c018f90b497caaec834cb214
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
This should help when things are moving fast, and downloads and
network object are destroyed before the callbacks finishes.
Change-Id: I1f65965bd61efc2e641d03eb071f23e684dd5c44
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
And send it in our 'SETTINGS' frame. Add an auto-test
for this and (as a bonus) - fix a bug accidentally
introduced by the previous change.
Task-number: QTBUG-77412
Change-Id: I4277ff47e8d8d3b6b8666fbcd7dc73c827f349c0
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
QHttp2Configuration is using entities (read definitions) from http2,
which is only conditionally included in the *.pri file (requires http).
So as a result we had linker errors.
Fixes: QTBUG-77759
Change-Id: I8b33b0a4802a295f67edad03da3743b24f7ce514
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@qt.io>
Reviewed-by: Paul Wicking <paul.wicking@qt.io>
It's temporarily storing QByteArrays before we copy them directly to
QIODevice's internal buffer. We can save the extra work by just push
them directly into the buffer.
The signal compression is no longer useful performance-wise, but is
kept as it will throttle the amount of readyRead emissions the users has
to handle.
Reorder some of the operations as a result to make it more natural.
Change-Id: Ifc0481d56fec42545e95970125d883a5c68d647a
Reviewed-by: Alex Trotsenko <alex1973tr@gmail.com>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Upon reading the Apple's documentation it would appear the indexing
is more complicated (they name the specific fields and which kind
of indexing must be applied to each). This requires a finer level
of configuration/control and probably a separate class (aka
QHpackConfiguration? ;). We'll provide it in future, if requested
by our users.
Fixes: QTBUG-77412
Change-Id: I6e9461b3966ed59c8b70873eab999a07a04eb289
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
And either compress or not.
Task-number: QTBUG-77412
Change-Id: I3b09385d2b3caf4f7de0455ad6e22c0f068c33a9
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Similar to TLS configuration that we can use on QNetworkRequest,
we can configure different options in our HTTP/2 handling by
providing QNetworkAccessManager with h2 configuration. Previously,
it was only possible internally in our auto-test - a hack with
QObject's properties and a private class. Now it's time to provide
a public API for this.
[ChangeLog][QtNetwork][QNetworkRequest] Add an ability to configure HTTP/2 protocol
Change-Id: I80266a74f6dcdfabb7fc05ed1dce17897bcda886
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
The source code is still only using the old negative define check.
Change-Id: I7e2775d6d4664cb5b802b0a40fb3f1711c8f6915
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
As is said in RFC7301 in section 3.1 [1]:
Protocols are named by IANA-registered, opaque, non-empty byte strings
[...]. Empty strings MUST NOT be included and byte strings MUST NOT be
truncated.
[1]: https://tools.ietf.org/html/rfc7301#section-3.1
Change-Id: I38168ac570a433807e16121d5dec46d4ac73c4bf
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
As is said in RFC7301 in section 3.1 [1]:
Protocols are named by IANA-registered, opaque, non-empty byte strings
[...]. Empty strings MUST NOT be included and byte strings MUST NOT be
truncated.
[1]: https://tools.ietf.org/html/rfc7301#section-3.1
Change-Id: I2c41fa99984a53cc58803e5a264d06edac964cc6
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Some servers seem to be unable to properly calculate our window size
from a delta we send via WINDOW_UPDATE frame immediately after our
client preface and the SETTINGS frame. The remote replies with a
GOAWAY frame blaming flow control error. Guessing what's this
magic number they use seems to be not feasible, we now use a
half of what we had before.
Fixes: QTBUG-77308
Change-Id: I41dacfd25a395a27003f330d01b6d8d60b8f407c
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Remove "final" since it has problems with 10.0.15063.0:
qnetconmonitor_win.cpp
C:\Program Files (x86)\Windows Kits\10\include\10.0.15063.0\winrt\wrl/client.h(61): error C3246: 'Microsoft::WRL::Details::RemoveIUnknownBase<T>': cannot inherit from 'QNetworkConnectionEvents' as it has been declared as 'final'
with
[
T=QNetworkConnectionEvents
]
kernel\qnetconmonitor_win.cpp(102): note: see declaration of 'QNetworkConnectionEvents'
kernel\qnetconmonitor_win.cpp(373): note: see reference to class template instantiation 'Microsoft::WRL::Details::RemoveIUnknownBase<T>' being compiled
with
[
T=QNetworkConnectionEvents
]
C:\Program Files (x86)\Windows Kits\10\include\10.0.15063.0\winrt\wrl/client.h(61): error C3246: 'Microsoft::WRL::Details::RemoveIUnknownBase<T>': cannot inherit from 'QNetworkListManagerEvents' as it has been declared as 'final'
with
[
T=QNetworkListManagerEvents
]
kernel\qnetconmonitor_win.cpp(468): note: see declaration of 'QNetworkListManagerEvents'
kernel\qnetconmonitor_win.cpp(650): note: see reference to class template instantiation 'Microsoft::WRL::Details::RemoveIUnknownBase<T>' being compiled
with
[
T=QNetworkListManagerEvents
]
Change-Id: Ia35545b65acaebea3fcff194884be8a156974123
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Now that QRecursiveMutex is getting split off of QMutex, QMutexLocker
will stop working on QRecursiveMutex once the split has been finalized
in Qt 6.
Even today, QMutexLocker contains casts from QBasicMutex to QMutex
that some reviewers are uncomfortable with. One way to carry
QMutexLocker forward is to template it on the mutex type, possibly
with aliases like QBasicMutexLocker and QRecursiveMutexLocker. C++17
code would then not require a port, thanks to CTAD.
But we have the problem now, and we can't template QMutexLocker in Qt 5.
The alternative is to look at std and realize that they have surpassed
QMutexLocker in expressiveness already. A scoped_lock cannot be
unlocked again, a unique_lock can be moved around. QMutexLocker
doesn't do either. The only "problem" is that the std lock classes are
already templates, but we can't, yet, rely on C++17 CTAD to make them
look as if they weren't.
So, prepare for a future with C++17 CTAD by writing factory functions,
qt_scoped_lock and qt_unique_lock, which will later port mechanically
to their C++17 equivalents (mostly).
The functions are added to a new private qlocking_p.h becauee we don't
want to make them public. These are for use in Qt's own implementation,
or for users that don't care about compatibility and will not mind them
to be removed once we depend on C++17. Originally, I planned to use
qmutex_p.h instead, but that header is not self-contained and causes
build errors when we started to include it into libraries other than
QtCore.
Regarding the return value of qt_scoped_lock: Ideally, we'd like to
return a std::scoped_lock, but two things stand in the way: First,
scoped_lock was only added in C++17 (we fall back to lock_guard if
scoped_lock is not available). Second, returning one from a function
requires C++17 guaranteed copy elision, because neither scoped_lock
not lock_guard have a copy ctor. In order for code not to come to
depend on a particular lock class, we return any of lock_guard,
unique_lock or scoped_guard, depending on what the compiler supports,
and therefore wrap the functions in the unnamed namespace to avoid
running into ODR if (private) headers are used from different projects
(autotests, e.g.). By the time we can drop them, however, qt_*_lock
will be semantically 100% identical to their replacements.
Port some initial users.
Change-Id: I2a208ef2a4a533ee8e675812273986460e6b4d00
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
There are two cases:
In a BlockingQueuedConnection, QMetaCallEvent doesn't allocate memory
and instead passes already existing pointers through. A QSemaphore
is used to serialize data access between threads. So the constructor
taking a QSemaphore can be simplified to only accept an existing arg
array.
In a QueuedConnection, QMetaCallEvent needs to make deep copies of
the arguments, and memory needs to be allocated based on the number
of arguments. The previous code put the burden of memory allocation
on the code generating the event, while the memory was free'd by
~QMetaCallEvent. Instead, make it QMetaCallEvent's responsibility
to allocate and free the memory as needed, and adjust the code
generating QMetaCallEvents.
We can allocate the memory for types and pointers to arguments in a
single block, starting with the space for the array of void*, followed
by the space for the array of integers to avoid byte alignment issues.
By pre-allocating the space that's needed by three arguments, we can
avoid all mallocs for the majority of QMetaCallEvents.
Until this change has propagated through qt5.git, we need to keep the
old API that is still used by QtDeclarative around. Once QtDeclarative
has migrated to the new API, it can be removed.
Change-Id: Id7359ffc14897237ea9672dabae9ef199a821907
Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
The three cases - with classic slot, with functor and context object,
and with lambda - are all doing the same work, they just differ in
how they signal the application code about the results.
The detour through an explicitly posted QMetaCallEvent is needed
if we have a functor or lambda; making sure that the temporary
QHostInfoResult object lives in the right thread guarantees that the
event is received in the correct thread, so we can directly call the
functor (as long as the context object is still alive).
Since we guarantee that the QHostInfoResult object lives in the
thread of the receiver, we can simply emit the signal for old-style
signal/slot connections; the regular signal/slot mechanism will do
the work for us.
Change-Id: I584df17df879af01c653e354490c4691dbedd3fa
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Move location of stream ID from a dynamic property to an internal QHash.
Change-Id: I9bab4cbfaebe6a04d54afa7889aac748070e1f2e
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
While it is correct not to call the functor when the context object has
been destroyed, we still need ot clean up the slotObj. It's a low-
probability memory leak: the context object has to disappear while
waiting for a host resolution, and for repeated requests for the same
host the cache takes over anyway.
Task-number: QTBUG-76276
Change-Id: Id9daf391353b8252443f3186a7d504d70c553b24
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Now that all our supported compilers know char16_t, we no longer need
QStringViewLiteral, whose only purpose in life was to turn u"" into
L"" for MSVC < 2015.
Change-Id: I25a094fe7992d9d5dbeb4a524d9e99e043dcb8ce
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Lookups performed via QHostInfoRunnable must not synchronously call
the user-code's receiver objects, as that would execute user-code in
the wrong thread. Instead, post a metacall event through the event
loop of the receiver object, or the thread that initiated the lookup.
This was done correctly for the trivial cases of empty host name or
cached results, so the code generally existed. By moving it from a
global function into a member function of QHostInfoResult, we can
simply access the required data to construct and post the event.
As we process that posted event, we need to check that the context
object (which is already guarded via QPointer) is still alive, if
we had one in the first place. If we had one, and it's deleted, then
abort.
[ChangeLog][QtNetwork][QHostInfo] Functors used in the lookupHost
overloads are now called correctly in the thread of the context object.
When used without context object, the thread that initiates the lookup
will run the functor, and is required to run an event loop.
Change-Id: I9b38d4f9a23cfc4d9e07bc72de2d2cefe5d0d033
Fixes: QTBUG-76276
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
OpenSSL 1.1.x libs must be suffixed otherwise it will use the system ones which on API-21 are OpenSSL 1.0 not 1.1
Fixes: QTBUG-76884
Change-Id: I7d4052be68cf7dc65f74a48da8e1e37182056a5e
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
QNetworkAccessManager::connectToHostEncrypted()/connectToHost()
creates 'fake' requests with pseudo-schemes 'preconnect-https'/
'preconnect-http'. QHttp2ProtocolHandler should handle this
requests in a special way - reporting them immediately as
finished (so that QNAM emits finished as it does in case of
HTTP/1.1) and not trying to send anything.
We also have to properly cache the connection - 'https' or
'http' scheme is too generic - it allows (unfortunately)
mixing H2/HTTP/1.1 in a single connection in case an attribute
was missing on a request, which is wrong.
h2c is more complicated, since it needs a real request
to negotiate the protocol switch to H2, with the current
QNetworkHttpConnection(Channel)'s design it's not possible
without large changes (aka regressions and new bugs introduced).
Auto-test extended.
Fixes: QTBUG-77082
Change-Id: I03467673a620c89784c2d36521020dc9d08aced7
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
QObjects are not even reentrant, but this class must be thread-safe,
so it's always ... icky ... to have to analyze a "thread-safe
QObject", because for all intents and purposes, that's an oxymoron.
The QObject-ness isn't even used, except for defining a private slot,
connected to QCoreApplication::destroyed(). That slot just calls
waitForDone() on QThreadPool, which is a QObject itself, so use it
as the context object for the signal-slot connection, using a lambda
as slot.
So, strip the class of it's base class, convert the private slot to a
lambda and connect to that. Finally, remove the moveToThread() call,
because this new class can be destroyed from any thread, not just the
main one.
Change-Id: I0e33983aa7afd0ad621ece4afd10d9e4adad38c1
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
If it was used in the past, it no longer is, and can't be, because
it's not exported.
Change-Id: Ifb9c353e756add5b57bf0c5706c075bb2eb41d83
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Equality for QNetworkConfiguration is defined by identity of the
Private object, so even if there are QNCs with the same properties,
they will not compare equal. This probably happens when a user roams
through different WLANs (e.g. from work via public transport to home
and then back), and each time the home WLAN pops up, it will beget a
new QNetworkConfiguration, not comparing equal to the previous one.
So, over time, we might collect a sizeable amount of
QNetworkConfiguration objects just sitting in the cache without ever
being able to actually use an associated network session, because they
all have long since expired.
To fix, prune expired network sessions, and thus their associated
QNetworkConfigurations, from the cache. To not run every time, prune
only when this size of the cache is larger than 16 (arbitrary number).
Change-Id: I11a636f45ccf67420f84b1c79a4453a144de7c5c
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
It turns out that the only reason a recursive mutex is used was
because work() was locking it. But work() was only ever called from
functions that already had locked the mutex themselves, and kept it
locked across the call to work(). Clearly mark work() as expecting to
be called with the mutex held (by renaming the function to
rescheduleWithMutexHeld(), then drop the mutex locking from it.
After this change, a non-recursive mutex suffices, so save the memory
allocation and extra complexity involved with recursive mutexes.
Looking at the non-QT_CONFIG(thread) code in rescheduleWithMutexHeld(),
one might be tempted to expect a recursive mutex, since that code
calls QHostInfoRunnable::run() directly, which, in turn, recurses into
QHostInfoLookupManager whence control came, under mutex lock. But in
non-QT_CONFIG(thread) builds, QMutex is but an empty shell, all of its
operations are no-ops, so no possibility for deadlock, either.
Change-Id: Ic01d90c2ed3995b66ccf946d146fdaa6f9af3d8b
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
QHostInfo isn't implicitly shared. The more important to optimize away
copies by providing a move constructor.
Port from QScopedPointer to Q_DECLARE_PRIVATE, as otherwise the move
ctor can't be inline, and we don't implement move ctors out-of-line in
Qt.
[ChangeLog][QtNetwork][QHostInfo] Added move contructor.
Change-Id: I6b63a04e36f63e299205830fdc590ff7e2af338b
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Somehow I didn't test using QNetwork{Status,Connection}Monitor together
with Gui. In qwindowstheme.cpp we call CoInitialize and as such we
cannot use CoInitializeEx with a different thread mode.
Change-Id: If4a4441cc2616371d8b7cda72cfad11187d8f153
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
As noted in RFC7230 Section 6.1 the 'Connection' options are case
insensitive.
https://tools.ietf.org/html/rfc7230#section-6.1
Change-Id: I80b98d82eaa5572d38a6c3f99383bc059ec91c54
Fixes: QTBUG-76990
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
As in QVarLengthArray (c34242c679), this will
probably bite someone someday, so fix it before there's a bug report.
Use ctor delegation to keep code size increase small.
There's also the benefit that default-constructing a QNetworkRequest now
no longer creates an expensive QUrl object just to destroy it unused again.
Add an auto-test.
Change-Id: I5ceb5402ca3946048d695244d1b1c36564a1e80a
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
On the WASM platform, the macro CMDG_NEXTHDR, which is not under our
control, emits a warning about comparing ulong and long with each
other, which -Werror turns into an error:
qnativesocketengine_unix.cpp:1004:24: error: comparison of integers of different signs: 'unsigned long' and 'long' [-Werror,-Wsign-compare]
cmsgptr = CMSG_NXTHDR(&msg, cmsgptr)) {
^~~~~~~~~~~~~~~~~~~~~~~~~~
emsdk/emscripten/1.38.30/system/include/libc/sys/socket.h:286:44: note: expanded from macro 'CMSG_NXTHDR'
__CMSG_LEN(cmsg) + sizeof(struct cmsghdr) >= __MHDR_END(mhdr) - (unsigned char *)(cmsg) \
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Fix by locally disabling the warning.
Change-Id: Ia2ed4318b2ef679b84ac8544835d1e383568ccac
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Conflicts:
configure.pri
Also required s/solid\.color/solidColor/ in a couple of places in:
src/gui/painting/qpaintengine_raster.cpp
Change-Id: I29937f63e9779deb6dac7ae77e2948d06ebc0319
The previous code handled only some error codes, in a very inefficient
way, for some code paths. This change standardizes error handling using
a helper function that maps winsock WSAE* codes to Qt error codes.
The test for connecting to unreachable hosts or ports is now more
generic, and enabled on Windows, where it passes in local tests,
but dependency on network configuration still makes it fragile,
so ignoring some failures without completely skipping the test.
[ChangeLog][Network][Windows] Correctly emit errors when trying to
reach unreachable hosts or services
Change-Id: Icaca3e6fef88621d683f6d6fa3016212847de4ea
Fixes: QTBUG-42567
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Also fix the name mismatch between the Windows- and non-Windows
versions of loadOpenSsl(), which, presumably, were caused by having
two different return values, something easily fixed by defining a
small struct instead of using a QPair.
Some #ifdef'ery saved, and a lot of brittle deletes on early returns.
Change-Id: I77440de2f6fa51759510506ff4ef51917eb5b3ea
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
In auto-test-enabled builds, QHostInfoCache can be enabled and
disabled using qt_qhostinfo_enable_cache() at any time. We cannot rule
out that users use this function, or, indeed, that the auto-test never
gets a threading stress-test. Under the assumption, then, that
QHostInfoCache::enabled can be set by any thread at any time, and is
read by any thread using QHostInfo::lookupHost(), we're presented with
a data race, thus UB.
Fix by making the accesses to QHostInfoCache::enabed atomic. Relaxed
operations are suffcient, as the bool is the only data of interest in
these situations. In particular, access to the cache itself is
protected by the cache's mutex.
We use std::atomic<bool> because QAtomicInteger<bool> doesn't exist on
all implementations, but std::atomic<bool> must. Commit a0faf9e236 set
the precedent that it works.
Change-Id: Ia1766753bb54c5fe8d8447b51a49a96a7a853eef
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Use a scope guard to reliably mark the runnable as finished with the
manager, lest a deleted runnable lingers in the manager's
currentLookups for too long/ever.
Change-Id: I89eff49931d0428f4e75789a0a1188edb1f66220
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
The code contained a copy of std::any_of from a time when we couldn't rely
it's availability in all compilers. We now can, so remove the copy.
Change-Id: I356077f58ae6a48b71f2dd98a2dab4e2acf985c7
Reviewed-by: Anton Kudryavtsev <antkudr@mail.ru>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Simplistic console application with one test for each of the two
classes. Simply tests that we receive the signal when the connection
is disrupted in some way.
This patch also exports the classes for tests/developer builds
so that we can actually link with them.
Change-Id: I8066312274350984110c3f3ad3e94854adca7c2a
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Currently not available for mingw because it's missing all the
interfaces. But should start working once it has them.
Change-Id: I231e8b69e008c5300a83087fe9cd071acd0687cb
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
QNetwork{Status,Connection}Monitor isn't used for QNetworkReplyImpl yet,
so we should connect the signal until it is.
Change-Id: I40f1483608195418d43c442f46184882f64596ff
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Also reduce the pool size from 131 to 17, and use QBasicMutex
instead of recursive ones.
Change-Id: I3bf0374cce5ff2c07427070aba6128a22c9b70e4
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
The use of QMutexPool caused an #ifdef, because, lacking an object,
some OpenSSL function pointer was used as the address required as
input for the mutex pool. Sadly, the names of the functions differ
between OpenSSL versions, thus the need for an #ifdef.
By simply using a QBasicMutex (defined at namespace scope to evade
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79561), the #ifdef can go
away.
Another level of #ifdefs goes away because, even for
non-QT_CONFIG(thread) builds, Q*Mutex etc are defined, just as
no-ops. So we can freely use them without QT_CONFIG(thread) guard,
unlike QMutexPool.
Finally, optimize for the (common?) case of already-loaded libraries
by making symbolsResolved an atomic variable, and checking that before
taking the mutex (double-checked locking, done right).
For reasons of said GCC bug, again, the QBasicAtomic is defined at
namespace scope. And then move the other boolean there for symmetry.
Change-Id: Ic5f44871fb200e5368b9af327e4d1e852fbc586c
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
In one case, added NSDMI to avoid churning the ctor-init-list.
Change-Id: I5587d5cb7e393f60ad29cb5186127304d27d1d46
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The plain bool variable wasDeleted is set to true in the
QHostLookupManager dtor before the call to clear(), which calls
waitForDone() on the thread pool performing the lookups. All tasks on
the thread pool start by checking this variable so as to return early
when destruction is in progress. But the check was outside the
mutex-protected area, so as a non-atomic load, without a
happens-before relation to the write, this is a Data Race, thus UB.
Fix by moving the check past the mutex locking into the critical
section. This way, tasks that were waiting for the mutex after seeing
no wasDeleted before get the message reliably.
This does not introduce a dead-lock, since the call to waitForDone()
is outside any mutex protection leaving a window for the tasks to
obtain the mutex and react on wasDeleted.
Change-Id: Ied4b9daa7dc78295b0d36a536839845c4db2fb78
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Reviewed-by: David Faure <david.faure@kdab.com>
- Replaced QOperatingSystemVersion::WindowsVersion,
QSysInfo::windowsVersion(), QSysInfo::macVersion(),
QSysInfo::MacintoshVersion with QOperatingSystemVersion::current().
- Added QOperatingSystemVersion::WindowsVista for convenience, as it
is used in lots of places.
Change-Id: If9c4ac496005b2e70b5c70be160747afa74b98c1
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
The QNetworkReply implementation for Qt for WebAssembly now supports
usage of the QNetworkAccessManager::sendCustomRequest, making it
possible to send requests with custom verbs.
[ChangeLog][QtNetwork][QNetworkAccessManager] Fixed
QNetworkAccessManager::sendCustomRequest for Qt For WebAssembly.
Fixes: QTBUG-76775
Change-Id: I9394ffef110fce4ed2c877893631bedc7631f71e
Reviewed-by: Lorn Potter <lorn.potter@gmail.com>
When calling waitFor{ReadyRead,Disconnected} it will wait for data but
if the data is already received and the read notification has been
queued (and there's no more data coming in) it will return false.
By checking if a read notification has been queued and then handling
this we can easily take care of this scenario.
Fixes some flaky tests which missed the read data in waitForDisconnect
and similar.
Fixes: QTBUG-38385
Change-Id: Ic05d59883c1175783e56ff1822b6636c35aec874
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
If we end up in the connected state then we should pass on any
remaining data immediately instead of waiting until the next time we get
a read notification.
The other `case`s in the switch might be able to do something similar,
but I don't want to introduce that logic now in case it breaks something
else, the Connected branch is small and simple to deal with.
Should severely reduce flakiness with socks proxy in CI under pressure.
Task-number: QTBUG-76367
Change-Id: I0965d4c62a29a25ce6b8dd60862a464279aef0b4
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Reviewed-by: Ryan Chu <ryan.chu@qt.io>
Because it is.
Change-Id: I8d5204c30884b2c8656615a7d82428c539672d28
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Ville Voutilainen <ville.voutilainen@qt.io>
This helps to fix build without feature.properties
Change-Id: Ia1fd2a1ca88105048e75694874058bb1292899a0
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
not to resolve merge conflicts on every 5.13->dev merge.
Change-Id: Id41a7efff52148fe46bedcde828646694fd1764d
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Fix warnings showing in network tests:
tst_QNetworkReply::headFromHttp(rfc+socks):
Using QCharRef with an index pointing outside the valid range of a QString. The corresponding behavior is deprecated, and will be changed in a future version of Qt.
introduced by qtbase/c2d2757bccc68e1b981df059786c2e76f2969530 (5.14).
Replace index access by QByteArray::append().
Change-Id: I0b4aed563d076237d5f9cc6aa438c7502eb3568c
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
The loop iterates over a collection of QExplicitSharedDataPointer, which
traditionally doesn't propagate const. In Qt 6, it will, so prepare the
code for this change, by taking the loop variable by non-const reference.
Since the loop is followed by container.clear(), make it a consume-loop
by looping over qExchange(container, {}).
Change-Id: If12ab005544183598fd76a0c486b2df1582710d5
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Generate the ${MODULE}-android-dependencies.xml for the androiddeployqt
tool. This will ensure all the right plugins and dependencies are
packaged when generating the apk.
This change also changes the visibility for executable to default/public.
Not having this will cause the application to crash as we can't locate
main() in the executable (shared library).
Additionally pro2cmake conversion script has been updated to perform
the required conversions for the Android settings.
Finally, the 6 projects in QtBase that have Android dependencies have
been updated with the new script and the step that produces the xml
files in run at the end in QtPostProcess.cmake.
Change-Id: I9774ba1b123bc11cae972fa37054ef2c51988498
Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
When using a http proxy (and presumably other proxies) we might have
failed/aborted (aka "finished") the request and _then_ receive a
"proxy authentication required" message from the proxy. In this case
there is no spdy/http2 reply in the queue, so asserting is wrong.
Change-Id: Id9b76b580299f6a6cd6efad62d6aaf63183816fb
Fixes: QTBUG-76426
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
In some scenarios with QNAM we call socket->close, leading to a flush,
leading to an error, leading to another error emission... To work
around this scenario we stop trying to close the socket if the network
channel is already closing.
Change-Id: Id15504f476484ce61f11ba83a5755ceb5f581f9b
Fixes: QTBUG-76567
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
and also plain wrong comments: since we don't set verification callback
on a store - we don't have to lock (our q_X509Callback never gets called).
This change should simplify the merge with change I have in 5.12 (where
I completely got rid of locking). Since I don't care about exact errors
found (relying on the fact it's the same chain of certs we check in
SSL_connect/SSL_accept), for now we don't try to extract them from
OCSP_basic_verify. In fufure, if these chains are different, we
can create a temporary store (see how it's done in 'verify', for example)
and set a VF callback on this store.
Change-Id: I4a36e19836d19c2ea95c869dcfe85f49fe723ff0
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
This changes many different CMake places to mention Qt6 instead of
Qt5.
Note that some old qt5 cmake config files in corelib are probably not
needed anymore, but I still renamed and kept them for now.
Change-Id: Ie69e81540386a5af153f76c0242e18d48211bec4
When our QSslSocketBackendPrivate (OpenSSL backend) was developed,
the ancient versions of OpenSSL did not have an API needed to pass
an application-specific data into verification callback. Thus the
developers resorted to the use of global variables (a list with errors)
and locks. Some of our auto-tests use QNAM and in-process server.
Whenever the client (essentially qhttpthreadeddelegate) and the server
live in different threads, any use of 'https' is dead-lock prone,
which recent events demonstrated and which were previously observed
but not understood properly (rare occasions, not always easy to
reproduce). Now we fix this for good by removing locking.
There are two places (in 5.12) where these locks are needed:
1. Before calling SSL_connect/SSL_accept (handshake) - here
we reuse the same trick we do in PSK callback ('SSL' has
an external data set, and it's 'this', meaning an object
of type QSslSocketBackendPrivate).
2. The static member function 'verify', here we do not have
'SSL', but we have our temporary 'X509_STORE', to which
we can directly attach an external data - a pointer to
a vector to collect verification errors.
Note, this change assumes that OpenSSL Qt is build/linked
against is at least of version 1.0.1 - we set external data
on SSL unconditionally (no version checks).
Fixes: QTBUG-76157
Change-Id: I05c98e77dfd5fb0c2c260fb6c463732facf53ffc
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
There were some changes done in pro2cmake, so regenerate the main
qtbase modules to keep everything up to date.
Some new whitespace got added.
Some special cases were removed (ZLIB).
Some opengl code got moved around.
Some plugin types were added.
And some other minor things.
Change-Id: Ie8cba4a9aa6b4b163c39d6cf546ea9e0bfc05c01
Reviewed-by: Qt CMake Build Bot
Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
Fix build configuration for qdnslookup implementations for android as
UNIX is also set when building for Android. We should only have the more
specific conditions enabled.
Change-Id: Ib891df34e5569a20b03ce270e1eab080325549b9
Reviewed-by: Qt CMake Build Bot
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
The code didn't really use the queue as a queue, because it enqueued
one at a time, but dequeued potentially many. Even if it actually
decays to actual queue behavior: for the small number of elements
expected, a vector would still be preferable over the per-element
allocation performed by QQueue.
Because std::vector doesn't have pop_front(), I took the liberty of
making the dequeue code easier to read by using find_if with an
aptly-named lambda.
Change-Id: I09bd7ede50cb8ab5af5d1fc2ede37a3731f67070
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Semi-automated, just needed ~20 manual fixes:
$ find \( -iname \*.cpp -or -iname \*.h \) -exec perl -pe 's/(\.|->)load\(\)/$1loadRelaxed\(\)/g' -i \{\} +
$ find \( -iname \*.cpp -or -iname \*.h \) -exec perl -pe 's/(\.|->)store\(/$1storeRelaxed\(/g' -i \{\} +
It can be easily improved (e.g. for store check that there are no commas
after the opening parens). The most common offender is QLibrary::load,
and some code using std::atomic directly.
Change-Id: I07c38a3c8ed32c924ef4999e85c7e45cf48f0f6c
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Under no circumstance must Qt code define a non-public qHash()
overload of a public Qt type. If you can't provide a public qHash()
for the type, you can't use QHash. Period. This is because if we don't
provide a qHash() function, the user will have to provide one, and
then we have an ODR violation, iow: UB.
So, port away from QHash and qHash and use std::unordered_map, which
allows to pass a custom hash function - just what we needed.
Also fix other issues of the code:
- Use a function object as the deleter of the QSharedPointer, to allow
inlining the function call.
- Avoid double lookup in case of a cache miss.
- Don't use direct initialization syntax for pointer variables. That's
just ... weird.
Change-Id: Ica8bff728c1bd1cbb2fb43aa03a99bcb7ac2f7cc
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Assigning past the end used to be supported, but isn't a good way to
do the job. Exposed by a qtxmlpatterns test that exercised this code,
which gets a warning thanks to Giuseppe's recent changes to QCharRef.
Task-number: QTBUG-76070
Change-Id: Ic254c7ffce60e3b2aafce76ab03ea5db8c68fafc
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
this also optimizes network post method handling
Task-number: QTBUG-75660
Change-Id: Ibb0d01f2cc2b2bc7802598c4f6f04b04882c12ca
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
There are no QLatin1String overloads, so using QLatin1String as the
format string produced a temporary QString. Use QStringViewLiteral
instead.
Change-Id: I682eb8cfaa98fdcfb491edd290460db636a98a9f
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
1) Remove a use-after-move. It was benign because the move didn't
actually trigger move assignment, as the original object was const.
2) Remove a usage of insert(hint), as the hint was always end(),
and there is no reason to believe that that's the insertion place.
Change-Id: I71aac8cdc9fb85b6ecef3695bae7b21f022bb60b
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
src/corelib/tools/qstring.cpp:9505:clang found diagnostics parsing \fn int QLatin1String::indexOf(QLatin1String l1, int from Qt::CaseSensitivity cs) const error: expected ')' error: out-of-line definition of 'indexOf' does not match any declaration in 'QLatin1String'
src/corelib/tools/qstringview.cpp:723:clang found diagnostics parsing \fn qsizetype QStringView::indexOf(QLatin1String l1, qsizetype from Qt::CaseSensitivity cs) const error: expected ')' error: out-of-line definition of 'indexOf' does not match any declaration in 'QStringView'
src/corelib/tools/qstringview.cpp:822:Unknown command '\t'
src/corelib/tools/qstringview.cpp:825:Unknown command '\t'
src/corelib/tools/qstringview.cpp:826:Unknown command '\t'
src/corelib/tools/qstringview.cpp:831:Unknown command '\t'
src/corelib/global/qnamespace.qdoc:2448:Undocumented enum item 'MarkdownText' in Qt::TextFormat
src/corelib/tools/qstringmatcher.cpp:183:No such parameter 'pattern' in QStringMatcher::QStringMatcher()
src/network/ssl/qsslerror.cpp:58:Undocumented enum item 'CertificateStatusUnknown' in QSslError::SslError
src/gui/kernel/qevent.cpp:5321:Undocumented parameter 'screenOrientation' in QScreenOrientationChangeEvent::QScreenOrientationChangeEvent()
src/gui/kernel/qevent.cpp:5321:Undocumented parameter 'screen' in QScreenOrientationChangeEvent::QScreenOrientationChangeEvent()
src/gui/kernel/qevent.cpp:5321:No such parameter 'orientation' in QScreenOrientationChangeEvent::QScreenOrientationChangeEvent()
src/gui/text/qtextformat.cpp:532:Undocumented enum item 'BlockCodeLanguage' in QTextFormat::Property
src/gui/text/qtextformat.cpp:532:Undocumented enum item 'BlockQuoteLevel' in QTextFormat::Property
src/gui/text/qtextformat.cpp:532:Undocumented enum item 'BlockMarker' in QTextFormat::Property
src/gui/text/qtextdocument.cpp:3294:Undocumented parameter 'features' in QTextDocument::toMarkdown()
src/gui/painting/qcolorspace.cpp:659:Undocumented parameter 'colorSpace1' in QColorSpace::operator!=()
src/gui/painting/qcolorspace.cpp:659:Undocumented parameter 'colorSpace2' in QColorSpace::operator!=()
src/gui/painting/qcolorspace.cpp:659:No such parameter 'colorspace1' in QColorSpace::operator!=()
src/gui/painting/qcolorspace.cpp:659:No such parameter 'colorspace2' in QColorSpace::operator!=()
examples/widgets/doc/src/icons.qdoc:269:Command '\snippet (//! [43])' failed at end of file 'widgets/icons/iconpreviewarea.cpp'
src/widgets/styles/qstyle.cpp:2026:Undocumented enum item 'SP_DialogRetryButton' in QStyle::StandardPixmap
src/widgets/styles/qstyle.cpp:2026:Undocumented enum item 'SP_DialogYesToAllButton' in QStyle::StandardPixmap
src/widgets/styles/qstyle.cpp:2026:Undocumented enum item 'SP_DialogIgnoreButton' in QStyle::StandardPixmap
src/widgets/styles/qstyle.cpp:2026:Undocumented enum item 'SP_DialogNoToAllButton' in QStyle::StandardPixmap
src/widgets/styles/qstyle.cpp:2026:Undocumented enum item 'SP_DialogAbortButton' in QStyle::StandardPixmap
src/widgets/styles/qstyle.cpp:2026:Undocumented enum item 'SP_RestoreDefaultsButton' in QStyle::StandardPixmap
src/widgets/styles/qstyle.cpp:2026:Undocumented enum item 'SP_DialogSaveAllButton' in QStyle::StandardPixmap
src/testlib/qtestcase.qdoc:439:Undocumented parameter 'TestClass' in QTest::QTEST_HIGHDPI_SCALING_MAIN
src/testlib/qtestcase.qdoc:452:Undocumented parameter 'TestClass' in QTest::QTEST_NO_HIGHDPI_SCALING_MAIN
Change-Id: Ib0e9bf81c5caaa6b1fc644ac92085af47c600e0e
Reviewed-by: Shawn Rutledge <shawn.rutledge@qt.io>
Reviewed-by: Anton Kudryavtsev <antkudr@mail.ru>
Reviewed-by: Paul Wicking <paul.wicking@qt.io>
Conflicts:
src/corelib/tools/qlocale_data_p.h
(Regenerated by running the scripts in util/local_database/)
src/gui/opengl/qopengltextureuploader.cpp
Done-With: Edward Welbourne <edward.welbourne@qt.io>
Done-With: Allan Sandfeld Jensen <allan.jensen@qt.io>
Change-Id: I12df7f066ed0a25eb109f61c4b8d8dea63b683e2
Make sure "this" still exists when we're done sending the
readNotification.
The crash manifested itself when connecting to certain websites as they
would reply with status 403, then close the connection. On our end we
would then handle this "remote host closed" followed by handling the
data we received. The http code handles the data successfully and sees
we are done and there is nothing more to do, so it closes the
connection. Which leads to closing QAbstractSocket, which closes
native socket again and then deletes it.
Fixes: QTBUG-75620
Change-Id: I233c67f359aa8234f1a2c4ea9463108b08c9165f
Reviewed-by: Oliver Wolff <oliver.wolff@qt.io>
Try to replace all wordings like '.. to 0' with '.. to \nullptr'. Also
checked for 'null pointer' and similar.
Change-Id: I73341f59ba51e0798e816a8b1a532c7c7374b74a
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
They are going to be deprecated.
Change-Id: Ib021aad108dc021df76ae21d1db6c8a1a734893d
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Merge all data related to mapping libraries into one data structure
in helper.py.
Use that data for everything related to library mapping.
This change enables way more features now like e.g. adding find_package
calls into generated files.
Change-Id: Ibbd2a1063cbeb65277582d434a6a672d62fc170b
Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
Do not set properties of packages to type OPTIONAL. That is the default
anyway.
Update generator script and generated files.
Change-Id: I7a4d043b69c93ce8c2929a2e27ac6a07e4e6d8cc
Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
Instead of generating .html page for each file in an example
project, generate links to code.qt.io, under the correct
path and branch, where the user can browse the example source.
Store all URLs under QT_INSTALL_DOCS/config where other qt5
submodules can access them. The repository name appears in
the URL, so we cannot define a single URL for all modules.
Task-number: QTBUG-74391
Change-Id: I63d4d6d2c352877797b1ee8e057d48c0cd789bff
Reviewed-by: Paul Wicking <paul.wicking@qt.io>
If this callback is not set then OpenSSL will call the callback
used for <= TLS 1.2 unconditionally when connecting. If using PSK it
will call it again later once the preshared key is needed.
We don't currently handle the TLSv1.3 PSK, but we definitely should.
But for now we can work around it - when psk_use_session_callback is
called we simply change the PSK callback to a dummy function whose only
purpose is to restore the old callback.
This is mostly done to keep behavior the same as it is now for users
(and to keep our tests running).
Later we can add a new signal and handle this new feature properly.
Reviewed-by: Simo Fält <simo.falt@qt.io>
(cherry picked from commit d8efc8d718)
Task-number: QTBUG-67463
Change-Id: I4aca4ae73ec4be7c4f82a85e8864de103f35a834
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
From Qt 5.12.5 and 5.13.0 on we ship OpenSSL 1.1.x libraries with Qt
through the binary installers, as Qt is a general purpose toolkit and
as such not subject to import/export restrictions.
However, application developers still have to take care of the
relevant requirements and compliance before distributing OpenSSL.
Change-Id: I1c3622116eadda270d638becfa23a5493976e919
Fixes: QTBUG-75814
Reviewed-by: Paul Wicking <paul.wicking@qt.io>
Reviewed-by: Akseli Salovaara <akseli.salovaara@qt.io>
also fix data progress
Task-number: QTBUG-75489
Change-Id: I5222fda64d258a6ae78ba0ca20194b81c289c27e
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
Previously there were two issues:
- A QNetworkReply could be aborted but be in NoError state.
(GOAWAY frame with 0 as error)
- Streams in a connection would be aborted prematurely when a GOAWAY
frame with a lastStreamId of 2^31-1 was received.
Fixes: QTBUG-73947
Change-Id: Iddee9385c1db3cc4bb80e07efac7220fff787bf3
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
This change introduces a new function called qt_find_package()
which can take an extra option called PROVIDED_TARGETS, which
associates targets with the package that defines those targets.
This is done by setting the INTERFACE_QT_PACKAGE_NAME and
INTERFACE_QT_PACKAGE_VERSION properties on the imported targets.
This information allows us to generate appropriate find_dependency()
calls in a module's Config file for third party libraries.
For example when an application links against QtCore, it should also
link against zlib and atomic libraries. In order to do that, the
library locations first have to be found by CMake. This is achieved by
embedding find_dependency(ZLIB) and find_dependency(Atomic) in
Qt5CoreDependencies.cmake which is included by Qt5CoreConfig.cmake.
The latter is picked up when an application project contains
find_package(Qt5Core), and thus all linking dependencies are resolved.
The information 'which package provides which targets' is contained
in the python json2cmake conversion script. The generated output of
the script contains qt_find_package() calls that represent that
information.
The Qt5CoreDependencies.cmake file and which which dependencies it
contains is generated at the QtPostProcess stop.
Note that for non-static Qt builds, we only need to propagate public
3rd party libraries. For static builds, we need all third party
libraries.
In order for the INTERFACE_QT_PACKAGE_NAME property to be read in any
scope, the targets on which the property is set, have to be GLOBAL.
Also for applications and other modules to find all required third
party libraries, we have to install all our custom Find modules, and
make sure they define INTERFACE IMPORTED libraries, and not just
IMPORTED libraries.
Change-Id: I694d6e32d05b96d5e241df0156fc79d0029426aa
Reviewed-by: Tobias Hunger <tobias.hunger@qt.io>
Using getaddrinfo, which implements RFC 6724, implies that addresses
that are not needed will be trimmed. In particular, IPv6 addresses are
often not returned.
Also move the implementation detail documentation down in the text, it's
a detail with little relevance for the usage of the class, but makes for
a good opener regarding this behavior.
Change-Id: I516a64f0b39a6a06621a63c1d5236544b7758049
Fixes: QTBUG-31865
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Each module that has publically declared classes or structs
that are not meant to be documented is given a dontdocument.qdoc
file to tell qdoc that these classes are not meant to be
documentented. Then qdoc will not print warnings about missing
\class comments for these classes and structs.
Change-Id: I9195f0b546032e1c7642c9da34d85a0a4a9bfb08
Reviewed-by: Paul Wicking <paul.wicking@qt.io>
Add a helper function to QtBuild that generates Foo_nolink versions
of Foo library targets.
Map 'Foo/nolink' libs found in qmake to Foo_nolink.
Automatically run helper function to create _nolink targets as
part of extend_target.
Change-Id: I4c23ea68b3037d23c9a31d4ac272a6bd0565f7c0
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
QNetworkDatagram was introduced together with
these methods in Qt 5.8 (commit 4da2dda2aa)
Change-Id: I454c26ebf6f94988cada8ac9315db1d43a31a595
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The current code only tries to load the required functions from
LIBRESOLV_SO (if defined) and resolv, but on FreeBSD they are in libc:
https://www.freebsd.org/cgi/man.cgi?query=res_query&sektion=3&apropos=0&manpath=freebsd
This commit changes the code so that, after failing to load the
non-existent libraries, it attempts to load the functions with dlsym()
using the special handle RTLD_DEFAULT, which searches for the specified
symbol in the loaded libraries.
This is a follow-up to 8eeb5150ed.
Change-Id: I19d90b0ca8703398bf4f5f4edd5ae31e346ef251
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The current code only tries to load the required functions from
LIBRESOLV_SO (if defined) and resolv, but on FreeBSD they are in libc:
https://www.freebsd.org/cgi/man.cgi?query=res_query&sektion=3&apropos=0&manpath=freebsd
This commit changes the code so that, after failing to load the
non-existent libraries, it attempts to load the functions with dlsym()
using the special handle RTLD_DEFAULT, which searches for the specified
symbol in the loaded libraries.
Task-number: QTBUG-74844
Change-Id: If97aaae233cabbfa01c30d26d9a7fb01ec3ff5c2
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The actual variable that contains the architecture is
TEST_architecture_arch. TEST_architecture only contains the value
if the test was performed or not.
Fix the conversion script and all the generated files.
Change-Id: Icb3480832cab894948f4fef03b8bc8187cab6152
Reviewed-by: Tobias Hunger <tobias.hunger@qt.io>
After 'h2c' mode was implemented with the proper protocol
upgrade, the previously working 'direct connection' mode
was lost for clear text connections due to the erroneous
logic in the constructor: having !channel->ssl does not
necessary mean we started with HTTP/1.1 request, including
protocol upgrade header; it can also mean we connected a
plain socket and immediately sending h2 frames, without
any H2 negotiation at all.
Fixes: QTBUG-74765
Change-Id: Ice466d6bffb40048b7ab46fb064f2d3d795a12aa
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
And export the required symbols in OpenSSL so we can run the test there
as well even if it's not needed for any functionality.
Change-Id: I4246d2b0bbdd42079d255f97f3c66ce8bb37390b
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Prevent namespace collisions and make sure Qt functions
are grouped together.
Change-Id: I217188ee93e4300e273d10a79d6014179fc5a1ef
Reviewed-by: Lorn Potter <lorn.potter@gmail.com>
In the QNetworkAccessManager machinery we would treat "no-cache" as if
it meant "don't cache" while in reality it means "don't return these
cached elements without making sure they're up-to-date"
At the same time as this change is made let's add test data for
"no-store", which replaces the "no-cache" test data.
Fixes: QTBUG-71896
Change-Id: Ieda98f3982884ccc839cac2420c777968c786f6e
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Reviewed-by: Mikhail Svetkin <mikhail.svetkin@qt.io>
If this callback is not set then OpenSSL will call the callback
used for <= TLS 1.2 unconditionally when connecting. If using PSK it
will call it again later once the preshared key is needed.
We don't currently handle the TLSv1.3 PSK, but we definitely should.
But for now we can work around it - when psk_use_session_callback is
called we simply change the PSK callback to a dummy function whose only
purpose is to restore the old callback.
This is mostly done to keep behavior the same as it is now for users
(and to keep our tests running).
Later we can add a new signal and handle this new feature properly.
Task-number: QTBUG-67463
Change-Id: I4aca4ae73ec4be7c4f82a85e8864de103f35a834
Reviewed-by: Simo Fält <simo.falt@qt.io>
If the app is finished and going to shutdown,
qNetworkConfigurationManagerPrivate() returns nullptr.
Change-Id: I01915021d8698802b3a1d0dee43203cd3d4aba74
Task-number: QTBUG-76090
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
It's never set anywhere, neiher in QtBase nor in any other Qt module.
So, remove.
Change-Id: If616d350a1c1c74f6f3e87f8cd98ccb3bff5cf70
Reviewed-by: Lorn Potter <lorn.potter@gmail.com>
- Fix qmake build
- Fix QtNetwork moc-ing, by including the moc files
inside the cpp files
- Fix sql odbc plugin by including QT_PLUGIN define
- Fix Boostrap to link against the Platform target, to get the
correct Unicode and WIN64 defines.
- Fix vulkan headers to be found
- Fix freetype bzip and png unresolved symbols / linker issues
when building minimal platform plugin (also need to make
sure to use the vcpkg toolchain instead of CMAKE_PREFIX_PATH
because then find_package is overridden, which does magic
to properly propagate static library dependencies).
- Fix qfilesystementry test not to be built without private
tests feature (it led to undefined symbols issues).
- Make sure to remove QT_NO_CAST_TO_ASCII define when building
QtCore, so that the qstringbuilder3 test builds
successfully.
Task-number: QTBUG-74140
Change-Id: I353d08392b604d55f8e62cdd8696d1e19a3c084a
Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
The new files represent the json features more closely.
This helps with various build issues as well, like dynamicgl
feature not being present before.
Change-Id: I433592f48e0550700226a635492a3ce15af743c1
Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
We can pass these through to to the browser, which
will resolve them relative to the current html file.
Task-number: QTBUG-74289
Change-Id: I595f30456de55da4f1432fb997d381940f51c5f9
Reviewed-by: Lorn Potter <lorn.potter@gmail.com>
src/network/ssl/qsslsocket.cpp:1501: (qdoc) warning: Can't link to 'QSslConfiguration::defaultCaCertificates()'
src/network/ssl/qsslsocket.cpp:1482: (qdoc) warning: Can't link to 'QSslConfiguration::defaultCaCertificates()'
src/network/ssl/qsslsocket.cpp:1513: (qdoc) warning: Can't link to 'QSslConfiguration::defaultCaCertificates()'
src/network/ssl/qsslsocket.cpp:1119: (qdoc) warning: Can't link to 'QSslConfiguration::setDefaultCiphers()'
src/network/ssl/qsslsocket.cpp:1119: (qdoc) warning: Can't link to 'QSslConfiguration::defaultCiphers()'
src/gui/opengl/qopengltexture.cpp:4137: (qdoc) warning: Can't link to 'setComparisonFunction()'
src/widgets/dialogs/qfilesystemmodel.cpp:1215: (qdoc) warning: Undocumented return value (hint: use 'return' or 'returns' in the text
src/widgets/graphicsview/qgraphicswidget.cpp:1229: (qdoc) warning: Undocumented return value (hint: use 'return' or 'returns' in the text
src/widgets/graphicsview/qgraphicsscene.cpp:3313: (qdoc) warning: Undocumented return value (hint: use 'return' or 'returns' in the text
src/gui/painting/qpaintengine_raster.cpp:3438: (qdoc) warning: Undocumented return value (hint: use 'return' or 'returns' in the text
src/widgets/widgets/qtextedit.cpp:2544: (qdoc) warning: Overrides a previous doc
src/widgets/widgets/qplaintextedit.cpp:2932: (qdoc) warning: (The previous doc is here)
Change-Id: I0c68c59a87eb6a5d9e974f857af6aca0c6e0672d
Reviewed-by: Paul Wicking <paul.wicking@qt.io>
Needed to disable QT_NO_UNSHARABLE_CONTAINERS, as this
triggers asserts.
QMetaType also has some Qt 6 specific code disabled to
get things to compile.
Fix various details in autotests to accommodate for
the changes with Qt 6.
Add a workaround for black lists on macos, where
QSysInfo::productType() now returns 'macos' and not
'osx' anymore.
Change-Id: Ie26afb12a2aac36521472715934a7e34639ea4d0
Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
Users should use range constructors instead to do the conversion.
Keep conversion methods between QList and QVector as these will turn
into a no-op in Qt 6, whereas forcing people to use range constructors
would lead to deep copies of the data.
Change-Id: Id9fc9e4d007044e019826da523e8418857c91283
Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
Private classes to replace broken or even not working at all 'session management'
and 'bearer manager' (on at least two major platforms we support). This
implementation is macOS/iOS-specific and uses SystemConfiguration framework,
or more precisely SCNetworkReachability's part of it.
Task-number: QTBUG-40332
Change-Id: Iac5f44c4063c4092b93b8cf2bde3fb2c524855b3
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Remove remaining handling of missing support for rvalue refs.
Change-Id: I78bab8bccfeeb9c76f464f345874364a37e4840a
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The implementations were practically identical, so we can just move the
code into qhostinfo.cpp, cleaning up things along the way.
Since QHostInfoAgent is an internal class, add the shared code as
additional static helper functions. And since half the code already used
QCoreApplication::translate anyway, we can remove the QObject
inheritance which was only added for getting a tr().
Change-Id: I58eafbdc3e7d06d2e898486a1add63cd63d98c96
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Ignores errors produced by the HELP command. It will continue
executing the commands sent to QFtp when a server does not have a
HELP command implemented. Commands SIZE, MDTM, and PWD are not going
to be used if the HELP command failed.
[ChangeLog][QtNetwork][QNetworkAccessManager] Don't fail when FTP does
not implement the HELP command.
Task-number: QTBUG-69477
Change-Id: I0ebd51b134535730c6bef83de1abf1a427b8d2ce
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
It's what's returned when calling request() on a QNetworkReply, but for
QNetworkReplyImpl (the implementation used for e.g. FTP) it was not set,
so it returned a default object.
Change-Id: Id82f7920e4268c1a7743e681fc503bcd28889b6e
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Markus Goetz (Woboq GmbH) <markus@woboq.com>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Since we require getaddrinfo and freeaddrinfo, those symbols are
not resolved dynamically, and will never be null, which causes compile
errors.
Correctly appending a string literal shuts up compiler warnings.
Change-Id: I0096dca6a77ad86e0c89890444b8d7bbb38a40f2
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QRegExp includes can be found in several files where there's not even a
use of the class. This patch aims to avoid needless includes as well as
follow the "include only what you use" moto.
This patch removes a QRegExp include from the QStringList header which
means that there is likely going to be code breaking since QStringList
is used in many places and would get QRegExp in.
[ChangeLog][Potentially Source-Incompatible Changes] qstringlist.h no
longer includes qregexp.h.
Change-Id: I32847532f16e419d4cb735ddc11a26551127e923
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Timur Pocheptsov
Mårten Nordheim
Sona Kurazyan
Alexandru Croitor
Liang Qi
Friedemann Kleint
Qt Forward Merge Bot
Rainer Keller
Karsten Heimrich
Lorn Potter
Paul Olav Tvete
Allan Sandfeld Jensen
Marc Mutz
Volker Hilsheimer
Lars Knoll
Tasuku Suzuki
BogDan Vatra
Felix Barz
Simon Hausmann
Leander Beernaert
Cristian Adam
Giuseppe D'Angelo
Edward Welbourne
Christian Ehrlicher
Tobias Hunger
Topi Reinio
Christian Kamm
Martin Smith
Andre Hartmann
Davide Beatrici
Alexey Edelev
Morten Johan Sørvig
Val Doroshchuk
Kevin Funk
Joerg Bornemann
Jesus Fernandez
Samuel Gaist