| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
| |
Tab information was hidden by the closing buttons.
Closing is implented by using doAction and actionText methods, as it's not possible to
use QAccessibleActionInterface to send actions to child widgets in complex widgets.
Change-Id: I0f55d5084737fa13c81df402f5dd54f11e164f8a
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
NTFS mount points are not treated as symlinks, because they might
not have a link target.
This is the case when a volume is mounted as a single mount point
without a drive letter.
This patch fixes building Qt in an NTFS mount point.
Task-number: QTBUG-20431
Change-Id: Ie2e15212e1a7ca7fa0067b7ca8857e243e42c21a
Reviewed-by: Thomas Hartmann <Thomas.Hartmann@nokia.com>
|
|
|
|
|
| |
Change-Id: Iea6ca8fe764de85a7bb3d955384e72762163db1e
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fix a couple of cases where WildcardUnix was not matching when the
string was just [] or ended with a \.
The testWildcardEscaping() test has been extended to account for these
two cases too.
Task-number: QTBUG-20897
Change-Id: I9ddb10f5a1150427e8ccefae0dc87a3d4e104854
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
Reviewed-by: Giuseppe D'Angelo <dangelog@gmail.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
|
|
| |
On Mac OS X, if all text in the QLineEdit was selected and then deleted, cursor
visibility was not updated, and so the cursor remained hidden. Fixed
to update cursor visibility also when the text is empty.
Task-number: QTBUG-13169
Change-Id: Id52a20b07bb96609a78c42eb630ee2b20ed7cbcb
Reviewed-by: Jiang Jiang <jiang.jiang@nokia.com>
|
|
|
|
|
|
|
|
|
|
| |
In this case these string are not input through a composing process
(no marked text is set) but they should not be handled by Qt key
event handler.
Change-Id: I9ea96c29db69d51abc6821e72a3228158012bebc
Task-number: QTBUG-23216
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
|
|
|
|
|
|
|
|
|
|
| |
the change is bogus; people were actually building it successfully.
This reverts commit 0c70be512bfc40e8a2a09f60a66b7ecb114e748e
Change-Id: I641fac34b78d6c23ead143d34a8ae15bf191cae3
Reviewed-by: Jonathan Liu <net147@gmail.com>
Reviewed-by: Ray Donnelly <mingw.android@gmail.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
|
|
|
|
|
|
|
|
|
| |
Since the coreservices test should be part of the QPA ones after all,
this reverts d317182e and fixes the original problem correctly by
checking if qpa is on before testing for coreservices
Change-Id: I476f56924ff2d9bbc2f290563aaff9528fe72766
Reviewed-by: Eike Ziller <eike.ziller@nokia.com>
|
|
|
|
|
|
|
|
|
|
| |
The GPU in question supports GL 2 but not framebuffer objects. Since we
anyway have a font rendering path that doesn't use FBOs we might as well
not require framebuffer objects in order to use the GL 2 engine.
Task-number: QTBUG-22483
Change-Id: I2a80343fedda276e73e603ffe54edff58801af5b
Reviewed-by: Kim M. Kalland <kim.kalland@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Specifying qt_ntfs_permission_lookup++ in application code didn't
make qfilesystemengine_win.cpp respect Windows ACL as it was supposed
to. This was because GetTokenInformation for TokenUser failed always in
resolveLibs() function, because the TOKEN_USER struct that was given to
it wasn't large enough to contain both TOKEN_USER and SID structs that
GetTokenInformation wants to return in this case.
Fixed by calling GetTokenInformation twice, first to determine the
required size, and then another time to get the actual token info.
Additionally, the SID returned as part of the token info needs to be
stored for the lifetime of the application, as the TRUSTEE_W struct
has a pointer to it (currentUserTrusteeW).
The worldTrusteeW initialization also required a change to properly
store the SID.
Note: The dynamic resolution of FreeSid and other SID manipulating
functions doesn't appear to be necessary, as they are found on the
same ifdef level (in winbase.h) as the GetTokenInformation, which
already isn't dynamically resolved.
Task-number: QTBUG-247
Change-Id: If5b43faeded351c45b52a63854a10d9f618e33ef
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Change 54613aec was using the value 1 for QGLFormat::samples when it
wanted multisampling off. But this was actually resulting in
multisampling being turned on. It should have been using 0, which it
now does.
Task-number: ou1cimx1#951223
Change-Id: Ie68367f083f7ab4bae4143a59996c71add3e5c06
Reviewed-by: Pasi Pentikäinen <ext-pasi.a.pentikainen@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Crash is a regression from the commit
7cd29ef169b94b0182b4207b37715f7427e4e4d6.
The commit 7cd29ef169b94b0182b4207b37715f7427e4e4d6 creates two
fep contexts, another one is created while first one being created.
When application exit, other fep context is deleted
after coeEnv destroyed. Fep context destructor uses method from
coeEnv and when second fep Context deleted application crashes.
To fix the crash, flaged fep context creation while being created.
Change-Id: I3079179df2c3a59a0117d5de775116e06d7e3cff
Reviewed-by: Satyam Bandarapu <ext-satyam.bandarapu@nokia.com>
Reviewed-by: Murray Read <ext-murray.2.read@nokia.com>
Reviewed-by: Pasi Pentikäinen <ext-pasi.a.pentikainen@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The Qt drawing system is not designed to handle failures in beginPaint
and endPaint. But the Symbian port is creating EGL contexts and surfaces
on demand in these functions. These can fail, eg in the defect reported
the app gets a visibile=true event for such a short period that it
fails to create the surface before it is invisible again. These failures
have to be coped with.
Brush image drawing is failing if the app is asked to draw
while in the background, because the EGL surface cannot be created.
After this, if the app comes to the foreground at the wrong time, the
failed drawing is shown on the screen as a flat colored area.
A failedToAlloc variable in the openvg paint engine was locking in this
failed drawing, so that it would never redraw. This is fixed by removing
all use of that variable.
Failed surface creation during a draw in response to a visibility
change event was also leaving the window backing store with bad content.
This is now cleared on a surface creation failure so that drawing
is retried when the next visibility change event arrives.
The retry mechanism for creating a surface on failure also
makes the problem circumstances more likely to occur and can block the
app unnecessarily. Now, instead of blocking for 1 second every time on
failed surface creation, it only blocks on the first failure in a
sequence thereby allowing other parts of the app to run freely while
it is in the background.
Task-number: ou1cimx1#951921
Change-Id: I1e27746957ff624b0d9a1cdc9b319d0a3477135d
Reviewed-by: Gareth Stockwell <gareth.stockwell@accenture.com>
Reviewed-by: Shane Kearns <ext-shane.2.kearns@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When QNetworkAccessManager::get is called from the slot connected to
QNetworkAccessManager::finished, this is equivalent to calling
QNetworkSession::open from the slot connected to QNetworkSession::error.
Close handles before emitting the error signal, instead of afterwards.
Otherwise the session start would be silently cancelled in the above
situation.
Task-number: ou1cimx1#960395
Change-Id: I48cb376722e0169721fd33614a0b9280c17bdd62
Reviewed-by: Murray Read <ext-murray.2.read@nokia.com>
Reviewed-by: Pasi Pentikäinen <ext-pasi.a.pentikainen@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This solves the bug with using custom paper sizes on Mac, if a QSize
was specified then with a Unit it would not respect this on Mac as it
would not convert it correctly. Also paperSize() would return the
wrong value.
There is already a test for this in tst_QPrinter::testCustomPageSizes()
Task-number: QTBUG-18723
Change-Id: I47c0cf2dd48ea3e0daa3b748f337ec38714aaf34
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Reviewed-by: Jiang Jiang <jiang.jiang@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
QFileSystemEngine::currentPath() returns QFileSystemEntry, but
this method expects QString return value.
-> Code does not compile for WinCE
Switched code to use "QFileSystemEngine::currentPath().filePath()"
correspondigly as Windows desktop does if filename does not contain
drive letter. This is ok, since WinCE does not support drive letters.
Task-number: QTBUG-22499
Change-Id: I211b56a5511e6c3e1b1c4009c8ee2a9c6d2f1746
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
Reviewed-by: Andreas Holzammer <andreas.holzammer@kdab.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Commit 5be6cf0a6e306ed3a51ed5ba89317b1317544eea introduced an implicit
cast from const char* to QString in QDBusServer's constructor, which
breaks the compilation of applications which use QtDBus when
QT_NO_CAST_FROM_ASCII is defined and clang is used.
Task-number: QTBUG-23398
Change-Id: Iaba66ab05ae6232f69a3981c2ca168f1ffd7bb05
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
| |
Change-Id: I019d468ef683a0d4040782df9d22b198b4cca86b
Reviewed-by: hjk <qthjk@ovi.com>
|
|
|
|
|
|
|
|
| |
Manually update strings that did not get fixed by previous
search-and-replace.
Change-Id: Ie9ff57f08ce0dcf91d28620fa2dc9eeb8e98481e
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
|
|
|
|
|
|
|
|
|
| |
Simple search and replace. This commit doesn't touch 3rd-party files,
nor translations (where the change is not so simple and will be handled
in a separate commit).
Change-Id: I4e48513b8078a44a8cd272326685b25338890148
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
|
|\
| |
| |
| |
| |
| |
| | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/qt-air-staging:
Updated WebKit to 7b7ded42f42a558bc5672c2523f9c5689bb975d0
|
| | |
|
|\ \
| |/
| |
| |
| |
| |
| | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/qt-air-staging:
Fix crash when creating a QScriptEngine in a native thread
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The change in http://trac.webkit.org/changeset/48412/ introduced a fix
to avoid leaking thread specific data by ensuring get() on ThreadSpecific
works even during the thread destruction phase. The fix worked by setting the
local data again.
However as we can see in the backtrace from QTBUG-22926, the local data should
not be set unconditionally, otherwise our destroy function will be called
recursively when the local data is still set.
Task-number: QTBUG-22926
Reviewed-by: Kent Hansen
Tested-and-Reviewed-by: Andy Shaw
|
|\ \
| |/
|/|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/qt-symbian-staging:
Fix http authentication to a different realm on the same server
Fix race in http connection channel
Don't fetch credentials from cache following a failed proxy authentication
Handle plain socket write errors in SSL
Fix for assertion failure
Fix faulty logic in http connection pipelining
Test case for QTBUG-22875
QThreads on Symbian are named to allow them to be opened externally
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This is a regression caused by the NTLMv2 authentication patch.
I have manually tested NTLMv2 authentication against MS IIS and reverting
these two lines does not break it.
Task-number: QT-5209
Change-Id: I64159cbe468e1a7f834f8726fd0c9d4ab4c54b38
Reviewed-by: Peter Hartmann <peter.hartmann@nokia.com>
(cherry-picked from 4954f71648aa7f74a4cb8b1dd26470b5da44459e)
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When authentication is cancelled, close the channel instead of the
underlying socket.
The previous behaviour could result in further requests being sent
on the closed socket, which caused errors in case of https over a proxy.
Change-Id: I3dbfc164de4fb29a426c06acaac8f29b9da1d705
Reviewed-by: Peter Hartmann <peter.hartmann@nokia.com>
(cherry picked from commit a7b99151f4445755c91d5227607d9ea2f785301f)
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Add variable to QAuthenticatorPrivate for tracking failure
Track authentication success/failure in http proxy socket engine
Track authentication success/failure in http connection channel
Task-number: QTBUG-22875
Change-Id: Id5d39e839428271ad687e9da12fbbdea9c478f4f
Reviewed-by: Peter Hartmann <peter.hartmann@nokia.com>
(cherry-picked from d24aad82896addce88f1ffb4040560e406acf083)
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When an ssl socket is closed during connecting, and it is using a proxy
then it is possible for the plain socket to be in pending close state
when transmit() is called.
As errors were not handled, this caused the socket (and https request)
to "hang".
It now propagates the error from plain socket.
Change-Id: I6fb86815a2a63e197cea582f4b153e487543477c
Reviewed-by: Peter Hartmann <peter.hartmann@nokia.com>
Reviewed-by: Richard J. Moore <rich@kde.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
(cherry picked from commit 2cc78885b0b7d08f965998d156945a077e56c1d8)
|
| |
| |
| |
| |
| |
| | |
Change-Id: I97b9ecc37e938a3050793fc746288243a1cb40b7
Reviewed-by: Peter Hartmann <peter.hartmann@nokia.com>
(cherry picked from commit 96cda705dcbeb79429055c1acca91f149d318820)
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The code which prevents pipelining of requests when authentication is
in use had || where && should have been used.
Also check for blank user with a password.
Change-Id: Ic278cedd370c9d81377f49a0af43aef415cb49ad
Reviewed-by: Peter Hartmann <peter.hartmann@nokia.com>
(cherry picked from commit 058fb94afff8a1a9989ab6d18dacc1fe43769fdb)
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The Qt 4.7 implementation of QThread on Symbian used libpthread. This
automatically generated a random name for a thread. The Qt 4.8
implmentation was leaving threads unnamed. This is a change in
behaviour, in that unnamed/anonymous threads cannot be opened outside
of the owning process. This was causing a bug in some client/server
situations.
The fix is to generate a name for the RThread underlying the QThread
as follows:
QThread object name + QThread object address + random number
Task-number: ou1cimx1#959586
Reviewed-by: Shane Kearns
|
|\ \
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/qt-qml-staging:
Adapt autotests to commit e2773a9
Declarative: Use qDebug() for debugger status updates
Declarative: Unify output of warning messages in debug server
|
| |\ \ |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Although "Qml debugging is enabled. Only use this in a safe environment!"
is a warning, using qWarning() for it let every app crash that is run with
QT_FATAL_WARNINGS. Fix that by using qDebug() for the updates that
happen during 'normal' operation.
Reviewed-by: Aurindam Jana <aurindam.jana@nokia.com>
Change-Id: Ie37a35ca27ec6e507f7de140484ca8cf96410be2
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Use the stream operator instead of toAscii().constData(), or qPrintable.
Also use QString::fromLatin1() instead of QString::fromAscii().
Task-number: QTBUG-22860
Change-Id: Iabe37778f0fd50ca450f0504939441c633d6354e
|
|\ \ \ \
| | |_|/
| |/| |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/qt-symbian-staging:
Symbian: Fix exporting and importing of some functions for autotests
Symbian: Fix SRCDIR usage in qlayout autotest.
Symbian: Fix qcssparser autotest
Symbian: Fix qscriptextensionplugin autotest
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Symbian builds seem more strict about properly declaring exports than
other builds, so added proper exporting and importing to some
function declarations.
Task-number: ou1cimx1#957184
Reviewed-by: mread
|
|\ \ \ \
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/qt-fire-staging:
Fix crash in positionInLigature
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Check boundary of pos before accessing attributes.
Task-number: QTBUG-23104
Reviewed-by: Eskil
|
|\ \ \ \ \
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/qt-earth-staging:
SSL: fix build with -openssl-linked for OpenSSL 0.9.8*
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
OpenSSL's SSL_ctrl() always took a "void *" argument as 4th parameter,
since at least version 0.9.7.
I have no idea why we had "const void *" in there.
Reviewed-by: Richard J. Moore <rich@kde.org>
Task-number: QTBUG-23132
|
|\ \ \ \ \ \
| | |_|/ / /
| |/| | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/qt-symbian-staging:
Fix missing part from commit dae052cb11c0018121f2c4028aed9db17769fd77
Fix m_longTapDetector causing crash.
Fix to incorrect ApplicationActivate event in QtOpenGL
Symbian - Fix QFile::map with non page aligned offsets
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Forgot to amend commit before pushing, sorry.
Task-number: QTBUG-23252
Reviewed-by: TRUSTME
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
GL shared widget calls null poiter m_longTapDetector.
These call shouldn't happen because GL shared widget
isn't part of UI stack anymore. This patch adds null
pointer checks
Task-number: QTBUG-23252
Reviewed-by: TRUSTME
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
QEvent::ApplicationActivate is sent incorrectly when application
goes to background if OpenGL graphics system is used. The problem
is that hidden global shared QGLWidget used by QtOpenGL for root
context is added to CONE stack. Qt destroys shared GL widget
when application goes to background and underlying CCoeControl
is removed from CONE stack which causes CONE to handle
stack changes. CONE tries to focus next control in stack which
causes incorrect focus events in Qt leading to
ApplicationActivate event. GL global share widget must not
be added to CONE stack because it's hidden utility widget
and don't belong to UI widget stack.
Task-number: QTBUG-23195
Task-number: ou1cimx1#946477
Reviewed-by: Murray Read
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Although it is not documented, RFileMap native API requires the offset
into the file to be a multiple of the page size.
Round down the offset requested by the user to the page size.
Add back the difference before returning the base address to the user.
e.g. if the offset passed to QFile::map was 5000:
The offset passed to RFileMap will be 4096
The address returned to the user will be the RFileMap::Base() + 904
The modified address is used to key the hash of file mappings, so that
QFile::unmap() can be passed the modified address without needing changes.
Reviewed-by: mread
Task-number: ou1cimx#953054
|
|\ \ \ \ \ \
| |/ / / / /
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/qt-symbian-staging:
Symbian: fix tst_QMenuBar::task256322_highlight() regression
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
This fixes Symbian-only regression caused by
ab0a2466d7e3998caad226197355b729f523764a.
Reviewed-by: Miikka Heikkinen
|
|\ \ \ \ \ \ |
|