summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Implement setDoubleStep for QInputDialogAndy Shaw2017-07-113-0/+112
| | | | | | | | | | | | setDoubleStep works in the same manner as for setIntStep in that it is just available for input dialogs getting a double. [ChangeLog][QtWidgets][QInputDialog] Added setDoubleStep to enable changing of the step amount for getDouble(). Task-number: QTBUG-17547 Change-Id: I5cabcfceb23324f8045f2b1e49017644418db01a Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
* macOS: Deliver screen change on window creation instead of on first exposeTor Arne Vestbø2017-07-111-10/+5
| | | | | | | | We know the resulting screen based on the geometry mappings we now do, so no need to wait until expose to deliver the screen change. Change-Id: Ibb84948ab091d8f74d31cdd2d300b381e2e1e7cc Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
* macOS: Rename NSWindow variable in QCocoaWindow::createNSWindowTor Arne Vestbø2017-07-111-11/+11
| | | | | | | To distinguish it from window(). Change-Id: I90eb5207c6d9f178f009ecf755e2be8d29149a6b Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
* macOS: Don't special-case QTestLib when reacting to occlusion eventsTor Arne Vestbø2017-07-111-12/+4
| | | | | | | | The tests (and Qt) should be able to handle occlusion of windows. Change-Id: I5e93e032f6a5282f19a20d0e230863d2a165f4e3 Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org> Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
* tst_QSslSocket::protocolServerSide - fixes for OpenSSL 1.1Timur Pocheptsov2017-07-111-3/+17
| | | | | | | | | | | | | | | | Several tests are not valid for 1.1 anymore: 1. SSL2 was removed, but there is no OPENSSL_NO_SSL2 and the 'protocolServerSide' test is trying to use QSsl::SSLv2 and thus is failing. 2. We now use the generic TLS_server/client_method instead of version specific methods we have in pre-1.1 back-end. So, for example, a client socket with QSsl::TLS_V1_0 in its SSL configuration will be able to negotiate TLS 1.2 if our server socket wants it, while with TLSv1_client_method (OpenSSL < 1.1) our test was expecting SSL handshake to fail. Change-Id: I18efd5921c79b189e4d9529be09299a361a8a81d Reviewed-by: Edward Welbourne <edward.welbourne@qt.io> Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
* QSslCertificatePrivate: make use of QByteArray::toHex(separator)Andre Hartmann2017-07-101-11/+6
| | | | | | | | | | Qt 5.9 introduced a QByteArray::toHex() overload that inserts a separator char after each byte. The function colonSeparatedHex() could not be replaced completely, as it skips leading zeros also. Change-Id: Ic1d5d4771a0a5171f0e7e9813b83eb4c1f59f085 Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
* QMacStyle: Simplify logic for SC_GroupBoxLabel rectangleGabriel de Dietrich2017-07-102-29/+18
| | | | | | | | | | | This removes the single occurrence of HIThemeGetTextDimensions. Since we have access to the actual platform font through QPlatformTheme::font(), it's possible to use QFontMetricsF instead. We also cache small and mini system fonts to be reused in other places. Change-Id: I799c61eef0a2a6880a1c51aa7d3105c7e5b0ed4f Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
* QMacStyle: Simplify rendering SC_GroupBoxLabelGabriel de Dietrich2017-07-101-27/+12
| | | | | | | Removes the last occurrence of HITHemeDrawTextBox. Change-Id: I13804a16672c692a4405d8b2f9d5fb6a132a6b7b Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
* QMacStyle: Simplify CE_PushButtonLabel, CE_TabBarTabLabelGabriel de Dietrich2017-07-102-140/+57
| | | | | | | | | | We don't need the calls to HIThemeDrawTextBox after all. These render simple text nowadays, so we can do this with Qt APIs. Also fixes TabBarPalette in QCocoaTheme. Change-Id: Ib9436be83f03bd62fca40feb110a86602794c5b7 Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
* macOS QComboBox: Don't show scrollers if they aren't neededDaniel Teske2017-07-102-0/+14
| | | | | | | Task-number: QTBUG-13925 Task-number: QTBUG-18788 Change-Id: Id155ea26824e1cb470cc6174704436d99fb7ac87 Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@qt.io>
* macOS: Use same naming for QPA event forwarding functionsTor Arne Vestbø2017-07-092-22/+26
| | | | | | | Expose events and others will follow. Change-Id: I8e11a133381a678517b54ad1872fe302515d4104 Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
* macOS: Simplify QCocoaWindow::handleGeometryChangeMorten Johan Sørvig2017-07-093-11/+20
| | | | | | | | | | | | | | There are really only two cases here, where the difference is the coordinate system of the window position. 1) Child QWindow and embedded QWindow: The position is relative to parent view/window origin. 2) Top-level QWindow: The position is relative to screen origin. Change-Id: I867133a5adbbf3a690f574aec06b70c2bc64ad95 Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io> Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
* Move endian-specific integer types to public qtendian headerAllan Sandfeld Jensen2017-07-093-81/+490
| | | | | | | | | The classes have recently been cleaned up to be ready for being made public, and this change adds the final documentation. The classes are already used in qjson, qtdeclarative and some image format decoders. Change-Id: Ib6df89f1119162c5bad7a08e9b6f843a2db4040f Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
* macOS: Move geometry reporting from QNSView to QCocoaWindowTor Arne Vestbø2017-07-084-55/+61
| | | | | | | | | | | There's no longer any reason to call out from QCocoaWindow to QNView for this, as the geometry events from AppKit are delivered directly to the QCocoaWindow. Most of the data used in the implementation are coming from QCocoaWindow anyways, and this enables geometry events for foreign windows in the future. Change-Id: Idd724d078e9981304dcbe6742b9ddc71640a2350 Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
* Add support for parameters in QShaderGeneratorKevin Ottens2017-07-082-8/+26
| | | | | Change-Id: Iff8fec51ddbf9b018cb5ec3aff357125134013eb Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
* Add support for node parameters in QShaderGraphLoaderKevin Ottens2017-07-082-32/+48
| | | | | | | | Now the parameters from the prototypes are used as default values and the graph file can overload them. Change-Id: I9a8a73963c7e578bd0a34e96f76f147a12224f4f Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
* Add support for node parameters in QShaderNodesLoaderKevin Ottens2017-07-082-13/+34
| | | | | | | | This will allow to make the prototypes more extensible when referred to from a graph file while providing some sane defaults. Change-Id: I1ae10182427d8a7d29c51a64e19e99139494ea92 Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
* Add parameters to QShaderNodeKevin Ottens2017-07-083-2/+73
| | | | | | | | | This will allow to create nodes having an entry point for tuning, still need to be exploited by the loaders and the shader generator. Coming in further commits. Change-Id: I8384b0f528c9919e9f8d35102adde2b307f08b80 Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
* Complete the QShaderGraphLoader tests to introspect nodesKevin Ottens2017-07-081-0/+45
| | | | | | | | We forgot to check that the node content was correct and complete when loading the graph. Change-Id: Id4ee6aaba6ca268b0785e7fa3fb51a19817e7c9c Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
* Allow QImage with more than 2GByte of image dataAllan Sandfeld Jensen2017-07-0824-66/+115
| | | | | | | | | | | | Changes internal data-size and pointer calculations to qssize_t. Adds new sizeInBytes() accessor to read byte size, and marks the old one deprecated. Task-number: QTBUG-50912 Change-Id: Idf0c2010542b0ec1c9abef8afd02d6db07f43e6d Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
* Move the readlink(2) wrapper to qcore_unix.cppThiago Macieira2017-07-087-34/+50
| | | | | | | This deduplicates the code between QFileSystemEngine and QLockFile. Change-Id: I1eba2b016de74620bfc8fffd14cd005d5fd9beaa Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@qt.io>
* QLockFile: Move the check for Linux /proc to a central placeThiago Macieira2017-07-082-2/+17
| | | | | | | | | | | | | | | | | I'll need it soon in QTemporaryFile. This also reduces the overhead, since QFile::exists -> QFileInfo::exists needs to check if there's a file engine that handles "/proc/version" (there isn't), convert the UTF-16 filename to UTF-8 and QFileSystemEngine::fillMetaData will issue a more expensive stat(2) syscall. access(2) is cheaper. This commit also introduces QT_LINUX_ALWAYS_HAVE_PROCFS that isn't defined anywhere, but could be passed during configure with -D if a build wants it. Change-Id: I1eba2b016de74620bfc8fffd14ccfd1593d59ade Reviewed-by: Lars Knoll <lars.knoll@qt.io>
* Stop using QVector in QFileInfoPrivateThiago Macieira2017-07-072-15/+14
| | | | | | | | | | | | This was added in 32629676b977d98e853fc6101a63c0d888ff5598 for Qt 5.1 because QDateTime always allocated memory, so QVector was much faster for default-created objects. Since Qt 5.7, QDateTime no longer allocates memory in the default constructor, and in Qt 5.8 on 64-bit systems, it won't allocate memory at all for most reasonable dates, so construction times are acceptable now. Change-Id: If0ad4d988da143b3b1b2fffd1480e83121cddc8c Reviewed-by: Lars Knoll <lars.knoll@qt.io>
* QAbstractFileEngine: make setFileTime not a pure virtualThiago Macieira2017-07-079-35/+9
| | | | | | | | Most engines just had to override the virtual to return false. So do it in the base class. Change-Id: I8d96dea9955d4c749b99fffd14cd738a48aa818c Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@qt.io>
* macOS: Add LLDB debug script when building with separate debug info (dSYM)Tor Arne Vestbø2017-07-073-0/+112
| | | | | | | | | | | | | | | | | | | | | | | | | | The script will look for the most recent Qt Creator version on the system, and pick up the LLDB summary providers from there, allowing pretty-printing of Qt types inside LLDB/Xcode. LLDB will detect the file when loading the dSYM, and inform the user that the file can be loaded to enable the formatters. The script can be loaded automatically by adding the following setting in ~/.lldbinit: settings set target.load-script-from-symbol-file true Which comes as a slight security risk, as other libraries might have scripts of their own. The alternative is to load the script directly from ~/.lldbinit: command script import "<path to debug script in dSYM>" With an optional target.load-script-from-symbol-file set to false, to silence the warning when loading the dSYM bundle. Change-Id: I01ba51dab725a8d0a58f1ad1749742443b639cc5 Reviewed-by: hjk <hjk@qt.io> Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@qt.io> Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
* Mark an unlikely test as such and condition in positive orderEdward Welbourne2017-07-071-7/+10
| | | | | | | | Also wrap two blocks in braces, since the formerly-else block spreads across many lines; and split those lines differently to limit length. Change-Id: Ib89329b11aad6599926f0338d6546f4141d2c002 Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@qt.io>
* Ensure all platforms send geometry events if request was not fulfilledTor Arne Vestbø2017-07-077-55/+57
| | | | | | | | | | | | | | | | | | The logic of deciding whether or not to send resize and move events has been centralized in QGuiApplication. This ensures that if a window with geometry 100,100+200x200 is moved and resized to e.g. 0,0+100x100, but the window manager denies the request (because the window would e.g. overlap with system UI), and issues a geometry update with the original geometry, 100,100+200x200, we will still treat that as warrant of a move/resize event to the application, so the application knows that its position and size is as before. [ChangeLog][Qt Gui][QPA] QWindowSystemInterfacePrivate::handleGeometryChange no longer takes the old geometry as an argument. Task-number: QTBUG-57608 Change-Id: I1d471cc7a257fef958bdb1e56184fa95489403a3 Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
* tst_QDateTime::toTimeSpec(): verify round-tripping works cleanlyEdward Welbourne2017-07-071-0/+4
| | | | | Change-Id: I811fa4249dae923e5e4e434910d8bf96c15bff25 Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
* tst_QFileInfo: generalize old-file testEdward Welbourne2017-07-071-49/+23
| | | | | | | | | | | | | Now that QFileDevice::setFileTime() provides a portable way to prepare our test file, we can verify QFileInfo does handle the distant past, notably including negative time_t values. The old MS-specific code used a time back in 1601, which we can't hope to support cross-platform, so use one in 1901 that's a little inside the range of 32-bit time_t. Task-number: QTBUG-47985 Change-Id: I2de3e79d8c7864221f92395813b63f373e4d8a3b Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
* Exploit QVector3D::length() instead of duplicating its workEdward Welbourne2017-07-071-1/+1
| | | | | | | | | Calling the class-method dotProduct(x, x) in order to then take its square root is a clumsy way to obfuscate x.length() - and lacks its efforts (clumsy though they are) to limit rounding issues. Change-Id: I1dc1f38764651bc70c0620e286cb5625f505ddbf Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
* Q(Date|Time)+: in QDebug's <<, handle invalid explicitlyEdward Welbourne2017-07-072-18/+32
| | | | | | | | | | The output didn't previously make clear that the datum was invalid. It's now explicitly invalid. At the same time, use QDebug's space() and nospace() methods to make spacing choices explicit. Revised a QDate test to match. Change-Id: I4699f5897530b4caa31c22fdb07de149832b30f4 Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
* Clarify tst_QDateTime::fromStringToStringLocale() using a macroEdward Welbourne2017-07-071-7/+9
| | | | | | | | | | | Separate the part varying between tests from the common form of all the tests, so the reader can see the common pattern and know for sure that there's not a typo or copy-and-paste glitch. Change-Id: I3145a26ab42c104eb27756d906ac87f937024bad Reviewed-by: Mitch Curtis <mitch.curtis@qt.io> Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org> Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
* Call sinh() rather than implementing it using exp()Edward Welbourne2017-07-071-1/+1
| | | | | | Change-Id: I8dda2f94524b19bcdec640e7541b04d46a6c3941 Reviewed-by: Jesus Fernandez <Jesus.Fernandez@qt.io> Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
* Fix an egregious misnaming, s/deg([XY])/rad$1/gEdward Welbourne2017-07-071-4/+4
| | | | | | | | The angles in question were in radians, not degrees; we get them from std::atan(). Change-Id: Ifd60efbd975997ffca02e45ae884fb75e59806b4 Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
* tst_qstyle: Don't fail on empty scroller subcontrols on macOSGabriel de Dietrich2017-07-071-5/+11
| | | | | | | | | | | macOS hasn't had SC_ScrollBarAddLine and SC_ScrollBarAddLine since 10.7, when transient scrollbars first appeared. We also make the warning message a bit more informative. Change-Id: Idef4684162456d9bc274eea77908a6afe24fa0f5 Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io> Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
* QMacStyle: De-Carbonize QScrollBarGabriel de Dietrich2017-07-071-472/+385
| | | | | | | | | | | | | | | | | | | | Addresses drawing, hit testing, subcontrol rects and one metric. Also fixes the knob width on hovered transient scrollbars. Since Cocoa won't help for this (and never will), we do it manually. For non-transient scrollbars, no more HITheme. That's why we're doing this after all. It also comes with its own small hack; see how we darken the knob when hovered. We had to de-intertwine the logic with QSlider in drawComplexControl(), which now gets its own full case CC_Slider statements. QSlider will be addressed next. Task-number: QTBUG-49585 Change-Id: Iced58d52fff0c11866bdf6eb562dbab36c8f3ef2 Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@qt.io> Reviewed-by: Jake Petroules <jake.petroules@qt.io>
* Relieve platform plugins of having to persist geometry on WM callbacksTor Arne Vestbø2017-07-0711-32/+21
| | | | | | | | | We can offload this to QGuiApplication, just like the geometry of the QWindow is set. This ensures that all platforms behave the same, and that the documentation of QPlatformWindow::setGeometry is adhered. Change-Id: I19dbc32cb4fb146d716ec289c28030a547d3afaa Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
* Make calling QCoreApplication::translate() thread-safeAram So2017-07-063-20/+73
| | | | | | | | | | | | | | Fixed crash on QCoreApplication::translate() call from qqmlThread while QCoreApplication::{install,remove}Translator() is called from the GUI thread. [ChangeLog][QtCore][QCoreApplication] Calling QCoreApplication::translate() is now thread-safe. Task-number: QTBUG-57095 Change-Id: Ie5340a42040a829f311c01332e05d4bbaf60462c Reviewed-by: Olivier Goffart (Woboq GmbH) <ogoffart@woboq.com> Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
* macOS: Deliver native events to windows, as well as to event filtersTor Arne Vestbø2017-07-062-0/+8
| | | | | | | | | | | | | | | | | | | | | | Qt has two APIs to intercept native events today: - QAbstractEventDispatcher::installNativeEventFilter() - Q{Window|Widget}::nativeEvent() On macOS we only implemented one of them, the native event filter code path, by calling filterNativeEvent from the Cocoa event dispatcher. We now also propagate the native event to the corresponding QWindow, and QWidget if applicable. It would be nice if there was only one Qt API for this, or at least only one codepath for platform plugins to care about, but since the event filter might catch more event types than gets delivered to the window, we probably need both code paths going forward. Task-number: QTBUG-40116 Change-Id: I0796bd62a2b7c08b2eaaf6f15db8088e9703af02 Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
* macOS: Don't deliver events via NSWindow for deleted QCococaWindowsTor Arne Vestbø2017-07-061-0/+8
| | | | | | Change-Id: Icb3794f37c929019de1e997e15f7d975492224c2 Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io> Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
* Merge remote-tracking branch 'origin/5.9' into devLiang Qi2017-07-06146-273/+632
|\ | | | | | | | | | | | | Conflicts: .qmake.conf Change-Id: I43531e087bb810889d5c1fbfcdffb29b78804839
| * Extend tst_macgui::nonModalOrder blacklist to include 10.12Simon Hausmann2017-07-041-1/+2
| | | | | | | | | | | | Change-Id: I9cfe9e8d79a3223c9cdbb039cc3e243b173f5e9c Task-number: QTQAINFRA-1333 Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
| * QSslSocket - fix two memory management issues (Secure Transport)Timur Pocheptsov2017-07-041-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | I noticed we never release 'items' imported by SecPKCS12Import. But looking at the actual code (SecImportExport.c), it appears we own these 'items' and must release them. And this leads to a crash (on over-release) which reveals another bug: a value from a dictionary obtained with 'Get' method should follow the 'get rule' - we do not own it and QCFType RAII object is not needed. Change-Id: I219015fadedb256c401e50cf7e955f3d7e0a6c5f Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io> Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
| * Make qt_check_pointer more OOM-safeThiago Macieira2017-07-032-6/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | First, it can never return, so we can mark it Q_NORETURN and add an std::termianate at the end. Though if it did, we'd end up in a null- pointer dereference crash in the caller. Second, add Q_DECL_NOTHROW to it. It can't throw, but it terminates execution. This also prevents both puts and fprintf from escaping via pthread asynchronous cancellation on Linux/glibc. Third, don't use QMessageLogger, since that allocates memory and actually uses QString. If we really are in an OOM situation, then QString's failed allocation would recurse back into qt_check_pointer. We'd compound the OOM situation with a stack overflow... Change-Id: Ia53158e207a94bf49489fffd14c81c47971f4e82 Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io> Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
| * Use QMessageLogger context properly in qglobal.cpp redirectsThiago Macieira2017-07-031-3/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | qt_assert and qt_check_pointer get the function name and line number from the caller (the functions are called from the Q_ASSERT and Q_CHECK_PTR macros, respectively), so we don't need to capture the context from those two functions. Instead, pass the context to QMessageLogger for proper logging. I've left the file name and line number in the assertions, for users who did not add them to their message log pattern, but I've removed from the almost never used qt_check_pointer function. Note: how useful is it that we allocate memory in response to failing to allocate memory? Change-Id: Ia53158e207a94bf49489fffd14c81b359c5b6537 Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io> Reviewed-by: Kai Koehne <kai.koehne@qt.io>
| * Fix build with MSVC 2015 Update 2 if constexpr is enabledThiago Macieira2017-07-032-2/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This compiler seems to require explicit initialization of all member variables in a constexpr constructor, even if they have an implicit default constructor of their own. We probably fixed the rest of Qt a couple of years ago, but not these two places because they were arrays and those require the C++11 syntax for uniform initialization. All compilers that support constexpr do support uniform initialization. MSVC 2015 fixed our issues with it on the same update. Change-Id: Ibc1eb23e3ae093f5c6928ded3a041be35eb9baae Reviewed-by: Marc Mutz <marc.mutz@kdab.com> Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
| * Disable the -Wstringop-overflow warning from GCC 7Thiago Macieira2017-07-031-0/+2
| | | | | | | | | | | | | | | | | | | | | | It prints the warning even if we surround the affected code with QT_WARNING_DISABLE_GCC("-Wstringop-overflow") (see e4eaa629439fe1ba1e), so we have no alternative other than to disable the warning completely. Change-Id: Ia3e896da908f42939148fffd14c488c4006040e6 Reviewed-by: Ville Voutilainen <ville.voutilainen@qt.io> Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com> Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
| * Fix parsing of 0E+1 and 0E-1 (capital 'E')Thiago Macieira2017-07-032-2/+58
| | | | | | | | | | | | | | | | | | | | Since the result is an actual zero, this section of code looking for underflows kicks in. But we forgot to take the capital letter into account when parsing the number. Task-number: QTBUG-61350 Change-Id: Ia53158e207a94bf49489fffd14c6abbd21f0bac0 Reviewed-by: Olivier Goffart (Woboq GmbH) <ogoffart@woboq.com>
| * Windows QPA: Correctly check for fixed sized windows in WM_DPICHANGEDFriedemann Kleint2017-07-031-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | Use Qt's flags instead of WS_DLGFRAME which matches WS_CAPTION as well (WS_BORDER | WS_DLGFRAME). Amends 886ce572d628e7cd98cc39edcc930ffae951e95e. Task-number: QTBUG-58959 Change-Id: Ifdc106667d67cc6f5d3611806aae1035742fb882 Reviewed-by: Olivier Goffart (Woboq GmbH) <ogoffart@woboq.com> Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
| * Android: Replace QMutex with QBasicMutex in androidjnimainChristian Strømme2017-07-031-1/+1
| | | | | | | | | | | | | | QBasicMutex is sufficient and provides reduced initialization cost. Change-Id: I79ae61daaed4f5edd9b21d913f78e78e7ba14c94 Reviewed-by: BogDan Vatra <bogdan@kdab.com>
generated by cgit v1.2.3 (git 2.25.1) at 2024-05-10 05:51:13 +0000