summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* Replace a few hardcoded paths with defaults from paths.hThiago Macieira2017-12-244-14/+34
| | | | | | | | This removes at least one special-case we had to have, in Android's lack of /etc/mnttab. Bionic's _PATH_MOUNTED is already /proc/mounts. Change-Id: I9407dcf22de6407c83b5fffd14fedc638586d0f9 Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
* Doc: attempt to fix some qdoc errors for QRandomGeneratorThiago Macieira2017-12-241-5/+7
| | | | | | | | | | | | | This fixes only those that look fixable. There are a number of errors in the online documentation that look like qdoc tool bugs, like missing functions (operator==) and documentation text that does not exist in the source code. This fixes: - QRandomGenerator(System) constructor showing up - Links for C++ reference documentation not showing up as links Change-Id: I9e2892cb6c374e93bcb7fffd14fe21db5a6969d9 Reviewed-by: Martin Smith <martin.smith@qt.io>
* QMacStyle: Minimize size of scrollbar transparency layerGabriel de Dietrich2017-12-211-1/+1
| | | | | | | | | Instead of allocating the full backing store size for the transparency layer, we make sure we only allocate exactly as much as needed by the scrollbar. Change-Id: I55c3172fe3dd2a1f3fd46828463497f9f35cb1ae Reviewed-by: Jake Petroules <jake.petroules@qt.io>
* QMacStyle: Use light color on expanding scrollbarGabriel de Dietrich2017-12-211-3/+4
| | | | | Change-Id: I4e9c870c8acaaa690f530f847c9927d61a508a94 Reviewed-by: Jake Petroules <jake.petroules@qt.io>
* QMacStyle: Bring back submenu indicatorGabriel de Dietrich2017-12-211-19/+34
| | | | | | | | | | | | | | | | This one fell off the truck while we were removing HITheme calls. We add it back by simply rendering a BLACK RIGHT-POINTING TRIANGLE character. We also fix smaller issues, such as not displaying any shortcut related to a submenu action — this is simply not a thing. The spacing between the menu item's text and the submenu indicator has also been slightly improved. Change-Id: I6c768a5506a5eb9528b0dd76acd52b561266d67b Task-number: QTBUG-64405 Reviewed-by: Jake Petroules <jake.petroules@qt.io>
* Fix drawing color fonts with shear and perspective transformsAllan Sandfeld Jensen2017-12-201-2/+20
| | | | | | | | | Fixes the fallback painting used with complex transforms to be able to handle color bitmap fonts which can't be converted to a path. Change-Id: Id2851607f673b8fc1aea63f92043d0cdebc0fb9d Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com> Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@qt.io>
* Windows QPA: Check validity of device when handling tablet leaveFriedemann Kleint2017-12-191-0/+2
| | | | | | | | | | Huion GT-191 has been observed to send spurious leave events. Amends 127483b5e30de6c1905ea3280dd45a0b7d6a3813. Task-number: QTBUG-65120 Change-Id: I5dfa003a71be137a7b40cc9c27d7cf2cada6922d Reviewed-by: Shawn Rutledge <shawn.rutledge@qt.io> Reviewed-by: Oliver Wolff <oliver.wolff@qt.io>
* bearermanagment: fix build for --no-feature-bearermanagementLorn Potter2017-12-191-2/+1
| | | | | Change-Id: I9e3f44b924e09a8d1a5095dc55d585b3cab33c15 Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
* Fix scan reply dbus signature in connman bearer backendLorn Potter2017-12-191-1/+1
| | | | | | | Task-number: QTBUG-57844 Change-Id: I1f3035f32d213ec6da95650a946c17c64becf549 Reviewed-by: Sami Nurmenniemi <sami.nurmenniemi@qt.io> Reviewed-by: Lorn Potter <lorn.potter@gmail.com>
* Mask potentially undefined alpha in blend_transformed_argbAllan Sandfeld Jensen2017-12-161-2/+3
| | | | | | | | Makes sure the ARGB32PM that is painted on always have a defined alpha. Task-number: QTBUG-55645 Change-Id: Ifcf5fcc2127d255518eca4763845a197da6c7914 Reviewed-by: Olivier Goffart (Woboq GmbH) <ogoffart@woboq.com>
* macOS: Make retina OpenGL work on VMwareMorten Johan Sørvig2017-12-153-1/+20
| | | | | | | | | | | | | | | | | | | | | | Qt requests full resolution OpenGL surfaces by default. However, when running as a VMware guest it looks like the OS silently creates low-resolution surfaces. This is not possible to detect using the standard NSWindow APIs for converting to backing coordinates or for reading the backing scale factor. The result of this is that Qt will incorrectly display one quarter of the window content only. Fall back to detecting if the OpenGL renderer is the Apple software renderer, which it will be on VMware. Cancel the high-resolution surface request if this is the case. This needs to be done while we have a valid OpenGL context. Task-number: QTBUG-62357 Change-Id: I33bf12b3bb0408249e6d66e0a8ca86b044bea781 Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
* Paint text decoration where the text isAllan Sandfeld Jensen2017-12-131-2/+2
| | | | | | | | | | | For multi text-item texts we should render the text decoration at the same position as the text-item part is, and not always from the beginning of the line. Task-number: QTBUG-60422 Change-Id: I9aa58fc164122ad1fae9716b8b18bdfbbbd778a9 Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com> Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@qt.io>
* QtGui/Windows: Fix wait cursors remaining activeFriedemann Kleint2017-12-121-0/+2
| | | | | | | | | | | In QWindowPrivate::applyCursor(), do not apply override cursors when the platform supports QPlatformCursor::OverrideCursor. Complements b05d1c2ebfebf0f427a92668c0a7b177d0952012. Task-number: QTBUG-65001 Change-Id: Ie84cc30ad99b22e037aae829a2ce847ec4bf900f Reviewed-by: Oliver Wolff <oliver.wolff@qt.io>
* Windows QPA: Fix handling of override cursorsFriedemann Kleint2017-12-122-2/+6
| | | | | | | | | | | | | | | | | | | Override cursors were not restored when nested or in a dual monitor setups. The default cursor stored in QWindowsCursor::m_overriddenCursor was clobbered by subsequent calls to QWindowsCursor::setOverrideCursor(). This caused for example the wait cursor to remain active when switching to Help Mode in Qt Creator. Add a check preventing that. Make the variable static so that it is shared between the cursors of multiple screens. Amends b05d1c2ebfebf0f427a92668c0a7b177d0952012. Task-number: QTBUG-65001 Change-Id: Iead5804d317f73dedd78d22c1c85c62b5349ab83 Reviewed-by: Oliver Wolff <oliver.wolff@qt.io>
* Fix Android Service thread getting stuck in service startupHeikki Haveri2017-12-121-1/+1
| | | | | | | | | | | | | | | | This change allows services using QAndroidService to function properly. A service using a plain QCoreApplication is still affected. The original code uses postfix decrement, causing an off-by-one error in reference counting, which, in turn, fails to properly release a semaphore. Fix by using deref() instead, which is also more readable, and consistent with the use of ref() elsewhere in the code. Task-number: QTBUG-64728 Change-Id: I99b59307638d619506c594f86de6d7d202d755f2 Reviewed-by: Marc Mutz <marc.mutz@kdab.com> Reviewed-by: BogDan Vatra <bogdan@kdab.com>
* QStandardPaths: fall back to a default if PATH isn't setThiago Macieira2017-12-121-0/+29
| | | | | | | | | | | | | | | | | | | | | | | | | | Change 28666d167aa8e602c0bea25ebc4d51b55005db13 simplified QProcess::start{Process,Detached} by using findExecutable() instead of using execvp, but this introduced an unintended side effect when the PATH environment variable isn't set. It turns out that most libc have a default fallback in that situation, which we didn't apply. This commit applies the default directly to findExecutable(), which seems sensible. [ChangeLog][QtCore][QProcess] Fixed a regression that made QProcess be unable to find executables when the PATH environment variable on some Unix systems wasn't set. This behavior should not be relied upon since many systems do not have sensible fallback values for PATH. [ChangeLog][QtCore][QStandardPaths] findExecutable() will now apply the default value for the PATH environment variable (as returned by the POSIX confstr(3) function or found in <paths.h>) if the variable isn't set in the environment. Task-number: QTBUG-65076 Change-Id: I9407dcf22de6407c83b5fffd14feda6c1f509210 Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@qt.io> Reviewed-by: David Faure <david.faure@kdab.com>
* Windows QPA: Emulate aboutToHide(), aboutToShow() signals of QPlatformMenuFriedemann Kleint2017-12-111-1/+8
| | | | | | | | | | | Wrap the call to TrackPopupMenu accordingly. Complements 7849aa6e96aa923fca5523afc8cf88edcc0bcf90. Task-number: QTBUG-64628 Change-Id: Ia370e566266e96ab690ce5ed41d06dea7cafd4e4 Reviewed-by: J-P Nurmi <jpnurmi@qt.io> Reviewed-by: Oliver Wolff <oliver.wolff@qt.io>
* Merge remote-tracking branch 'origin/5.10.0' into 5.10Liang Qi2017-12-0849-159/+285
|\ | | | | | | Change-Id: I6dcf85067ec226136c207ea69ca9d66736c84db5
| * Android: use a canonical data dir for Qt environment variablesv5.10.0-rc3v5.10.0J-P Nurmi2017-11-301-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ApplicationInfo.dataDir holds "/data/user/0" (Blackberry PRIV running Android 6.0.1), which is in fact a soft link to "/data/data". This directory is used as a prefix for various Qt environment variables, including QML2_IMPORT_PATH, which in turn is used for resolving QML type URIs, looking up Qt Quick Controls 2 styles, and so on. The QML engine is not happy with "/data/data" and "/data/user/0" being wildly mixed for QML types in the same module. Use the canonical path instead to avoid such conflicts. Change-Id: I1fd45736728ee662942d7ef48c3fbc553981c59b Task-number: QTBUG-64868 Reviewed-by: Lars Knoll <lars.knoll@qt.io>
| * Improve readability of code that uses the Qt signed size typev5.10.0-rc2Simon Hausmann2017-11-2823-107/+107
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | During the container BoF session at the Qt Contributor Summit 2017 the name of the signed size type became a subject of discussion in the context of readability of code using this type and the intention of using it for all length, size and count properties throughout the entire framework in future versions of Qt. This change proposes qsizetype as new name for qssize_t to emphasize the readability of code over POSIX compatibility, the former being potentially more relevant than the latter to the majority of users of Qt. Change-Id: Idb99cb4a8782703c054fa463a9e5af23a918e7f3 Reviewed-by: Samuel Gaist <samuel.gaist@edeltech.ch> Reviewed-by: David Faure <david.faure@kdab.com>
| * Fix namespaced linux buildsAllan Sandfeld Jensen2017-11-282-1/+8
| | | | | | | | | | | | | | | | | | Using "struct statx" as argument type is forward declaration which then will expect the namespace where it was first encountered. Change-Id: I2d4ba930bd5b4e264228f2549bd6ef75e5cf3a67 Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@qt.io> Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
| * Fix naming of new property QSortFilterProxyModel::recursiveFilteringDavid Faure2017-11-282-7/+7
| | | | | | | | | | | | | | | | | | | | | | Add "Enabled" to match Qt API naming rules. + fix \since tag. Change-Id: Iaf312648f7385cd7e8d3d101b561fbd4e955df25 Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io> Reviewed-by: Edward Welbourne <edward.welbourne@qt.io> Reviewed-by: Thiago Macieira <thiago.macieira@intel.com> Reviewed-by: Lars Knoll <lars.knoll@qt.io>
| * Fix regression in painted emoji offsetv5.10.0-rc1Allan Sandfeld Jensen2017-11-251-1/+4
| | | | | | | | | | | | | | | | | | Factor out translation from the matrix applied on bitmap glyphs, as that gets applied as position when painted. Task-number: QTBUG-64313 Change-Id: Iab8d995c00ee02eda0896242903312d837b6d790 Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@qt.io>
| * Fix the build if st_atimensec is definedThiago Macieira2017-11-231-2/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | It's a trick used by some C libraries to keep compatibility with BSD extensions while supporting POSIX.1-2008 API. st_atimensec is defined as st_atim.tv_nsec, so the code would expand to invalid C++ In substitution of 'template<class T> typename std::enable_if<((&T::st_atim.tv_nsec), true), long long int>::type{anonymous}::GetFileTimes::atime(const T&, int) [with T = stat]': error: invalid use of non-static data member 'stat::st_atim' Change-Id: I38341f8155354cc4a776fffd14e20f4fc0f6d5bb Reviewed-by: BogDan Vatra <bogdan@kdab.com> Reviewed-by: Lars Knoll <lars.knoll@qt.io>
| * QTemporaryFile: fix issues with removing a file twiceThiago Macieira2017-11-234-5/+25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The assertion in isUnnamedFile() we had was incorrect after the file was removed, since we cleared the name and possibly reset back to the template. Since ~QTemporaryFile() calls remove(), this was easy to trigger if you attempted to remove the temp file and leave QTemporaryFile like that. Take this opportunity to add to the docs of setAutoRemove() explaining the possibility of unnamed files. #7 0x00007f69bcc2b50e in qt_assert ( assertion=assertion@entry=0x7f69bcf194a0 "unnamedFile == d_func()->fileEntry.isEmpty()", file=file@entry=0x7f69bcf19458 "io/qtemporaryfile.cpp", line=line@entry=514) at global/qglobal.cpp:3123 #8 0x00007f69bcd672cf in QTemporaryFileEngine::isUnnamedFile (this=this@entry=0x55cd60644df0) at io/qtemporaryfile.cpp:514 #9 0x00007f69bcd683f7 in QTemporaryFileEngine::remove (this=0x55cd60644df0) at io/qtemporaryfile.cpp:396 #10 0x00007f69bcd48654 in QFile::remove (this=this@entry=0x7fffb393f7e0) at io/qfile.cpp:513 #11 0x00007f69bcd6653b in QTemporaryFile::~QTemporaryFile (this=0x7fffb393f7e0, __in_chrg=<optimized out>) at io/qtemporaryfile.cpp:719 Change-Id: I57a1bd6e0c194530b732fffd14f4ed28ca8185b2 Reviewed-by: Andreas Hartmetz <ahartmetz@gmail.com> Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
| * Doc: Update the list of highlighted examplesTopi Reinio2017-11-226-2/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Update the list of highlighted examples for modules in qtbase, based on which examples have been updated to use C++11 features, the new signal/slot connection syntax, and documentation improvements. Not all the modules have highlighted examples yet as some of the work is still ongoing. Task-number: QTBUG-60641 Change-Id: If28d59c10ca1a30e5db408970f20159434ac94f8 Reviewed-by: Nico Vertriest <nico.vertriest@qt.io> Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io> Reviewed-by: Venugopal Shivashankar <Venugopal.Shivashankar@qt.io>
| * Merge remote-tracking branch 'origin/5.9.3' into 5.10.0Liang Qi2017-11-212-151/+158
| |\ | | | | | | | | | Change-Id: I3224e0b877ec24b587e299083feb793518b0568c
| * | QFileSystemEngine/Linux: fix when qt_lstatx() succeedsThiago Macieira2017-11-201-15/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When qt_lstatx() succeeds and the target is not a link, we'd erroneously mark the file as non-existent during the pass to check qt_statx(). All flags besides the file's modes were cleared. This is unit-tested, but only happens on Linux kernels 4.12 or later. It didn't happen to me because I already had this fix applied as part of a later change relating to QSystemResult. Task-number: QTBUG-64514 Change-Id: I938b024e38bf4aac9154fffd14f893506a1ef55b Reviewed-by: Lars Knoll <lars.knoll@qt.io>
| * | QRandomGenerator: Enforce the use of 32-bit integers in the engineThiago Macieira2017-11-201-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | std::mt19937 is defined as operating on uint_fast32_t, which is usually just a 32-bit integer. That's not the case on 64-bit Linux, where it is actually 64-bit wide, meaning sizeof(std::mt19937) jumps from 2504 to 5000 bytes, with exactly 50% of it filled with zeroes. The seed() function also needs a large zero-extending loop. Change-Id: Icaa86fc7b54d4b368c0efffd14efa911e2a40b44 Reviewed-by: Lars Knoll <lars.knoll@qt.io>
| * | Merge 5.10 into 5.10.0Oswald Buddenhagen2017-11-2078-1237/+1883
| |\ \ | | | | | | | | | | | | Change-Id: Ibfbaa8ef89cf45b87a2c65f1da4a708e5464f259
| * | | Adjust screen scale factor for certain screen configurationsVictor-Andrei Variu2017-11-171-2/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pixel density reported by the screen is sometimes not precise enough, so recalculate it: divide px (physical pixels) by dp (device-independent pixels) for both width and height, and then use the average if it is different from the one initially reported by the screen Task-number: QTBUG-62191 Change-Id: Ia2f485c7ce8849db6e7c1d2ac08f5e008aea2ff8 Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
| * | | macOS: Fix OpenGL context sharingMorten Johan Sørvig2017-11-161-1/+24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | See code comments for problem and fix description. Task-number: QTBUG-63180 Change-Id: I6c6381f2c77c246bd975f66f9baa0165e32de777 Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
| * | | QDateTime: Move the deprecation marker to Qt 5.10Thiago Macieira2017-11-141-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit b6a61211280aa6ddd56f107a9795c9824b7702b0 went into dev at around the time of the 5.10 branching. We apparently got the side of it wrong. Change-Id: Ic632b4163d784b83951cfffd14f67bec63fbc795 Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
| * | | Special-case parsing of Zulu time-zone in date-timesEdward Welbourne2017-11-141-1/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When parsing a date-time's zone, a stray Z denotes UTC (a.k.a. Zulu time), despite not being a valid name for the zone. Clients parsing such date strings had to treat the Z as a literal, rather than a zone-ID, but then they got back a LocalTime instead of the UTC the string actually described. So teach QTimeZoneParser to handle this special case and adapt an existing test (that used a time ending in Z, but had to treat it as a local time) to check this works. [ChangeLog][QtCore][QDateTime] When parsing a time-zone, "Z" is now recognized as an alias for UTC. Change-Id: Ib6aa2d8ea2dc6b2da526b39aec74dbc007f90fd8 Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
| * | | Fix BC breakage by adding an overload for the QSqlField constructorAndy Shaw2017-11-132-2/+16
| | | | | | | | | | | | | | | | | | | | Change-Id: I253bb8cd97b982fa5ed5dd546c38deb1f7995986 Reviewed-by: Lars Knoll <lars.knoll@qt.io>
| * | | Fix shader graph metatypes for static buildsLaszlo Agocs2017-11-112-8/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Cannot rely on Q_COREAPP_STARTUP_FUNCTION inside Qt since static builds have to be fully supported. Due to not registering those types, the shader builder silently generated incompilable shader code in static builds. This is critical especially on platforms where static builds are the only choice (INTEGRITY). Task-number: QTBUG-64365 Change-Id: I8820ded239ac160ab00c7fc34918fd3f273f0afb Reviewed-by: Lars Knoll <lars.knoll@qt.io>
| * | | QTemporaryFile: hide the O_TMPFILE feature behind a check for linkat()Thiago Macieira2017-11-114-2/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Some Linux libc (I'm looking at you, Bionic) use the system call but don't expose it to userspace. We could use syscall() to make the system call, but instead I decided to penalize users of those libc by not having the feature. It's probably a good thing, since there were likely to be more problems with Android anyway and I don't have an environment to debug. Task-number: QTBUG-64154 Change-Id: I57a1bd6e0c194530b732fffd14f3007a1062d935 Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@qt.io> Reviewed-by: Lars Knoll <lars.knoll@qt.io>
* | | | xcb: verify if xrandr present before using xcb_randr* APIsGatis Paeglis2017-12-071-12/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Not doing so might break the connection. We have had similar issues before, e.g. QTBUG-45312. Change-Id: I95f15d24773fc92b052578bd72d1ba264d0a5f63 Reviewed-by: Laszlo Agocs <laszlo.agocs@qt.io> Reviewed-by: Uli Schlachter <psychon@znc.in>
* | | | QAbstractSocket: clear error code to match the error descriptionAlex Trotsenko2017-12-062-0/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since 5.10, QIODevice resets the error string on opening. So, we should set the error code to UnknownSocketError accordingly. Change-Id: I0dd314788ffc182d6837f9d06b51e41d6de59d7e Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
* | | | fixup oversight in QStringView::lengthHelperPointerEric Lemanissier2017-12-061-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | amends 3b61cd6ad738b8479bf216dcf736bb935e8812df Change-Id: I3afa008299b8fcccae8943e545b536a68b17bd1a Reviewed-by: Marc Mutz <marc.mutz@kdab.com> Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
* | | | Merge "Merge remote-tracking branch 'origin/5.9' into 5.10" into ↵Liang Qi2017-12-0522-93/+118
|\ \ \ \ | | | | | | | | | | | | | | | refs/staging/5.10
| * \ \ \ Merge remote-tracking branch 'origin/5.9' into 5.10Liang Qi2017-12-0522-93/+118
| |\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Conflicts: src/plugins/platforms/ios/quiview.mm Change-Id: I88384c70047391c75d9ff166c8d9881ff6751dbf
| | * | | | ANGLE: D3D11: Fix shared handle support detection for WARP when MinGW is usedOliver Wolff2017-11-302-10/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The MinGW version we support supports IsWindows8OrGreater so that we can check the windows version properly. As the OpenGL detection falls back to WARP in case of RDP it was possible, that shared handles were wrongly stated as supported, which caused crashes in users' code. Task-number: QTBUG-64657 Change-Id: Iaca2bd169f2764cf6ec68a1d36112a735246b29a Reviewed-by: Andre de la Rocha <andre.rocha@qt.io> Reviewed-by: Andy Shaw <andy.shaw@qt.io>
| | * | | | xcb: Set executed drop action when drop is outside the applicationBłażej Szczygieł2017-11-292-0/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Task-number: QTBUG-62815 Change-Id: I13ee1a3a7e9515d827d29ada38bc0d396f4800d7 Reviewed-by: Shawn Rutledge <shawn.rutledge@qt.io>
| | * | | | iOS: Improve logging during application startupTor Arne Vestbø2017-11-291-5/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Change-Id: I15c1980d7c532c94b34e612bb781c8ed5bf096a0 Reviewed-by: Jake Petroules <jake.petroules@qt.io>
| | * | | | QPSQL: Remove semicolon after Q_DECLARE_SQLDRIVER_PRIVATERobert Szefner2017-11-291-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The semicolon is unnecessary and QtCreator warns of it. Change-Id: I20f803d1ea0136080ff4dc4f7d9863fd8028992e Reviewed-by: Andy Shaw <andy.shaw@qt.io>
| | * | | | Correct the vertex used to calculate v1/v2FracJames McDonnell2017-11-291-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The NoClip version of the drawTriangle code had the vertices swapped. Task-number: QTBUG-50845 Change-Id: I731dafee6cc140ea017b3b7d1051a27ad3081aa7 Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@qt.io>
| | * | | | [doc] Document QString{,Ref}::split() behavior with empty 'sep'Marc Mutz2017-11-292-16/+28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We actually test for this already in tst_QString::split(). Change-Id: I35fe8f90900ea9c8e6251facdb3326b9226348d0 Reviewed-by: Lars Knoll <lars.knoll@qt.io>
| | * | | | iOS: Don't treat AppleTV as an iDevice when resolving physical DPITor Arne Vestbø2017-11-291-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Change-Id: I07ac92a7b2d8c65b7d70a4f2ed5f96f8f4d99ef0 Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
| | * | | | iOS: Update screen metrics (physical DPI) for the latest set of devicesTor Arne Vestbø2017-11-291-13/+19
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We were missing some recent iPads, and the iPhone 8 Plus and X. Change-Id: Ib65644a277a1cbd75ccb360b79b9ac8af935c741 Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
generated by cgit v1.2.3 (git 2.25.1) at 2024-05-04 15:04:47 +0000