summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* Scale logicalDpi when turning on screen scalingPaul Olav Tvete2015-06-184-6/+22
| | | | | | | | | | | | | Assume that the logical DPI for the primary screen is sane, and keep font sizes constant as screens are scaled. The global scale factor will act as a zoom level, and will cause all fonts to be bigger. Note that since we do not change logicalDpi after application startup, the manual test sliders will not match what happens with auto scaling. We may want to fix that... Change-Id: I5a3daa57c2dacf0158836492d31573723e49399a Reviewed-by: Morten Johan Sørvig <morten.sorvig@theqtcompany.com>
* Only turn off font hinting when really scalingPaul Olav Tvete2015-06-184-19/+44
| | | | | | | | | | | | | | | | | | | We don't want to turn off font hinting based on whether auto-scaling is on, otherwise Qt Creator will look bad on a low-DPI screen. Instead, we turn it off only if we have at least one screen that is scaled. QHighDpiScaling::isActive() now means that there is actual scaling going on. There is currently no need for the old meaning, so this change does not include QHighDpiScaling::isEnabled(). Note that nothing can save us from the case where there is one high-DPI and one low-DPI screen. In that case we choose looking bad on the low-DPI screen instead of looking like crap on the high-DPI screen. Also note that our font system we doesn't allow us to change our minds on hinting later when screens are plugged in or removed. Change-Id: I14a4ec7a49f4ba74a4c74684c7b951d0a71b951d Reviewed-by: Morten Johan Sørvig <morten.sorvig@theqtcompany.com>
* Make drag and drop work again after the mergePaul Olav Tvete2015-06-173-3/+13
| | | | | | | | QBasicDrag does not have enough information to map the coordinates, so we move that responsibility to the plugin. Change-Id: I6eb87014b3259844bb3373ecf961d92debdb2b84 Reviewed-by: Morten Johan Sørvig <morten.sorvig@theqtcompany.com>
* dirty hack to make setScreenFactor workPaul Olav Tvete2015-06-171-0/+4
| | | | | | | | | This is a seriously dirty hack, but without this, screen geometries will be wrong when scale factors are changed after application startup. Change-Id: Ia39164303ef648ec3752d7f4b9db06477a2d3fb4 Reviewed-by: Morten Johan Sørvig <morten.sorvig@theqtcompany.com>
* Update doc.Morten Johan Sørvig2015-06-162-25/+36
| | | | | | | | | | Move to the class documentation in the cpp file, focus on documenting the class and not necessarily high-dpi scaling in general. Change-Id: I4f04398904325b2f9061db8d3a82cd362276e977 Reviewed-by: Friedemann Kleint <Friedemann.Kleint@theqtcompany.com> Reviewed-by: Paul Olav Tvete <paul.tvete@theqtcompany.com>
* Windows: Fix compilation after f5bdd92e5d0217bf3eec13395e4baa95bd8fda37.Friedemann Kleint2015-06-121-2/+16
| | | | | | | | | Bring back static versions of the constrained scaling functions. To be squashed with dbe2dd295cce19680f8ba7a7cca65150701051e2 . Change-Id: Id05b8085ab56dd777ce1ab3b7da3ce47c01dcb00 Reviewed-by: Morten Johan Sørvig <morten.sorvig@theqtcompany.com>
* Add QPlatformWindow::windowClosestAcceptableGeometry().Friedemann Kleint2015-06-103-2/+17
| | | | | | | | | | Add a function to determine the closest acceptable geometry (height-for-width) to QPlatformWindow similar to the existing QPlatformWindow::windowMinimumSize(), windowMaximumSize()... Clean up the hack in the Windows platform plugin accordingly. Change-Id: I96e943cfdbafdd0ff444e0d53ee0bdf8603d33df Reviewed-by: Paul Olav Tvete <paul.tvete@theqtcompany.com>
* Merge remote-tracking branch 'qt/dev' into dev-highdpiPaul Olav Tvete2015-06-09567-3487/+7363
|\ | | | | | | | | | | | | | | | | | | Conflicts: src/gui/kernel/qsimpledrag.cpp src/gui/kernel/qwindowsysteminterface.cpp src/gui/kernel/qwindowsysteminterface_p.h src/plugins/platforms/xcb/qxcbwindow.cpp Change-Id: Icd887552ade61d6a2b2527383970f7145aa00faf
| * Testlib: Export blacklist functionsCaroline Chao2015-06-081-3/+4
| | | | | | | | | | | | | | So they can be used in QtQuickTest. Change-Id: I64925feb243fb27234951276d81119bc1f6d5a56 Reviewed-by: Simon Hausmann <simon.hausmann@theqtcompany.com>
| * itemsviews: Add 15 QList::reserve() callsSérgio Martins2015-06-085-11/+37
| | | | | | | | | | | | Change-Id: I4ed8872ecd560d51c00cb3d2b93e082c5abdd8dc Reviewed-by: Olivier Goffart (Woboq GmbH) <ogoffart@woboq.com> Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
| * make the enums of QProcess known to the meta object systemJoerg Bornemann2015-06-081-0/+11
| | | | | | | | | | | | | | | | | | | | This enables users to construct enum values from strings and provides useful qDebug operators. Task-number: QTBUG-10777 Change-Id: Ia3592f00d559b980b4875ee8fb5e8ffb4328d464 Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com> Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
| * Export class QAccessible::ActivationObserver.Friedemann Kleint2015-06-081-1/+1
| | | | | | | | | | | | | | | | Change a21dfab54ed518266c42d3e87e02bd991de8486e de-inlined its destructor, so, the class needs to be exported. Change-Id: I32d21622a944a93718f549060e5e5f7e71fd2646 Reviewed-by: Simon Hausmann <simon.hausmann@theqtcompany.com>
| * Fix incorrect expression in directfb keyboard codeHolger Hans Peter Freyther2015-06-081-1/+1
| | | | | | | | | | | | | | | | | | | | Add missing equal sign to have the result of the OR stored inside the modifiers variable. Fixes Coverity CID#89071. Change-Id: I0ee1a080f5f6757ae7646733aa3bb5cf8c6cdeed Reviewed-by: Simon Hausmann <simon.hausmann@theqtcompany.com>
| * Remove exec bits from files that should not be executableDmitry Shachnev2015-06-0745-0/+0
| | | | | | | | | | | | | | Change-Id: I66f49c6db82eadc3b11cc9b1cf01375e9596a8e6 Reviewed-by: Lisandro Damián Nicanor Pérez Meyer <perezmeyer@gmail.com> Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com> Reviewed-by: Mark Brand <mabrand@mabrand.nl>
| * Fix WEC2013 build of sqlite3.Bjoern Breitmeyer2015-06-071-0/+2
| | | | | | | | | | | | Change-Id: Iaa953df941d888b8ae1a590dd728a1d819d7e604 Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com> Reviewed-by: Mark Brand <mabrand@mabrand.nl>
| * Keep the mouse timestamp for tests closer to the actual timeUlf Hermann2015-06-051-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Pass on the requested delay into the mouse events. We don't care about single msecs as the surrounding code also takes time that we don't consider and whenever we generate a release we add 500ms of time skew anyway. The initial press of a double click has to increment the timestamp, just like all other mouse events do. Change-Id: Ifb94830f9f878ce87d92bfd86c72a16b55052f3b Reviewed-by: Simon Hausmann <simon.hausmann@theqtcompany.com> Reviewed-by: Lars Knoll <lars.knoll@digia.com>
| * qdoc: fix invalid memory access in HtmlGeneratorMartin Smith2015-06-051-1/+2
| | | | | | | | | | | | | | | | | | | | HtmlGenerator::highlightedCode() contained a potential out-of-range- memory reference, which is prevented by this fix. Change-Id: I3bc87a8287e1d51c2786f5ec42384dbac04c9636 Task-number: QTBUG-45643 Reviewed-by: Martin Smith <martin.smith@digia.com>
| * qdoc: Just reformatting and removing dead codeMartin Smith2015-06-052-50/+6
| | | | | | | | | | | | | | Whitespace only. Change-Id: I6e4b55ad55e22da4150530457202f144fcc7f416 Reviewed-by: Martin Smith <martin.smith@digia.com>
| * doc: Remove uses of \returnsMartin Smith2015-06-051-2/+2
| | | | | | | | | | | | | | | | | | It is a doxygen command that is not recognized by qdoc. Change-Id: I26b0c54a1aa715fc9d8b70e411502f46fc6bc0f1 Task-number: QTBUG-46495 Reviewed-by: Martin Smith <martin.smith@digia.com>
| * doc: Remove uses of \mainclassMartin Smith2015-06-054-4/+0
| | | | | | | | | | | | | | | | This qdoc command is deprecated. Task-number: QTBUG-46476 Change-Id: Iee36f91bdcf500c2e88022a5f8a9c3accc6048c2 Reviewed-by: Martin Smith <martin.smith@digia.com>
| * Move the function helper to its own fileJorgen Lind2015-06-055-13/+19
| | | | | | | | | | | | | | | | Also make it a class so a forward header is generated by syncqt Change-Id: Ibab6b925dc6e9dab1b617b9b4027a4144e4a3773 Reviewed-by: Timur Pocheptsov <Timur.Pocheptsov@digia.com> Reviewed-by: Laszlo Agocs <laszlo.agocs@theqtcompany.com>
| * Fix crash in tst_qfiledialog on OS XSimon Hausmann2015-06-042-4/+14
| | | | | | | | | | | | | | | | | | | | The test aggressively shows and hides dialogs and popups, and we would end up installing this global event monitor frequently. However we never cleaned up properly, for example if the window didn't get hidden properly or if the monitor was already installed for some reason. Change-Id: I6fa28eaeb03e089ced735912dbe29b0b8ad75d58 Reviewed-by: Andy Shaw <andy.shaw@theqtcompany.com>
| * Merge "Merge remote-tracking branch 'origin/5.5' into dev" into refs/staging/devSimon Hausmann2015-06-04361-2448/+5051
| |\
| | * Merge remote-tracking branch 'origin/5.5' into devSimon Hausmann2015-06-03361-2448/+5051
| | |\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Conflicts: src/corelib/global/qnamespace.qdoc src/corelib/io/qwindowspipereader.cpp src/corelib/io/qwindowspipereader_p.h src/corelib/statemachine/qstatemachine.cpp src/corelib/statemachine/qstatemachine_p.h src/plugins/platforms/xcb/qxcbconnection.h tests/auto/network/access/qnetworkreply/tst_qnetworkreply.cpp tests/auto/tools/qmake/tst_qmake.cpp tests/manual/touch/main.cpp Change-Id: I917d694890e79ee3da7d65134b5b085e23e0dd62
| | | * Merge "Merge remote-tracking branch 'origin/5.5.0' into 5.5" into ↵Liang Qi2015-06-0222-241/+383
| | | |\ | | | | | | | | | | | | | | | refs/staging/5.5
| | | | * Merge remote-tracking branch 'origin/5.5.0' into 5.5Liang Qi2015-06-0222-241/+383
| | | | |\ | | | | | | | | | | | | | | | | | | Change-Id: I060b0e10e55487a4a4a101f77ad7c8dc8cf0f553
| | | | | * Merge "Merge remote-tracking branch 'origin/5.4' into 5.5" into ↵Jani Heikkinen2015-06-0221-238/+377
| | | | | |\ | | | | | | | | | | | | | | | | | | | | | refs/staging/5.5.0
| | | | | | * Merge remote-tracking branch 'origin/5.4' into 5.5Liang Qi2015-06-0121-238/+377
| | | | | | |\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Conflicts: qmake/generators/mac/pbuilder_pbx.cpp src/corelib/global/qglobal.h Change-Id: I2c0f7544bf194f2d0f59218fd583c822901487b0
| | | | | | | * Merge "Merge remote-tracking branch 'origin/5.4.2' into 5.4" into ↵Liang Qi2015-05-296-13/+19
| | | | | | | |\ | | | | | | | | | | | | | | | | | | | | | | | | | | | refs/staging/5.4
| | | | | | | | * Merge remote-tracking branch 'origin/5.4.2' into 5.4Liang Qi2015-05-296-13/+19
| | | | | | | | |\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Change-Id: Ied2a227a25859163a924c7b5717492a1f974c5ca
| | | | | | | | | * Revert some changes in light of GCC 4 -fPIE reversalv5.4.2Stephen Kelly2015-05-292-1/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The -fPIE option is now accepted when using GCC 4, which means it is available for backward compatibility for clients using CMake 2.8.11 or older which makes use of the POSITION_INDEPENDENT_CODE feature. Conditionally use that feature for old versions of cmake with GCC 4. Restore the tests for those versions, and clarify the situation in the ChangeLog. Change-Id: I5a06b155dda7db559d86841a2b34fd8ed95acbd0 Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
| | | | | | | | | * Make qglobal.h only complain for GCC >= 5 about -fPIEThiago Macieira2015-05-291-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit 3eca75de67b3fd2c890715b30c7899cebc096fe9 introduced the #error nagging about use of -fPIE, but it makes the transition quite difficult for people using other buildsystems. So let's give people a grace period and enforce only for GCC >= 5. Clang is affected, but differently. The problem only happens with -flto -- that is, it happens when the linker detects that it's creating a final executable. Maybe -Wl,-pie would fix it. Change-Id: If4d5ac8db0ed4a84a3eaffff13e275edc29a72b7 Reviewed-by: Simon Hausmann <simon.hausmann@theqtcompany.com> Reviewed-by: Dmitry Shachnev <mitya57@gmail.com>
| | | | | | | | | * Only add -fPIC flags for compilers known to require it.Stephen Kelly2015-05-272-6/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit 083c9269 (Try to ensure that -fPIC is used in CMake builds, 2015-05-11) added a raw -fPIC to the INTERFACE_COMPILE_OPTIONS of Qt5::Core, which affects all consuming compilers. Use the qmake variable $$QMAKE_CXXFLAGS_APP instead, which at least currently contains only the -fPIC variable or harmlessly expands to nothing. If the content of that qmake variable changes in the future, a $$QMAKE_CXXFLAGS_APP_PIC variable should be extracted in qmake and used here. Don't use the POSITION_INDEPENDENT_CODE feature of CMake. That adds the -fPIE flag for executables, which is explicitly what qglobal.h forbids since commit 3eca75de (Make qglobal.h complain if you use -fPIE, 2015-05-11). The current behavior of that CMake feature is tracked here: http://public.kitware.com/Bug/view.php?id=15570 Change-Id: I5c5bcc40fe4b310b55a681a3505f45c50adfa054 Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com> Reviewed-by: Simon Hausmann <simon.hausmann@theqtcompany.com>
| | | | | | | | | * Avoid false positives from google by storing OpenSSL version as Unicode.Richard J. Moore2015-05-251-1/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Google scan play store apps for the openssl version string which leads to false positives since we record the version we were compiled against even though we don't link it directly. Task-number: QTBUG-46265 Change-Id: Iefd0e0954149c17350d49f57f9f374938124d7b8 Reviewed-by: Olivier Goffart (Woboq GmbH) <ogoffart@woboq.com>
| | | | | | | | | * Do not modify decoder when determining image-formatAllan Sandfeld Jensen2015-05-211-7/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We should not configure the decoder when just determining the image- format. Doing so can cause all versions of libpng to print a warning, and some versions to fail to decode. The code appears to be a leftover from when the image-format logic was copied out of the introduction of the decoding method, where the proper settings are still applied. Task-number: QTBUG-46233 Change-Id: I6619728804f040ae6c9d637c7298a8586e22499e Reviewed-by: Andy Shaw <andy.shaw@theqtcompany.com> Reviewed-by: aavit <eirik.aavitsland@theqtcompany.com>
| | | | | | | | | * Workaround for Samsung keyboard bugPaul Olav Tvete2015-05-201-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Return null string instead of empty string when the selection is empty. It looks like Samsung just tests for selection == null when doing backspace. Task-number: QTBUG-45785 Change-Id: Iaa006a8ffe52b2704c7348646dde9ca4e1f78c5c Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@theqtcompany.com>
| | | | | | | * | | Fix no bearermanagement buildLorn Potter2015-05-292-20/+24
| | | | | | | |/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | As pointed out in the bug, it also fixes API use when configured with no bearermanagement. Task-number: QTBUG-46239 Change-Id: Ief8df85ad6acf61e8d5bb3eed54e7d6ecb84c1a0 Reviewed-by: Alex Blasche <alexander.blasche@theqtcompany.com>
| | | | | | | * | Doc: Added the missing \brief and \image for example docsVenugopal Shivashankar2015-05-272-0/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The Qt TestLib examples are just tutorials so updated the qdocconf to use the default thumbnail in the Qt Creator welcome screen. Task-number: QTBUG-41996 Change-Id: Ia04a42a92e414c97a426b6095a62621a348e7de0 Reviewed-by: Topi Reiniö <topi.reinio@digia.com> Reviewed-by: Martin Smith <martin.smith@digia.com>
| | | | | | | * | Fix documentation of obsoleted functions to show the right replacementAndy Shaw2015-05-261-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Change-Id: Ib008a5544d68d93e1f96ff6b7504e9a7ea4bb192 Reviewed-by: Friedemann Kleint <Friedemann.Kleint@theqtcompany.com>
| | | | | | | * | Use the new non-obsoleted functions for getting the paper detailsAndy Shaw2015-05-261-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This fixes a problem with the margins not being correctly respected as the functions introduced previously would set a different property to what was being queried in this case. Change-Id: I3458c8e46239276a296d17aa80da7330c85fcf0a Reviewed-by: Friedemann Kleint <Friedemann.Kleint@theqtcompany.com>
| | | | | | | * | Ensure the same behavior on Win as Unix re the host in isApparentlyStaleAndy Shaw2015-05-221-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When the hostname is empty then it is assumed that the lock file is from the same host as the one running the application. Change-Id: Iba8aefc171a209294371dc2022d93ede3035b242 Reviewed-by: Will Wagner <willw@carallon.com> Reviewed-by: David Faure <david.faure@kdab.com>
| | | | | | | * | Doc: Excluded qdoc files that caused unnecessary Qt Creator warnings.Venugopal Shivashankar2015-05-201-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The customcompleter and textcodes are widget examples, but they end up in the Qt Core exampledirs boundary because of a reference to the plugandpaint example in the docs. This resulted in a couple of wrong entries being written into the examples-manifest.xml, which is used by Qt Creator. This change explicitly exludes the qdoc pages for the two examples so that qdoc doesn't add the corresponding entries into examples-manifest.xml. Task-number: QTBUG-41996 Change-Id: I0e95b6d4d93e0ce18f5b34e5034b279598b4924f Reviewed-by: Topi Reiniö <topi.reinio@digia.com> Reviewed-by: Martin Smith <martin.smith@digia.com>
| | | | | | | * | Prevent bad-ptrs deref in QNetworkConfigurationManagerPrivateMikhail Lappo2015-05-181-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Prevent application to crash with segfault in Qt bearer thread. Corrected hardly reproduceable bug, when QNetworkConfigurationManagerPrivate in pollEngines slot dereferenced null and bad pointers and caused crash Task-number: QTBUG-44407 Change-Id: I2f0b11b2d10125a21a62588d76ad824f375e4a1d Reviewed-by: Richard J. Moore <rich@kde.org>
| | | | | | | * | Merge remote-tracking branch 'origin/5.4.2' into 5.4Liang Qi2015-05-185-184/+298
| | | | | | | |\| | | | | | | | | | | | | | | | | | | | | | | | | | | | Change-Id: Ic1e8b154b43c8d8e3f682c96b074b2b77dbfe2ac
| | | | | | | | * Try to ensure that -fPIC is used in CMake buildsEvangelos Foutras2015-05-141-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In commit 36d6eb721e7d5997ade75e289d4088dc48678d0d the -fPIE switch was replaced with -fPIC in an effort to avoid generating copy relocations which are incompatible with Qt5 when built with -reduce-relocations. Task-number: QTBUG-45755 Change-Id: I59a55ea15052f498104848c5fd867e563ddc2290 Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
| | | | | | | | * Windows: Fix exit crash of GUI applications when deleting argv[].Friedemann Kleint2015-05-131-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When passing Qt arguments followed by normal arguments, a double deletion may occur due to Qt shifting argv. For example: argv[] = app -qwindowgeometry +50+50 some_arg <null> becomes: argv[] = app some_arg <null> some_arg <null> Terminate deletion when encountering the null pointer. Change-Id: I5279955b6bd463f5858d6e5e8e16a1f5d0945652 Reviewed-by: Joerg Bornemann <joerg.bornemann@theqtcompany.com>
| | | | | | | | * Make qglobal.h complain if you use -fPIEThiago Macieira2015-05-131-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Prior to Qt 5.4.2 (commit 36d6eb721e7d5997ade75e289d4088dc48678d0d), we allowed it, but now we need to enforce that it is not used. Note that -fPIE does define __PIC__, so we need this to catch the use of -fPIE. [ChangeLog][Important Behavior Changes] On x86 and x86-64 systems with ELF binaries (especially Linux), due to a new optimization in GCC 5.x in combination with a recent version of GNU binutils, compiling Qt applications with -fPIE is no longer enough. Applications now need to be compiled with the -fPIC option if Qt's option "reduce relocations" is active. Note that Clang is known to generate incompatible code even with -fPIC if the -flto option is active. Task-number: QTBUG-45755 Change-Id: I66a35ce5f88941f29aa6ffff13dd210e0aa2728f Reviewed-by: Dmitry Shachnev <mitya57@gmail.com> Reviewed-by: Simon Hausmann <simon.hausmann@theqtcompany.com>
| | | | | | | | * Android: Store and use the class names as key when caching.Christian Strømme2015-05-082-180/+293
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Previously the jclass handle was part of the key used for caching the class' methods and fields. Using the jclass handle is not ideal, but it meant that we could easily create a key when the only identifier we had was the jobject or jclass handle. However, in Android 5.1, the re-use of handles seems to be more aggressive and therefore increasing the chance of a collision in the cache look-up. This change removes caching for all calls where we don't know the class name, as that is the only thing that guarantees that we create unique keys for each class. The consequence of this is that only calls that provide a class name will benefit from the internal caching. Task-number: QTBUG-45748 Change-Id: I0039d04e7c068debc9e3b3983632c45dc8e52309 Reviewed-by: Frank Meerkoetter <frank.meerkoetter@basyskom.com> Reviewed-by: Alex Blasche <alexander.blasche@theqtcompany.com>
| | | | | | | * | Doc: corrected autolink issues statemachine.qdocNico Vertriest2015-05-131-16/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Task-number: QTBUG-40362 Change-Id: Ia686ebdfd722f448aa30fb1f1f266b6148df4026 Reviewed-by: Topi Reiniö <topi.reinio@digia.com>
| | | | | | | * | Doc: clarify ownership after QStackedWidget::removeWidgetJoerg Bornemann2015-05-131-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Change-Id: Ia14b72cdac3205a3896c47ecc81b31adb508181b Task-number: QTBUG-44891 Reviewed-by: Leena Miettinen <riitta-leena.miettinen@theqtcompany.com>
generated by cgit v1.2.3 (git 2.25.1) at 2024-04-27 13:47:20 +0000