summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* Small fix in iteviews for sizeHintForColumn/RowThierry Bastian2012-01-311-3/+3
| | | | | | | We were trying to check the column/row count but always from the root of the model insteal of the root that can be defined in the view. Task-number: QTBUG-5773
* fix coding style for merge request re. utf8 characters in SSL certsPierre Rossi2012-01-311-7/+6
| | | | | | | | fixes minor coding issues for "Use OpenSSL X509_NAME_ENTRY API to parse UTF8 subjectName/issuerName" Task-number: QTBUG-7912 Reviewed-by: Peter Hartmann <peter.hartmann@nokia.com>
* Use OpenSSL X509_NAME_ENTRY API to parse UTF8 subjectName/issuerNameRaul Metsma2012-01-313-36/+29
| | | | | | | | ... to be able to display non-ASCII names from subject and issuerInfo. Task-number: QTBUG-7912 Merge-request: 922 Reviewed-by: Peter Hartmann <peter.hartmann@nokia.com>
* Don't destroy Qt's internal pthread_key_t if it was not initializedBradley T. Hughes2012-01-311-7/+9
| | | | | | | | | | | | | The destructor function destroy_current_thread_data_key() was always called on library unload, even if the pthread_key_t was not initialized. This can cause us to destroy a key that doesn't belong to us, as pointed out in the task below. Fix this by creating a function static instance of a class whose destructor will destroy the key. Task-number: QTBUG-10861 Reviewed-by: Morten Sørvig
* Fixed a problem with toolbars not relayoutingThierry Bastian2012-01-311-4/+4
| | | | | | | | This could happen when dragging an extended toolbar from a mainwindow PAtch was provided on IRC by ravek (#dev) Task-number: QTBUG-10920 Reviewed-by: Trust-Me
* Fixed the itemviews so that the editor is not considered opaque any moreThierry Bastian2012-01-313-26/+3
| | | | | | | The editor was always considered opaque and clipping was set on the painter. This assumption can't be verified. Task-number: QTBUG-11367
* QListView: speedup QListView::setRowHidden()Olivier Goffart2012-01-312-7/+13
| | | | | | | | | Basically the same optimisation that had been made long ago in the tree view: Use the right data structure for QListViewPrivate::hiddenRows (QSet instead of QVector) Task-number: QTBUG-11438 Reviewed-by: Gabriel
* Fixed partial update on QLineEditThierry Bastian2012-01-313-8/+19
| | | | | | | | | QLineControl passes a QRect to tell when a partial update is needed. We used to connect this slot to update. This made the whole lineedit update when the cursor blinked. Task-number: QTBUG-12709 Reviewed-by: gabi
* QGroupBox: Fix the focus rect of the title when using stylesheetMartin Pejcoch2012-01-311-0/+7
| | | | | Task-number: QTBUG-13015 Reviewed-by: Olivier Goffart
* Take Xft.hintstyle by default to match the behavior of GTK+Jiang Jiang2012-01-312-1/+10
| | | | | | | | For Qt apps running in GNOME, we use the GTK+ settings by default (instead of fontconfig settings). Task-number: QTBUG-13800 Reviewed-by: Samuel
* Fix fontconfig pattern mergingJiang Jiang2012-01-311-0/+7
| | | | | | | | | | | | | When merging fontconfig patterns, we need to clean up existing object first, otherwise the added object will be appended instead of replacing the existing object. When multiple font engines are using shaping, if fontconfig patterns are not correctly merged, it will result using the old font. For instance, in "<b>Fake Bold</b> Normal", "Normal" text will continue using the font for "Fake Bold" (that only happen with fallback fonts). Task-number: QTBUG-14455 Reviewed-by: Eskil
* Draw focusrect through style proxyJens Bache-Wiig2012-01-311-1/+1
| | | | | | | | This function call was simply not converted to use the new proxy style, hence preventing people from overriding it properly. Reviewed-by: Andy Shaw Task-number: QTBUG-16127
* Set no brush when the brush is a solid patern transparent color.Pierre Rossi2012-01-311-0/+2
| | | | | | | | Fix a bug in pdf print engine that was noticeable when printing a webpage: the missing image graphic didn't appear in the pdf. Task-number: QTBUG-16435 Reviewed-by: sroedal
* Fixed hslHue / hslHueF returning the wrong value for achromatic QColors.Samuel Rødal2012-01-311-1/+1
| | | | | | | | Made toHsl() work the same way as toHsv(), ensuring the hue gets set to USHRT_MAX and not 0. Task-number: QTBUG-16530 Reviewed-by: Eskil Abrahamsen Blomfeldt
* Fixed QPainterPath::pointAtPercent(t) for one-element paths.Samuel Rødal2012-01-311-1/+4
| | | | | | | Return the MoveTo element instead of QPointF(0, 0). Task-number: QTBUG-16788 Reviewed-by: Eskil Abrahamsen Blomfeldt
* Fixes: missing export of gzungetc() from internal zlibaavit2012-01-311-1/+1
| | | | | Task-number: QTBUG-17278 Reviewed-by: jiang
* Fail in a nicer way when QPixmap is used in a non-GUI application.Samuel Rødal2012-01-311-0/+5
| | | | | | | Show same fatal message as when trying to instantiate a QWidget. Task-number: QTBUG-17873 Reviewed-by: Gunnar Sletta
* fix QLocalSocket::isValid() on WindowsJoerg Bornemann2012-01-311-4/+1
| | | | | | | This function always returned false, even if the socket was connected. Task-number: QTBUG-18204 Reviewed-by: Martin Petersson
* QAccessibleTextEdit: Using x coordinate for calculate character widthJosé Millán Soto2012-01-311-1/+1
| | | | | | Merge-request: 1148 Task-number: QTBUG-18233 Reviewed-by: Frederik Gladhorn <frederik.gladhorn@nokia.com>
* Fix errors when compiling Qt with STLportCédric OCHS2012-01-311-1/+3
| | | | | | Task-number: QTBUG-18374 Reviewed-by: Olivier Goffart Merge-request: 1157
* qRound: do not do operation with double when qreal is floatOlivier Goffart2012-01-311-2/+2
| | | | | | | | 0.5 is double a literal, and they force computation on double while it would be more optimized on qreal if qreal is a float Task-number: QTBUG-16673 Reviewed-by: Thierry
* QSslConfiguration: fix equals operatorPeter Hartmann2012-01-311-1/+1
|
* Fix cursor position of one digit after RTL textJiang Jiang2012-01-311-1/+1
| | | | | | | | | A single digit after RTL text will be considered as RTL because it was treated as Other Neutral (DirON), it can be solved by changing it to DirAN. Task-number: QTBUG-2795 Reviewed-by: Lars Knoll
* Send the hoverLeave not properly sent on the widget inside QGPW.Alexis Menard2012-01-311-2/+2
| | | | | | | | We need to send the hover leave event on the proper receiver in case the widget we embed in a complex hierarchy not the widget itself. Task-number:QTBUG-6986 Reviewed-by:bnilsen
* Support exception handling in QtConcurrent::run()Morten Johan Sørvig2012-01-311-2/+24
| | | | | Use QFutureInterface::reportException() in the same way that we do in QtConcurrent::map().
* avoid infinite recursion in canonicalized() if cwd is a symlinkRitt Konstantin2012-01-311-3/+9
| | | | | | | | | | | | | if current directory is a symlink to another directory on the same drive (the simplest example; say, c:\current\dir is a symlink to c:\target) then the first valid prefix for "c:\target" in canonicalized() is "c:" (NOT "c:\") and would be treated like "c:\current\dir" again...and again... until stack overflow :) Merge-Request: 494 Task-number: QTBUG-7610 Reviewed-by: Zeno Albisser
* Fix memory leak in XSD component of XmlPatternsHonglei Zhang2012-01-3110-25/+25
| | | | | | | | | | | | There are couple of memory leaks in XmlPatterns when parsing XSD files. The reason is that the module over uses the shared data pointer and generates many cyclic reference loop. Some part of the code is refactored. Instead of using shared data pointer, normal C++ pointer is used to break the ownership loop. This is bug fix for QTBUG-8948. Task-number: QTBUG-8948 Reviewed-by: Sami Merila and Laszlo Agocs
* QNAM: Do not cache network connections per username.Robert Hogan2012-01-311-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | If user information is present in the Url, QNetworkAccessHttpBackend will cache the connection with a key in the format of user@hostname. This causes the following tests in QtWebKit to fail: LayoutTests/http/tests/xmlhttprequest/re-login.html LayoutTests/http/tests/xmlhttprequest/re-login-async.html The tests call the same url with user1 credentials, no credentials, user 2 credentials and no credentials. In the second and fourth calls Qt should use the most recently stored credentials. For the fourth call it currently uses user1's credentials instead of user2's. This is because it retrieves the cached connection from the third call and uses the user/pass in that connection's authenticator - which is user1's. In order to retrieve the connection with the most recently used credentials userinfo should be removed when caching the connection in QNetworkAccessCache. This will allow Qt to retrieve the connection used in the third call, in fact the same cached connection for all four calls. Merge-request: 623 Reviewed-by: Markus Goetz <Markus.Goetz@nokia.com>
* Fix painting of overlines in labels.Gunnar Sletta2012-01-311-1/+1
| | | | | | | | | | | If the size of the widget is an odd number, the layout rect will be positioned at a 0.5 offset. The clipping set further below will snap this to 1 and thus clip away the anything drawn on the first row of pixels. Fix this by aligning the rectangle we get by layout rect. Reviewed-by: Eskil Task: http://bugreports.qt.nokia.com/browse/QTBUG-9684
* Avoid repeatedly trying to load unloadable plugins, causing slownessaavit2012-01-311-2/+7
| | | | | | | | | | | | | | In certain uncommon situations, where different Qt versions are used on the same system, the plugin caching optimization may identify a dll as a valid plugin, even though it will later fail to load. This fix will avoid that Qt repeatdly tries to reopen such dlls, something which caused a significant performance hit in these cases. E.g. where Qt would unsuccessfully try to load a number of KDE image format plugins for every image loading operation. Task-number: QTBUG-10066 Reviewed-by: thiago Reviewed-by: janarve
* Fix to allow upper- and lowercase .dll suffix on windowsZeno Albisser2012-01-311-2/+2
| | | | | | | This fix is needed since windows uses case insensitive filesystems. Reviewed-by: Joao Task-number: QTBUG-10305
* Fix conversion between QVariant and QScriptValue.Jedrzej Nowacki2012-01-311-1/+1
| | | | | | | | | | Function converting from QVariant to QScriptValue is smart and tries to unwrap a value. So QVariant(1234) after conversion will be equal to QScriptValue(1234). It is ok, but QVariant(QVariant(...)) should be unpacked only once. Reviewed-by: Kent Hansen Task-number: QTBUG-10315
* QComboBox::modelColumn was not respected when selecting an itemGabriel de Dietrich2012-01-312-4/+11
| | | | | | | Auto-test included Reviewed-by: Olivier Task-number: QTBUG-10491
* Use WS_POPUP window style more often on windows.Prasanth Ullattil2012-01-311-11/+12
| | | | | | | | | | We dont use the WS_POPUP style when one of the following window flags are set (WindowSystemMenuHint, WindowTitleHint, WindowMinMaxButtonsHint WindowCloseButtonHint, WindowContextHelpButtonHint). None of these flags can conflict with this style, hence being removed. Task-number: QTBUG-10589 Reviewed-by: Denis
* Auto expand of QTreeWidget does not work correctly if animatedThierry Bastian2012-01-313-5/+7
| | | | | | | The problem was that the internal state of the tree was not correctly reset. Patch comes from the task. Task-number: QTBUG-10850
* Fix progressbar animation on VistaJens Bache-Wiig2012-01-311-5/+4
| | | | | | | | | | | This fixes two issues. - The indeterminate animation was sometimes incorrectly disabled when value was 0 - The progress animation was incorrectly stopped when progress bars were disabled Task-number: QTBUG-10957 Reviewed-by: richard
* Adjust handling of library .bundle on MacZeno Albisser2012-01-311-4/+8
| | | | | | | | | QLibrary::isLibrary() returns true for files with the sufix .bundle. But we did not use to determine the proper library within the bundle in that case. Reviewed-by: Fabien Freling Task-number: QTBUG-11217
* Fix the blank areas during resize with the raster engine on Mac OS X.Fabien Freling2012-01-313-18/+7
| | | | | | | | This is mainly done by reverting the commit 04e34fe3aecca482abeeabe2e31778e9102eeb08 Task-number: QTBUG-11518 Reviewed-by: Samuel
* Setting a timeSpec on QDateTimeEdit breaks the edit when only time is shownFrederik Gladhorn2012-01-311-0/+15
| | | | | | | | | | | Because of the time zone change we get an "overflow" in the min and max time. For example min 00:00:00 becomes 01:00:00 and max 23:59:59 becomes 00:59:59 which is an invalid time span and the displayed time gets fixed to the newly set min/max value. Autotest included. Task-number: QTBUG-12384 Reviewed-by: Gabriel
* Cocoa: QShortcut(Qt::Key_Enter) does not show in the menu barRichard Moe Gustavsen2012-01-312-71/+104
| | | | | | | | | | | | The reason is that cocoa expects the unicode control character for "enter", rather than the glyph showing an arrow. So rather than using the glyph-table from QKeySequence, we should use a table containing a mapping from Qt::Key to Cocoa control key. Luckily, such a table exists allready in qt_cocoa_helpers, but reversed ( cocoa key to Qt::Key). So this patch creates a function that binds all this together. Reviewed-by: cduclos
* ShortcutOverride has no effect on some shortcuts on Mac OS X (Cocoa).Prasanth Ullattil2012-01-312-8/+36
| | | | | | | | | | | | | | When generating the native key sequences for menu items, Qt prefers the non private unicode characters. But the characters in the NSEvent for keyboard events can contain characters form the unicode range reserved for Apple. For e.g. when user presses the "delete" key, the event contains NSDeleteFunctionKey, where in Qt is expecting NSDeleteCharacter. For now this is the only key identified for translation. If we find similar translations, those can be added to qt_mac_removePrivateUnicode(). Task-number: QTBUG-12495 Reviewed-by: Denis
* Disable selection of Libraries on Windows 7 in getExistingDirectory().Prasanth Ullattil2012-01-311-1/+1
| | | | | | | | | | | | The Libraries on Windows 7 are user-defined collection of folders, which is only a logical representation. There are no file system entries associated with it . This patch makes sure that user cannot choose a library when using the QFileDialog::getExistingDirectory(), in such cases it will show a standard windows warning message. Task-number: QTBUG-12655 Reviewed-by: Thierry Bastian
* QMetaType: add "signed char" as builtin.Olivier Goffart2012-01-312-0/+2
| | | | | | | | | | | | According to the c++ specificaiton, 3.9.1 [basic.fundamentals] Plain char, signed char, and unsigned char are three distinct types In QMetaType, I decide that 'signed char' alias with 'char' This allow qint8 to work nice (and which is already registered as an alias to char (QMetaType::Char) Reviewed-by: Joao Task-number: QTBUG-12920
* Mac Style: Compile FixMartin Petersson2012-01-311-1/+1
| | | | | Task-number: QTBUG-13055 Reviewed-by: Markus Goetz
* Fix corner of scroll area so it is stylable on Mac.Martin Petersson2012-01-311-1/+1
| | | | | Task-number: QTBUG-13055 Reviewed-by: Olivier Goffart
* Keep other text format with QTextOption::SuppressColors tag onJiang Jiang2012-01-311-2/+6
| | | | | | | | | | Setting the QTextOption::SuppressColors tag on a QTextDocument breaks all text format, not just colors. This patch fix this by selectively ignoring color properties in the format and keeping other properties. Task-number: QTBUG-13090 Reviewed-by: Eskil
* let generated flag control SQL generationMark Brand2012-01-315-32/+53
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Previously, if QSqlField's QVariant value had type "invalid", this caused the field to be omitted in data-changing SQL statements generated from QSqlRecord edit buffers. Complaints against this mechanism: - It precludes initializing value type to field type, which would otherwise seem sensible and useful, such as when using model.data() in contexts where the field type is not readily available. - QSqlField::clear() does initialize value type to match field type and so is incompatible with this mechanism. - Problems such as described in QTBUG-13211. - Unwanted distinction between "invalid" and "null" when mapping field values to SQL values. - QSqlField's generated flag already provides a mechanism for controlling SQL generation. These complaints are redressed here by replacing this mechanism with reliance on QSqlField's generated flag. The flag is initialized to false in new edit records and set to true when a value is set. Applications can still manipulate generated flags directly to control SQL generation. Generation of SELECT statements already used the generated flag and is unaffected by this change. Merge-request: 1114 Reviewed-by: Marius Storm-Olsen <marius.storm-olsen@nokia.com> Reviewed-by: Michael Goddard
* QNAM HTTP: Do not set cache headers when already set by userRobert Hogan2012-01-311-4/+6
| | | | | | | | | | | | | | Qt clobbers the cache-control headers set by WebCore when CacheLoadControlAttribute == QNetworkRequest::AlwaysNetwork. This causes at least one QtWebKit layout test to fail: http/tests/misc/refresh-headers.php Qt needs to detect if the client has already made a cache-control decision in the headers before overwriting them. See https://bugs.webkit.org/show_bug.cgi?id=45598 Task-Number: QTBUG-13564 Merge-request: 2472 Reviewed-by: Markus Goetz <Markus.Goetz@nokia.com>
* Mac: Center the Window title on Cocoa.Martin Petersson2012-01-311-0/+2
| | | | | Task-number: QTBUG-13593 Reviewed-by: Richard Moe Gustavsen
* Fix several errors with shaping of Indic scriptspravin.d.s2012-01-311-9/+14
| | | | | | | | | | | | | | | | 1. The opentype init feature was applied to some characters even when the character appeared inside a word. 2. Using the ZWJ to separate Ra and Halant Ya should connect the two latter characters and form a Yaphala 3. Some Oriya chars were not correctly marked as below-base form. Task-number: QTBUG-13620, QTBUG-13616, QTBUG-13542 Merge-request: 2484 Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@nokia.com>
generated by cgit v0.12 at 2024-11-29 10:19:38 (GMT)