summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* add recursion guard for GC markingKent Hansen2009-08-203-4/+10
| | | | | | | To achieve behavior of the old back-end. There, the recursion guard was automatic because a mark flag was set on the object as soon as marking begun, but in JSC it appears to only be set _after_ the marking is completed.
* provide line number information for innermost call frameKent Hansen2009-08-204-0/+11
| | | | | For the innermost frame, we don't have a returnPC, so use the line number that was last passed to the engine agent.
* provide filename in stack view if we have itKent Hansen2009-08-201-4/+7
| | | | | | In the JSC back-end it's currently possible that we know the file name, but not the line number (we can't just assume that lineNumber == -1 implies native function).
* set the current frame when getting callbacks in JSC DebuggerKent Hansen2009-08-203-30/+42
| | | | | If the debugger stops execution, we want the frame pointer to be in sync (so we get the full backtrace).
* disable executable line number info since we don't have it yetKent Hansen2009-08-202-4/+15
| | | | | The old back-end relied on a script-to-XML converter to get information about the script, but it's not implemented yet in the new back-end.
* make QScriptEngine::objectById() workKent Hansen2009-08-202-8/+5
| | | | | It needs to work even when there is no public QScriptValue that holds a reference to the object.
* Fix "fixme" in QtFunction::mark()Jedrzej Nowacki2009-08-201-3/+1
| | | | | | QtFunction::mark() does not cause infinite recursion any more. Reviewed-by: Kent Hansen
* add missing includeKent Hansen2009-08-201-0/+1
|
* Fix JavaScriptCore build on systems without anonymous mmappingNorbert Leser2009-08-202-3/+3
| | | | | | | | | | | | | | | | Use fastMalloc when neither MMAP nor VIRTUALALLOC are enabled RegisterFile constructor currently throws #error when both MMAP and VIRTUALALLOC conditions fail. On any platform that does not provide these features (for instance, Symbian), the fallback should be regular malloc (or fastMalloc). It is functionally equivalent in this case, even though it may have certain drawbacks such as lack of dynamic pre-allocation. Taken upstream from https://bugs.webkit.org/show_bug.cgi?id=27051 Reviewed-by: Simon Hausmann
* only enable QtScript on platforms where WebKit can be builtKent Hansen2009-08-201-1/+1
| | | | | QtScript needs to build JavaScriptCore (part of WebKit). If WebKit can't be built, we assume that JavaScriptCore can't be built.
* add platforms AIX and HPUXKent Hansen2009-08-201-1/+35
| | | | Also add some compilers, and define PLATFORM(UNIX) on Solaris.
* use PLATFORM(SOLARIS) instead of Qt-specific defineKent Hansen2009-08-202-2/+2
|
* Merge branch 'master' of git@scm.dev.nokia.troll.no:qt/qt into ↵Kent Hansen2009-08-2086-992/+2749
|\ | | | | | | qtscript-jsc-backend
| * XdndLeave events have incorrect source idBradley T. Hughes2009-08-201-2/+4
| | | | | | | | | | | | | | | | When sending XdndLeave events, we should use the effectiveWinId() of the drag source (just like we do when sending XdndEnter and XdndPosition events). Task-number: 255559
| * Instantiate the CP949 codec when building with -no-iconvBradley T. Hughes2009-08-201-0/+1
| | | | | | | | Otherwise this codec would not be available.
| * Add some more tests for QQuaternion multiplicationRhys Weatherley2009-08-201-0/+9
| | | | | | | | Reviewed-by: trustme
| * Improve memory usage of the connectionlists inside QObjectOlivier Goffart2009-08-1911-260/+353
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ... by not allocating space for slots in the vector. Before, the vector uses the signal index as index. The problem is that the slots and signal are mixed in the same index space. We solve the problem by having a different index space for the signal in the connectionlists vector. All we need to do is to add the information about the number of signals in the moc. Also, we are not connecting to cloned signal but only to the orginial ones. For example, destroyed(QObject * = 0) would generate two signal, we now only connect to the first one. This also improve a little bit the performence while activating signals since it removed one call to indexOfMethod. Reviewed-by: Brad
| * Doc - mentioning that the begin...() functions emit a signal that mustKavindra Devi Palaraja2009-08-191-0/+15
| | | | | | | | | | | | | | | | | | be handled by connected views/proxies. Otherwise, the views/proxies may end up in an invalid state. Task: 227718 Reviewed-By: Olivier Goffart
| * Document more of the behavior of QTouchEventBradley T. Hughes2009-08-191-43/+115
| | | | | | | | | | | | | | | | This includes docs on the default QWidget::event() behavior, how to use touch with QAbstractScrollArea subclasses, how the propagation and grouping works, as well as some caveats. Reviewed-by: David Boddie
| * Changed the streambookmarks example to use aggregationThorbjørn Lindeijer2009-08-197-71/+89
| | | | | | | | | | | | | | | | QXmlStreamReader and QXmlStreamWriter can be used conveniently without subclassing, which the example now demonstrates. Reviewed-by: mae Reviewed-by: David Boddie
| * Doc - Fixed whitespace issuesKavindra Devi Palaraja2009-08-191-15/+15
| | | | | | | | Reviewed-By: TrustMe
| * Doc - Cleanups on QAbstractTableModel, QAbstractItemModel, etc.Kavindra Devi Palaraja2009-08-191-388/+433
| | | | | | | | Reviewed-By: TrustMe
| * Carbon and Cocoa: Adding support for standard gestures.Richard Moe Gustavsen2009-08-192-4/+16
| |
| * Improved the documentation of saving and restoring window geometry.Denis Dzyubenko2009-08-193-23/+50
| | | | | | | | | | | | | | Mentioned in the doc that the preferred way to save/restore a geometry of a QMainWindow is to use both saveGeometry() and saveState(). Reviewed-by: Kavindra Devi Palaraja
| * Don't use pointers to temporary vars that go out of scope.Denis Dzyubenko2009-08-191-1/+1
| | | | | | | | | | | | | | | | Moved the dummy variable to the same scope as the 'motion' variable. This fixes Coverity defect CID 1528. Reviewed-by: Olivier Goffart Reviewed-by: Gabriel de Dietrich
| * Make the license test pass.Frans Englich2009-08-1955-91/+1479
| | | | | | | | | | | | | | This is partly done to address a review comment for S60. Reviewed-by: Marius SO Reviewed-by: Paul
| * Generate proper license header.Frans Englich2009-08-192-104/+189
| |
| * Revert re-adding of readUnknownElementThorbjørn Lindeijer2009-08-192-10/+0
| | | | | | | | | | Now that it is really no longer used. QXmlStreamReader::skipCurrentElement replaces it.
| * Fixed compile errorThorbjørn Lindeijer2009-08-191-1/+1
| | | | | | | | Apparently I forgot one occurrence.
* | Do not pass JSValue per const referenceOlivier Goffart2009-08-192-4/+4
| | | | | | | | It is a POD with the size of a pointer
* | CleanupsOlivier Goffart2009-08-193-24/+5
| | | | | | | | | | Remove useless variable. Use QBoolBlocker instread of custom QScript::InEval.
* | invalidate wrapped JSC::JSValue when script engine is deletedKent Hansen2009-08-192-5/+6
| | | | | | | | Otherwise we will crash if someone tries to do something with the value.
* | Fix QScriptValue::objectId().Jedrzej Nowacki2009-08-195-9/+12
| | | | | | | | | | | | | | QScriptValue id were made persistent. It depands on JSC:JSValue JSCell pointer not on QScriptValuePrivate attr. Reviewed-by: Kent Hansen
* | add more detailed tests for QRegExp <--> JS RegExp conversionKent Hansen2009-08-191-1/+19
| |
* | remove expected failureKent Hansen2009-08-191-1/+0
| | | | | | | | Works as of commit 5bca43cca3ac90429e3f9263d0d7ea8c9eb164d4.
* | add test to make sure a JavaScript object's ID persistsKent Hansen2009-08-191-0/+10
| | | | | | | | | | | | Even if the QScriptValue is destroyed, the underlying ID should not change; e.g. if a new QScriptValue is created and wraps the same object, the ID should be the same as before.
* | update expected output to match that of JSC-based qtscript back-endKent Hansen2009-08-191-2/+2
| | | | | | | | Output of backtrace has changed slightly (e.g. "at" instead of "@").
* | make QScriptEngine::isEvaluating() work for top-level evaluationKent Hansen2009-08-193-2/+45
| | | | | | | | | | | | Since QScriptEngine::evaluate() doesn't create a new stack frame anymore, we need to use a dedicated variable to keep track of whether the engine is currently evaluating or not.
* | move the builtinFunctionNames test to qscriptengine autotestsKent Hansen2009-08-192-198/+175
| | | | | | | | | | | | | | With the JSC-based back-end, stack frames aren't created when calling any of the built-in ECMA functions, so we can't base the test on that. Instead, just look up the "name" property of each function and check that it has the expected value.
* | make Qt Designer compile when Qt is configured with -no-scriptKent Hansen2009-08-198-4/+24
| |
* | only add qtscript autotests if qt is configured appropriatelyKent Hansen2009-08-191-13/+15
| |
* | make moc autotest compile without qtscriptKent Hansen2009-08-192-1/+4
| |
* | build on SolarisKent Hansen2009-08-192-2/+12
| | | | | | | | | | Work-around for "Error: A union member cannot have a user-defined assignment operator."
* | Add support for aligned buffers without alignment macrosThiago Macieira2009-08-191-1/+13
| | | | | | | | | | If we don't have alignment macros, we do what we can: overcommit 64 bytes and find the proper 64-byte-aligned position in the buffer.
* | Q_DISABLE_COPY were added to QScriptValuePrivate.Jedrzej Nowacki2009-08-191-0/+1
| | | | | | | | Reviewed-by: Kent Hansen
* | Clean upJedrzej Nowacki2009-08-191-1/+0
| | | | | | | | | | | | There is no point in setting class member in destructor Reviewed-by: Kent Hansen
* | Internal API clean up.Jedrzej Nowacki2009-08-193-16/+0
| | | | | | | | | | | | Static method QScriptValue::initFromJSCValue was removed. Reviewed-by: Kent Hansen
* | Clean up.Jedrzej Nowacki2009-08-193-49/+57
| | | | | | | | | | | | | | | | | | Internal API of QScriptValue were cleaned. toPublic() was removed and replaced by standard static Private::get(). All initFromXXX() methods were replaced by overload call of initFrom(). Reviewed-by: Kent Hansen
* | Fix potential building problem.Jedrzej Nowacki2009-08-191-7/+2
| | | | | | | | | | | | | | | | Comparison between pointer and int where removed. According to doc operator-> where created (class QScriptValueAutoRegister suppose to behave as pointer). Reviewed-by: Kent Hansen
* | remove QtScript from includepathKent Hansen2009-08-194-4/+1
| | | | | | | | qmake no longer uses QtScript.
generated by cgit v0.12 at 2024-11-15 00:13:01 (GMT)