summaryrefslogtreecommitdiffstats
path: root/src/script/api/qscriptengine.cpp
Commit message (Collapse)AuthorAgeFilesLines
...
* adapt to commit 014c4c63066fd3920594e6a58b02f314b5c88cdfKent Hansen2009-08-041-6/+7
|
* Do not crash if using popContext() while the current context have not been ↵Olivier Goffart2009-08-041-0/+5
| | | | | | pushed Reviewed-by: Kent Hansen
* Implement QScriptEngine::abortEvaluation()Tor Arne Vestbø2009-08-031-3/+15
| | | | | | We store the abort-state in the TimeoutChecker, since that's where we'll mostly access it, but the abort result is stored in the QScriptEngine's d-pointer.
* Implement QScriptEngine::setProcessEventsInterval()Tor Arne Vestbø2009-08-031-5/+35
| | | | | We're using a wrapper around TimeoutChecker, where we re-implement didTimeout() to call processEvents().
* Remove QScriptContextPrivate, QScriptContext is now a JSC::CallFrameOlivier Goffart2009-07-311-16/+2
| | | | | | | | | | | | | | There will not be instances of QScriptContext anymore. Pointer to QScriptContext are just pointer to JSC::CallFrame This simplifies the code as we do not need to manage the memory for the QScriptContext anymore. It is compatible because it is not possible to create QScriptContext. They were all created by QScriptEngine. QScriptContext constructor is private. Aknoweldged-by: Kent
* Move the declaration on some function into qscriptengine_p.hOlivier Goffart2009-07-311-1/+1
|
* Small Refactoring of QScriptOlivier Goffart2009-07-311-0/+20
| | | | | | | | | - Create a scope (activation object) for the native constructor in QScriptClass - put the isCalledasConstructor in the activation object (so i can clean up the QScriptContext - Remove the code duplication in all native functions. Aknoweldged-by: Kent
* QScriptValueIterator: fix missing non-enumerable valuesTor Arne Vestbø2009-07-311-6/+6
| | | | | | | | | | Added an extra argument to JSObject::getPropertyNames() that specifies if the non-enumerable properties (those with the DontEnum attribute set) should be included or not. Tried looking at using a unsigned as an attribute-inclusion or exclusion filter, but the semantics of either the calling or the callee code would be very strange so I opted out.
* attempt to return something sensible in uncaughtExceptionBacktrace()Kent Hansen2009-07-301-4/+10
| | | | | | | | JSC doesn't provide a way to get the backtrace after the exception has happened (you can only get it at the point it's thrown, by installing a debugger). As a least effort, we try to use the uncaught exception to provide a 1-line backtrace (the filename and linenumber of the innermost call where the exception happened).
* Implement QScriptEngine::pushContext end popContextOlivier Goffart2009-07-301-3/+25
| | | | Reviewed-by: Kent Hansen
* Fix license headersOlivier Goffart2009-07-291-4/+34
|
* Compile in release mode.Olivier Goffart2009-07-291-0/+1
| | | | | | Some function (such as JSValue::strictEquals) are declared inline in Operations.h and if that header is not included, that will produce link errors
* Enter a scope when enterning a native function.Olivier Goffart2009-07-291-12/+22
| | | | | | | | | | | | | | so native function that would call engine->evaluate("var b = 'foo'); would not change the global object. The change in qscriptengine.cpp makes sure that the correct scope is used for the execution of QScriptEngine::evaluate. The changes in qscriptfunction.cpp push a new scope for native function calls. We might want to move that into QScriptContext later Reviewed-by: Kent Hansen
* Fix tst_QScriptEngine::uncaughtExceptionOlivier Goffart2009-07-281-1/+0
| | | | | | | | | | | | The test used to test that passing 0 as line number works. There is no reason to impose the lineNumber to be >= 1, this is an arbitrary limitation. It even works when passing negative value, but as -1 is a magic number, an error on line '-1' will not have lineNumber Reviewed-by: Kent Hansen
* Fix tst_QScriptEngine::nestedEvaluateOlivier Goffart2009-07-281-1/+16
| | | | | | The 'this' value was not correctly set when evaluating Reviewed-by: Kent Hansen
* Make the 'arguments' object working when calling eval from native functionsOlivier Goffart2009-07-281-2/+7
| | | | | | | The 'arguments' is not handled by JavaScriptCore for native function. We have to do that manually Reviewed-by: Kent Hansen
* Change JavaScriptCore so it throw error when passing invalid regexp flagOlivier Goffart2009-07-281-1/+8
| | | | | | | | | As specified in the specification. The QScriptEngine::newRegExp on the other hand used to work with invalid flags. Reviewed-by: Kent Hansen
* compileKent Hansen2009-07-271-3/+3
|
* make QScriptEngine::setGlobalObject() work to some extentKent Hansen2009-07-271-20/+128
| | | | | | | | | | | | | | | | | | | | | | | | | JSC requires that the global object is actually a JSGlobalObject instance, whereas QScriptEngine::setGlobalObject() allows any object to be set as the global object. The way we solve this is by proxying from an internal global object to the custom (user-set) object. We need to take care that the internal global object is never actually exposed through our API; a brilliantly named helper function, toUsableValue(), makes that happen. Evaluating "var a = 10" with a custom global object doesn't work yet; the variable always ends up in the internal Global Object. For variable assignments, JSC appears to bypass the normal JSObject::put() and instead use JSGlobalObject::copyGlobals{From,To}(), which means I can't intercept and proxy the assignments. This commit enough to get the Context2D example working. There's another bug with iteration of the built-in Global Object's properties (non-enumerable properties are always skipped by the JSC C++ API, whereas with QScriptValueIterator they should not be), but that's a totally separate issue.
* don't rely on custom global object to get GC callbackKent Hansen2009-07-271-33/+28
| | | | | | | | | | | Install custom ClientData on JSGlobalData instance instead. Also some cleanups to avoid globalObject et al being accessed directly. Killed the proxying scheme employed in setGlobalObject() since it didn't work; if you stored the original Global Object and replaced it with another object, then added properties to the new object, they would show up in the old object, too (because the old object would always proxy to whatever the current Global Object was).
* The line number is correctly reported for exceptionBenjamin Poulain2009-07-271-0/+1
| | | | | Enable the tests checking the line number of exception. Add a lower bound on the lineNumber of QScriptEngin::evaluate()
* Remove the uncaughtException, use the JSC exception insteadBenjamin Poulain2009-07-271-9/+6
| | | | | | | | | Use the exception from JSC::exec instead of QScriptEngin::uncaughtException. A few more tests are passing for qscriptvalue and qscriptqobject. Reviewed-by: Kent Hansen
* Fix tst_QScriptValue::callOlivier Goffart2009-07-271-2/+5
| | | | | | | | | | We have to set the CallFrame correctly in QScriptEngine::evaluate() in order to ger the 'arguments' object and all the local stuff working. The code Assert if dynamicGlobalObject is not set, so set it to the global object. Reviewed-by: Kent Hansen
* use custom JSGlobalData clientdata to keep pointer to script engineKent Hansen2009-07-271-1/+11
| | | | Better than relying on a custom Global Object.
* move engine-specific GC marking to QScriptEnginePrivateKent Hansen2009-07-271-40/+44
|
* use scriptEngineFromExec() in a few more placesKent Hansen2009-07-271-3/+3
|
* introduce scriptEngineFromExec() helper functionKent Hansen2009-07-271-0/+27
| | | | | No need to expose the fact that we go via the Global Object to get an engine pointer.
* Implement QScriptString as a wraper around the JSC::IdentifierOlivier Goffart2009-07-231-1/+2
|
* Try best to convert a regexp to a ECMAScript expressionOlivier Goffart2009-07-231-1/+40
|
* set correct property flags in newFunction()Kent Hansen2009-07-171-2/+3
|
* do the correct prototype checkKent Hansen2009-07-161-2/+1
| | | | There was a bug in commit 1d56e14605d74aefb955bdd81ee4083131252c9e
* start implementing constructors for QMetaObject wrappersKent Hansen2009-07-161-7/+12
|
* fix bug in built-in print() functionKent Hansen2009-07-161-3/+2
| | | | D'oh, inverse logic.
* Convert the strings or number value to jscvalue when they are used with an ↵Olivier Goffart2009-07-161-9/+11
| | | | | | | engine Also change the stringValue not to be a pointer. This fixes a memory leak.
* only set default prototype if resulting object has built-in Object prototype ↵Kent Hansen2009-07-161-1/+4
| | | | | | as its prototype port of commit 4182e9c4b2cbfcd0604c57da210e97d68cbbbb52 in qt/master.
* compile when JIT is enabledKent Hansen2009-07-151-0/+1
| | | | NativeFunctionWrapper is typedef for JSFunction
* fix GC-related crashKent Hansen2009-07-151-2/+4
| | | | | | | For some reason our Global Object is not marked by the collector. This fixes the case when you invoke GC explicitly by calling collectGarbage(), but the issue is probably still there when GC is invoked implicitly.
* fix some memory leaksKent Hansen2009-07-151-2/+6
|
* initial attempt at implementing QScriptEngine::setGlobalObject()Kent Hansen2009-07-151-6/+66
| | | | | Doesn't actually replace the global object, but rather has the standard global object act as a proxy to the custom one.
* implement QObject wrapper caching (PreferExistingWrapperObject)Kent Hansen2009-07-151-2/+12
|
* implement QScriptEngine::isEvaluating()Kent Hansen2009-07-151-3/+2
|
* don't assert in QScriptEngine::setGlobalObject()Kent Hansen2009-07-131-2/+2
| | | | allow stuff to run, even though it's likely to crash/not work...
* add GC marking guardsKent Hansen2009-07-131-7/+13
| | | | | Caller is responsible for calling marked() before mark(), otherwise you might get infinite recursion.
* Merge branch 'qtscript-jsc-backend' of ↵Jedrzej Nowacki2009-07-131-183/+39
|\ | | | | | | git@scm.dev.nokia.troll.no:qt/qt-webkit into qtscript-jsc-backend
| * don't crash during GC if no prototype has been set for a metatypeKent Hansen2009-07-101-4/+7
| | | | | | | | | | | | Also, have pushContext() return currentContext() for now, to avoid crashing, and disable processing of __postInit__ property in importExtension() for same reason.
| * implement ability to dynamically change class of script objectsKent Hansen2009-07-101-179/+32
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | With an object created by QScriptEngine::newObject(), it should be possible to call QScriptValue::setClass() to dynamically change the behavior of that object. Similarly, it should be possible to promote plain script objects to QObject (QVariant) wrappers by calling the overload of QScriptEngine::newQObject() (newVariant()) that takes a script object as the first argument. This commit implements this capability. The premise is the (internal) QScriptObject class, which inherits JSC::JSObject. It reimplements all the methods for getting/setting properties etc. Then there's a level of indirection to facilitate dynamic change of the class: Each QScriptObject can have a delegate associated with it that will handle operations on the object. By default there is no delegate, so the object behaves as a normal JS object, as you expect. However, once a delegate is set (e.g., when QScriptValue::setScriptClass() is called), QScriptObject will give the delegate the chance to handle the object operation. In addition to a delegate implementation for QScriptClass-based objects, there are also delegates for QObject and QVariant wrappers. These replace the QObjectWrapperObject and QVariantWrapperObject classes.
* | fix checkSyntax autotest.Jedrzej Nowacki2009-07-131-17/+38
| |
* | fix autotest canEvaluate.Jedrzej Nowacki2009-07-131-4/+80
|/
* register QScriptValue as meta-typeKent Hansen2009-07-091-0/+2
| | | | The old engine did it as well. Makes more tests pass.
* Fix autotest checking QScriptValue::objectId method.Jedrzej Nowacki2009-07-091-4/+13
| | | | (QScriptValue::objectId() and QScriptEnigne::objectById)
generated by cgit v0.12 at 2025-07-27 16:17:19 (GMT)