| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
| |
Updated version of LGPL and FDL licenseheaders.
Apply release phase licenseheaders for all source files.
Reviewed-by: Trust Me
|
|\
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
master-integration
* 'master' of scm.dev.nokia.troll.no:qt/qt-qml-staging:
QmlViewer: Translate UI
Fix insert and scroll to bottom case
QDeclarativeDebug: Fix test case
QDeclarativeDebug: Handle case where client with ongoing queries is deleted
Remove compile warning
Fix QUnifiedTimer bug
Stop the animation driver when there are no more animations.
Optimization and benchmark for setting object properties from QML.
Add Constant and Final flags to QMetaObjectBuilder
Fixed QML TextEdit docs
Make -no-opengl exist on other platforms
Only ask for name when the user goes on the high score list.
|
| |
| |
| |
| |
| | |
Change-Id: Ib923e5d3946e99001ef682a9dd8ca6d7788818a3
Reviewed-by: Aaron Kennedy
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The code doesn't actually violate aliasing by doing type-punned
dereferencing. The objects are always accessed as the right type. So
disable the warning and pray that GCC doesn't optimise code out of
existence.
Reviewed-by: Trust Me
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This reverts commit 0d3044b547614cbd313d90021606af1f81fb10de.
I'm not sure if this is good for anything. I can't reproduce the
failures that happen on Mac and Windows, so let's try reverting the
only patch that touches QtDeclarative.
If this works, then we'll have found out that the code is broken and
my patch only revealed the errors.
|
|/
|
|
|
|
|
|
|
|
|
|
| |
GCC doesn't like any kind of reinterpret_cast or C-style cast with
pointers. So instead do the work with static_cast<>, which it seems to
like.
Also took the opportunity to change the generic payload type to void*,
so the alignment works as expected. I wonder how we haven't had
serious crashes so far on ARM...
Reviewed-by: Samuel Rødal
|
| |
|
|
|
|
| |
Broken by 43b8305367156c1ceb09eb4a056bdae3f325b5eb.
|
|
|
|
| |
Task-number: QTBUG-13849
|
|
|
|
|
|
|
|
| |
Properties can now be assigned a function that returns the binding
value.
Task-number: QTBUG-14964
Reviewed-by: Aaron Kennedy
|
|
|
|
| |
Task-number: QTBUG-13451
|
|\ |
|
| |
| |
| |
| | |
Reviewed-by: Trust Me
|
|/
|
|
|
|
|
|
| |
Use metaobject revisioning to exclude properties/revisions added in
later versions from interfering with earlier versions.
Task-number: QTBUG-13451
Reviewed-by: Aaron Kennedy
|
|
|
|
| |
Task-number: QTBUG-15697
|
|
|
|
|
|
|
|
|
| |
intptr_t is defined in some C header that we don't include. I don't
know which one it is, but without it, it fails to compile with:
declarative/qml/qdeclarativecontext.cpp:477: error: 'intptr_t' was not declared in this scope
Reviewed-By: Trust Me
|
|
|
|
| |
Task-number: QTBUG-11604
|
|
|
|
| |
Task-number: QTBUG-13999
|
|
|
|
| |
Reviewed-by: Thomas Hartmann <thomas.hartmann@nokia.com>
|
|
|
|
| |
QTBUG-13047
|
|
|
|
| |
QTBUG-12837
|
|
|
|
|
|
|
| |
QML will now invoke methods with enums as arguments, in the same fashion
as QtScript.
QTBUG-11313
|
|
|
|
| |
QTBUG-10816
|
| |
|
|
|
|
|
| |
Task-number: QTBUG-10412
Reviewed-by: Aaron Kennedy
|
| |
|
|
|
|
|
|
|
|
|
| |
This is not supported, and should not silently be converting the function
to a string. See QTBUG-10302 for why we check !isRegExp as well as
isFunction.
Task-number: QTBUG-10237
Reviewed-by: Aaron Kennedy
|
|
|
|
|
| |
Task-number: QTBUG-10291
Reviewed-by: akennedy
|
| |
|
|
|
|
|
|
|
| |
Real solution is probably to fix/rewrite/dispose of QDeclarativeOpenMetaObject.
Task-number: QTBUG-9420
Reviewed-by: Michael Brasser
|
|
|
|
| |
QTBUG-9872
|
| |
|
|
|
|
| |
QTBUG-9704
|
| |
|
|
|
|
|
|
| |
In QML "var"s are not the same as JavaScript vars - they are QVariants
instead. However, as they behave in a similar enough fashion to native
JavaScript it can be confusing to developers when they are called "var".
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
QML used to silently ignore a log of errors - such as a failed assignment
to a QObject property. These errors are now all reported as exceptions
in JavaScript.
Other questionable activities, like assigning a JavaScript array to a
"property var" property which appeared to work, thanks to QtScript's
transparent conversion of arrays to a QVariantList, are now blocked
entirely.
QTBUG-9152 QTBUG-9382 QTBUG-9341 QTBUG-6886
|
| |
|
|
|
|
|
| |
QDeclarativeScriptClass only existed to make compiling against 4.6 and
4.7 easier.
|
|\ |
|
| |
| |
| |
| |
| | |
Always use private/. The WinSCW compiler doesn't search the current
directory, for whatever reason.
|
|/ |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Changing value type bindings in state changes, and implicitly removing
them on property assignment was not reliable. Internally the system
considered a binding on "font" and one on "font.x" as a binding on two
separate properties, even though the "font" binding completely
overrides the "font.x" property.
Following this change a binding to "font.x" creates a proxy binding object
on the "font" property in addition to the "font.x" binding itself. This
allows behavior to be consistent across all operations.
QT-2920
|
| |
|
|
|
|
|
|
|
| |
QML now behaves in a way similar to QtScript when it comes to QObject
ownership.
QT-2803
|
|
|
|
|
|
|
| |
This support was accidentally removed as a consequence of
4a665ff5da05860f5eb46e3982ef3d8163a6cf59.
QTBUG-8001
|
| |
|