| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
| |
This was done by intercepting key events with text in them, and
temporarily submit them as preedit text instead of real input text.
Currently it does not work in WebKit, but that is because WebKit
hides preedit text as well, which is a bug of its own.
RevBy: Simon Hausmann
Autotest: Manual testing went fine
|
|
|
|
| |
Reviewed-by: Trust Me
|
|
|
|
|
|
|
|
| |
Calling SetFocus twice here was obsolete and caused intermittent panics.
Replaced them with proper queueInputCapabilitiesChanged call.
Task-number: QTBUG-6519
Reviewed-by: axis
|
|
|
|
|
|
|
|
|
|
| |
When using T9, FEP is always using black text color. This is due to
that we use default parameters for TFontPresentation. If we change
the text color according to style, before calling
GetFormatOfFepInlineText, it uses color value from style.
Task-number: QTBUG-4072
Reviewed-by: axis
|
|\ |
|
| |
| |
| |
| |
| |
| | |
The word 'module' was missing.
Reviewed-By: TrustMe
|
| |
| |
| |
| |
| | |
Task: QTBUG-5661
RevBy: Janne Koskinen
|
| |
| |
| |
| | |
RevBy: Trust me
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The native FEP handles the 'pencil' key on some platforms by opening an
edit menu on the screen that contains things like 'Insert Symbol' and
'Writing Language' for text input widgets. This was previously not
working in the Qt input method implementation because in order for the
FEP system to open the menu, it must be able to access the menu bar
(CBA). This is done my using the object provider mechanism (MOP) and
an implementation of the MopNext() function was missing. Adding this
and also setting the AppUi as the MOP parent for top level widgets
allowed the FEP framework to find it's way to the menubar and thus
open a menu.
Task-number: QTBUG-5606
Reviewed-by: axis <qt-info@nokia.com>
Reviewed-by: mread <qt-info@nokia.com>
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Qt key event was not handled properly in the case of long key press.
With long key press, QCoeFepInputContext::commitCurrentString gets
called 3 times("q", "", "1"). (Normal key press is causing one call).
This is how aknfep works, so commitCurrentString was modified
to replace first character if long key press event detected.
E.g. "q" is replaced with "1".
qlinecontrol modified to keep cursor position correct.
Signed-off-by: axis <qt-info@nokia.com>
|
|/
|
|
|
| |
Task: QT-1141
RevBy: mread
|
|
|
|
|
| |
Task: QT-2418
RevBy: Shane Kearns
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
QLineEdits with input masks report the cursor position relative to
displayed text via inputMethodQuery(), but the text returned is
the actual text of the control, which can differ from displayed text,
causing mismatch between FEP display and control display.
To properly fix this we would need to know the displayText of
QLineEdits instead of just the text, which on itself should be a trivial
change. The difficulties start when we need to commit the changes back
to the QLineEdit, which would have to be somehow able to handle
displayText, too.
Task made to fix this properly: QTBUG-5050
Task-number: QTBUG-4892
Reviewed-by: axis
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When switching away from the application, the focused widget is set to
null. When switching back, there are callbacks from S60 before the focus
has been restored.
These functions check for null widget, but the output function parameters
are left uninitialised, which causes a crash inside the S60 FEP.
1) GetXYZ functions now initialise the output parameters even when the
focused widget is null.
2) Return no input capability when there is no focused widget, as was
already done during destruction. This stops most of the callbacks
from S60.
Task-number: QTBUG-4618
Reviewed-by: axis
|
|
|
|
|
|
|
|
| |
1) Input methods caused crash due to using CCoeEnv::Fep() without
checking for NULL pointer
2) Autotest itself had Q_ASSERT where it should have used Q_VERIFY
Reviewed-by: axis
|
|\ |
|
| |
| |
| |
| | |
Reviewed-by: Trust Me
|
|/
|
|
|
| |
Task-number: 241223
Reviewed-by: Janne Koskinen
|
|
|
|
| |
Reviewed-by: Trust Me
|
|
|
|
|
|
| |
We export one extra private symbol from QtCore instead, and use that.
RevBy: Miikka Heikkinen
|
|
|
|
| |
RevBy: Trust me
|
|
|
|
| |
forwards the event and doesn't just send a fixed one.
|
|
|
|
| |
RevBy: Trust me
|
|
|
|
| |
RevBy: Trust me
|
|
|
|
|
| |
Contains some smaller fixes and renaming of macros. Looks big,
but isn't scary at all ;)
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Task: 257214
The problem happened if a user called setFocus() on an input capable
widget, and then tried to open the input panel by sending an event.
Since the call to InputCapabilitiesChanged is asynchronous, Symbian
would not yet know about the updated state, and the event would be
lost.
Now we generate our own asynchronous event, and ensure that it is
synchronous in the cases where it's needed.
|
|
|
|
|
|
|
|
| |
Task: 257215
The capabilities would not be updated if the IM hints were the same.
We still try to avoid that when we can, but now we update the
capabilities if we really have to.
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
We use a synchronous call rather than an asynchronous one, so we
don't risk the pointer being dereferenced after destruction is
finished.
|
|
|
|
| |
This reverts commit ba105741867dd96a9d58dcfcb78afef60e611bd6.
|
|
|
|
|
|
| |
We now commit the text if the mouse handler is invoked. This prevents
the case where the VK cannot be brought up because the widget is the
only focusable widget and is stuck in inline edit mode.
|
|\ |
|
| | |
|
| | |
|
| | |
|
|/
|
|
|
|
| |
In case the application is closing, we should not call process events
since event dispatcher is already deleted. Calling processEvents caused
application to oanic with KERN-EXEC 3 i.e. access violation.
|
|
|
|
|
|
|
|
|
|
| |
This was done by making sure that the format code is called from both
start and update, because S60 will often return nothing in start, but
then later on return something in update.
In addition, we fall back on our own format if S60 does not give us
anything. Currently it seems to give us a format only when using
predictive text.
|
|
|
|
|
|
| |
The input context *can* receive events for unfocused widgets, namely
when a CloseSoftwareInputPanel event is sent, or when a mouse event
is sent to a nonfocusable widget.
|
|
|
|
|
| |
Made it more robust, as well as added support for the new
ImhDigitsOnly and ImhFormattedNumbers flags.
|
| |
|
|
|
|
| |
RevBy: Trustme
|
|
|