Another way to implement this is to do it in a separate javascript file.
There you can use all the javascript you need, and not the "almost
javascript" of QML.
I find myself using those whenever I do real programming in javascript.
The inline JS in QML is fine for short things, but for real work, I use
those.
And remember to use ".pragma library" so it's the same scope for all QML
files.
Bo.
Den 27-01-2012 00:45, christopher.ad...@nokia.com skrev:
> Hi Harri,
>
>> Hi,
>>
>> Perhaps like most javascript coders, I'm mainly educated by experimenting
>> with it.
>>
>> I have a pretty strong Python background though, and most of the time I can
>> carry over some usage patterns just guessing how it probably works in js/qml
>> side.
>>
>> Now I was surprised though, and I wonder if the following behaviour is a bug,
>> or what is going on.
>>
>> I have some code, roughly like
>>
>> A.qml:
>> Item {
>> property variant cb // callback
>>
>> signal close()
>>
>> onClosed: cb()
>>
>> function popup(callback) {
>> cb = callback // callback() is called in this assignment!?
>> }
>>
>> ...
>> }
>>
>>
>> then some
>> B.qml:
>>
>> Item {
>> A { id: hello }
>>
>> function continueHere() {
>> print "should continue here"
>> }
>>
>> function setup() {
>> hello.popup(continueHere)
>> }
>> }
>>
>>
>> My problem is that callback is called at assignment time (see comment in
>> code).
>> Is it an artifact of cb being a variant property and I should not be doing
>> this?
>>
>
> There are a couple of reasons why this doesn't work for your code:
> Firstly, the property is a variant property (which internally is a QVariant
> and so cannot store a JavaScript function value).
> Secondly, (and the real reason), we've currently defined the semantic that
> assigning a function to a property causes a binding assignment. A binding
> assignment causes the bound expression to be evaluated, and all of the
> properties involved in the expression to be "captured" (which means we
> automatically re-evaluated the expression if any of those properties change).
>
> In short, the behavior you are seeing is expected.
>
> However, for QtQuick2.0, we are looking at ways to allow a function value to
> be saved to a "var" property (or perhaps to a new property type). We are
> currently considering several options:
>
> 1) adding a new property type: "property function someProp" to which js
> function values can be assigned
> 2) adding a new function to the Qt global object:
> Qt.assignableFunction("expr") -- if the result is assigned to a var property,
> it is NOT considered a binding assignment
> 3) changing the semantic so that a function assignment is NOT considered to
> be a binding assignment, and instead adding a new function to the Qt global
> object: Qt.binding("expr") -- if the result is assigned to any property, it
> is evaluated as a binding assignment.
>
> Note that we do need some way to define binding assignments (eg for initial
> property values in dynamic object instantiation (loader, component), and for
> generating bindings in imperative js code), but as you point out, there are
> some good use-cases for being able to assign a js function value to a var
> property.
>
> The current workaround is to store a js array with a single element (the
> function value) to a var property. But obviously that's not a desirable
> situation.
>
> Cheers,
> Chris.
>
>>
>> Thanks,
>>
>> Harri
>>
>>
>>
>> _______________________________________________
>> Qt-qml mailing list
>> Qt-qml@qt.nokia.com
>> http://lists.qt.nokia.com/mailman/listinfo/qt-qml
> _______________________________________________
> Qt-qml mailing list
> Qt-qml@qt.nokia.com
> http://lists.qt.nokia.com/mailman/listinfo/qt-qml
Bo Thorsen,
Fionia Software.
--
Expert Qt and C++ developer for hire
Contact me if you need expert Qt help
http://www.fioniasoftware.dk
_______________________________________________
Qt-qml mailing list
Qt-qml@qt.nokia.com
http://lists.qt.nokia.com/mailman/listinfo/qt-qml
