[Development] Another method of registering QML types

[Alan Alpert]

On Thu, Nov 8, 2012 at 2:16 PM, Thiago Macieira
<thiago.macieira at intel.com> wrote:
> On quinta-feira, 8 de novembro de 2012 11.02.23, Alan Alpert wrote:
>> I know commits go to Qt5 first, but with the issue of QtQuick there's
>> different implementations for QtQuick 1 and QtQuick 2. I don't think
>> it's a good idea for QtQuick 1 (Qt 5) to gain new features that are
>> withheld from QtQuick 1 (Qt 4), so if the feature is too big to add to
>> Qt 4.x, then it should just skip QtQuick 1. Which is not too appealing
>> until Qt 5 is out and QtQuick 2 is actually in use.
>
> I don't agree to that. I think the feature can be added to Qt Quick 1 in Qt 5
> and not added to the older version in Qt 4. But, if it's relevant, it should
> be added to Qt Quick 2 too (first?)

Definitely added to QtQuick 2 too (maybe not first, but that would be
purely an implementation detail).

But I'm not sure what the usecase is for extending QtQuick1 in Qt5
only? Isn't the primary point of QtQuick 1 in Qt5 to provide a smooth
migration path? QtQuick 1 in Qt5 is "Done" (unless this has changed),
so if a feature is important enough to ignore that classification in
Qt5, I think it would also be important enough to ignore the done
status of Qt 4. The impression I'm getting is also that it's probably
not important enough to add to QtQuick 1.

--
Alan Alpert