On 10.04.15 11:46, Alexander Nyßen wrote:
> Hi Tom,
>
>> Am 10.04.2015 um 11:00 schrieb Tom Schindl
>> <
tom.schindl@xxxxxxxxxxxxxxx>> <
mailto:tom.schindl@xxxxxxxxxxxxxxx>>:
>>
>> Hi,
>>
>> It looks like there's a misunterstanding.
>>
>> You can NOT use -Dorg.osgi.framework.bundle.parent=ext but still
>> need to use the adapter hook because you need swtfx - so you
>> still require min.runtime.feature.
>
> I see, then this was indeed misunderstood.
>
>>
>> I think I made the changes in a way that you can as a first
>> experiment leave your code unchanged but only switch to version
>> 2.0 of the feature.
>
> I can confirm that switching to version 2.0 of the
> min.runtime.feature does not have a bad impact on GEF, i.e. with
> the package imports in place things still work as with 1.2.
>
>>
>> Once we are sure this works you can start removeing ALL javax
>> imports from your MANIFEST.MF and verify if things are still OK.
>
> Would you recommend to remove the imports and increase our version
I would do so because this remove the confusion that:
* you import version 2.2 although you are on Java8
* you don't import a package but things still compile which might be
confusing to people.
It also has the advantage that if GEF4 runs in a pure OSGi+JavaFX env
that people don't require adapter hooks if they are not using the
SWT-integration.
> constraint for the min.runtime.feature to 2.0.0 at all? Does it
> have an impact on our transitive dependencies (up to now we still
> work with Java7 and JavaFX 2.2).
no - we still support 2.2 no impact for you!
I'd like to contribute our 2.0 version in the next few days to sim-rel
repo to fix copyright some problem reports from the sim-rel tools but
I guess this is ok for you now that we made sure things still operate
as before.
Tom
--
Thomas Schindl, CTO
BestSolution.at EDV Systemhaus GmbH
Eduard-Bodem-Gasse 5-7, A-6020 Innsbruck
http://www.bestsolution.at/Reg. Nr. FN 222302s am Firmenbuchgericht Innsbruck
