I just realized that I had messed up one of run configuration parameters.
Which is why I was getting 4.1 instead of AND for Capabilities.version.  I
reverted that and I am getting the correct value of AND.

However, there is still no way to tell the OS version.  I see that
spark.utils.PlatformMobileHelper does actually read an Android system file
to determine the version.  Obviously, this fails during simulation.

I am going to apply your [Mixin] approach to fix this during simulation.

Sorry for the noise.

Thanks,
Om


On Tue, Aug 12, 2014 at 11:42 AM, OmPrakash Muppirala <[email protected]>
wrote:

> On Tue, Aug 12, 2014 at 11:34 AM, Alex Harui <[email protected]> wrote:
>
>>
>>
>> On 8/12/14 11:05 AM, "OmPrakash Muppirala" <[email protected]> wrote:
>>
>> >Nice idea.  So, when we decide to remove this in the future (when Adobe
>> >hopefully fixes this bug) will the users get a compilation error if they
>> >are using this include?
>> If we don't remove the class, then it will still compile and do the
>> override.  Are you trying to find a way to make it smart enough to realize
>> the bug has been fixed?  Not sure that's worth it.  They can simply remove
>> the compile option.
>>
>
> No, that's fine.  I was just trying to figure out the repercussions of
> taking this approach.
>
>
>> It seems like that would be useful for testing different versions of
>> Android in the simulator regardless of this bug.
>>
>
> There is another problem though.  I have a mobile app with two run/debug
> configurations, one for iOS and one for Android.  AIR simulator works fine
> for iOS, i.e. it does correctly report "IOS xxx" for version.
>
> If I add this as a compiler option, it will apply to all run
> configurations, right?  Is there a way to include a compiler option for a
> particular run configuration in FB?
>
> Thanks,
> Om
>
>
>>
>> -Alex
>>
>>
>

Reply via email to