Seeing test failures in graphics scrum for ListViewKeyInputTest on Mac
jonathan.giles at oracle.com
Thu Sep 26 16:11:27 PDT 2013
On 27/09/2013 11:06 a.m., Richard Bair wrote:
>> I developed the tests but never saw the failures. David saw them and has
>> @ignore'd them until I can work out why they are intermittently failing.
>> I'm guessing it is a keyboard shortcut oversight (by me) between Windows
>> and Mac OS, with the only extra oddity being David saying that the tests
>> are not consistently failing.
> I can get it to consistently fail on my end, so I will see if I can find the cause.
There's a chance there needs to be conditionals in the test case to use
different keyboard commands on Mac OS. I haven't yet investigated, but
this would be my initial suspicion.
>> The other wrinkle is that I believe these tests were only @Ignore'd in
>> the controls repo, as the controls team has not integrated since that time.
> I think we should all start using the same repo after 8 (actually I don't really see why we couldn't just all starting working out of graphics repo now, and save ourselves a bunch of time integrating for the rest of the release). The purpose of team repos is to avoid issues like this, but they add a lot of overhead and actually it would have been easier and faster if we were in the same repo because then graphics wouldn't be broken for a week.
As you know this topic has been broached a number of times internally
(and something I have been actively advocating for some time now), but
my understanding is that it is gaining momentum. Hopefully we'll get
more progress on it in the next weeks / months once everyone recovers
from JavaOne and JavaFX 8.0 stops being so darn time consuming :-)
More information about the openjfx-dev