8u20: Request for feedback on unit test proposal
daniel.blaukopf at oracle.com
Thu Jan 30 10:16:39 PST 2014
It will run with whatever combination of Quantum, Glass and Prism you throw at it. If you run it with -Dheadless=true, FXUnit will run with the headless configuration of Monocle (not the VNC configuration, but you could just as well use that instead).
On Jan 30, 2014, at 8:09 PM, Tom Eugelink <tbee at tbee.org> wrote:
> Am I to understand that this is using the VNC server? Because I'm not seeing the tell-tale -D parameters in the gradle.build. Or is this actually running on-screen?
> On 2014-1-30 18:40, Daniel Blaukopf wrote:
>> I put up on https://javafx-jira.kenai.com/browse/RT-35396 a draft of what our unit tests might look like if we were to run them on top of the real Quantum Toolkit instead of StubToolkit. The main problem is how to solve the threading: tests that use StubToolkit are all accessing the toolkit and scene graph on the main thread, which you can’t do in Quantum.
>> A webrev is here: http://cr.openjdk.java.net/~dblaukop/webrev-20140123-1559-RT-35330/webrev
>> It’s not complete. I rewrote some tests to correct the threading, but many other problems remain, like that tests call methods on StubToolkit that are not part of the Toolkit interface. For a start I’d like feedback on how we get around the threading problems in JUnit.
More information about the openjfx-dev