<Swing Dev> RFR: 8258006: Replaces while cycles with iterator with enhanced for in java.desktop

[Phil Race]

On Tue, 15 Dec 2020 17:14:11 GMT, Andrey Turbanov <github.com+741251+turbanoff at openjdk.org> wrote:

>>> I've added few words in description.
>>> 
>>> About testing: as I can see testing is done via Github Actions. tier1 builds are ok.
>> 
>> OK that's better But about testing. The github actions won't run a single test that touches any code with these changes
>> All the client tests are in tier3-> tier5.
>
> I've run `test-tier3`. 15 tests failed. Looks like they are not related to my changes:
> 
>     JT Harness : Tests that failed
>     Tests are grouped by their final status message.
>     Execution failed: `main' threw exception: TestFailedException: TEST FAILED: ; nested exception is: TestFailedException: TEST FAILED: Rmid process exited with status 1 after 100ms.
> 
>         java/rmi/activation/ActivationSystem/modifyDescriptor/ModifyDescriptor.java: synopsis: need to modify registered ActivationDesc and ActivationGroupDesc 
> 
>     Execution failed: `main' threw exception: TestFailedException: TEST FAILED: ; nested exception is: TestFailedException: TEST FAILED: Test1 failed: a service would not restart
> 
>         java/rmi/activation/Activatable/forceLogSnapshot/ForceLogSnapshot.java: synopsis: Activatable objects cannot be restarted. 
> 
>     Execution failed: `main' threw exception: TestFailedException: TEST FAILED: ; nested exception is: java.rmi.activation.ActivateFailedException: activation failed; nested exception is: java.rmi.activation.ActivationException: timeout creating child process
> 
>         java/rmi/activation/Activatable/checkAnnotations/CheckAnnotations.java: rmid should annotate child process output
>         java/rmi/activation/Activatable/inactiveGroup/InactiveGroup.java: synopsis: rmid should not destroy group when it reports inactiveGroup 
> 
>     Execution failed: `main' threw exception: java.lang.Error: text should be 1.1
> 
>         java/beans/PropertyEditor/TestDoubleClassValue.java: Tests PropertyEditor for value of type Double
>         java/beans/PropertyEditor/TestDoubleTypeValue.java: Tests PropertyEditor for value of type double 
> 
>     Execution failed: `main' threw exception: java.lang.Error: values are not equal
> 
>         java/beans/PropertyEditor/TestDoubleClassJava.java: Tests PropertyEditor for value of type Double
>         java/beans/PropertyEditor/TestDoubleTypeJava.java: Tests PropertyEditor for value of type double 
> 
>     Execution failed: `main' threw exception: java.lang.Exception: failures: 1
> 
>         sanity/client/SwingSet/src/ColorChooserDemoTest.java: Verifies SwingSet3 ColorChooserDemo by performing simple interaction with all the controls that are shown in the ColorChooserDialog. 
> 
>     Execution failed: `main' threw exception: java.lang.StackOverflowError
> 
>         java/beans/XMLEncoder/Test6531597.java: Tests encoding of arrays of primitives
>         java/beans/XMLEncoder/Test6989223.java: Tests Rectangle2D.Double encoding
>         java/beans/XMLEncoder/java_awt_GridBagConstraints.java: Tests GridBagConstraints encoding
>         java/beans/XMLEncoder/java_awt_LinearGradientPaint.java: Tests LinearGradientPaint encoding
>         java/beans/XMLEncoder/java_awt_RadialGradientPaint.java: Tests RadialGradientPaint encoding
>         java/beans/XMLEncoder/java_awt_geom_AffineTransform.java: Tests AffineTransform encoding

These tests more or less *never* fail so how are you sure it is not your changes ?
And 8 beans tests failed and you made 3 changes in beans code so I'd like to see some evidence
to show these failures are unrelated and an explanation as to why these stable tests now fail ?
I don't think they are problem listed.

-------------

PR: https://git.openjdk.java.net/jdk/pull/1574